技术领域technical field
本发明涉及计算机领域，尤其涉及一种数据传输方法和装置。The invention relates to the field of computers, in particular to a data transmission method and device.
背景技术Background technique
传输控制协议(TCP，Transmission Control Protocol)是一种面向连接的、可靠的、 基于字节流的传输层通信协议，能够实现互联网可靠的端到端通讯，应用非常广泛。Transmission Control Protocol (TCP, Transmission Control Protocol) is a connection-oriented, reliable, byte stream-based transport layer communication protocol, which can realize reliable end-to-end communication on the Internet and is widely used.
互联网中交换机使用的通信信道，以及互联网中的缓冲存储空间一般是互联网上所有主 机端的共享资源。随着通信技术的不断丰富和提高，发送端的业务量不断增多，使得不同的 发送端之间会发生资源竞争，从而导致网络拥塞。为了解决网络拥塞问题，发送端会对丢包 事件进行监控。当发现丢包事件时，发送端即认为出现了网络拥塞，从而大幅度缩小拥塞窗 口。但是，目前，由于无线通信网络的普及，比如，发送端可通过无线4G、无线3G、无线保 真即Wi-Fi、公网或虚拟专用网络(VPN，Virtual Private Network)等接入方式接入公有 云网络或数据中心网络，在发送端通过无线通信网络与公有云网络或数据中心网络进行数据 交互的过程中会发生随机丢包，而并非为网络拥塞引起的丢包，但在此情况下发送端也会大 幅度减小拥塞窗口，从而大幅度降低了网络带宽的利用率。The communication channels used by switches in the Internet and the buffer storage space in the Internet are generally shared resources of all hosts on the Internet. With the continuous enrichment and improvement of communication technology, the business volume of the sending end is increasing, which causes resource competition between different sending ends, resulting in network congestion. In order to solve the problem of network congestion, the sender will monitor packet loss events. When a packet loss event is found, the sender thinks that there is network congestion, thereby greatly reducing the congestion window. However, at present, due to the popularization of wireless communication networks, for example, the sending end can access through wireless 4G, wireless 3G, wireless fidelity (Wi-Fi), public network or virtual private network (VPN, Virtual Private Network) and other access methods. In the public cloud network or data center network, random packet loss may occur during data interaction between the sending end and the public cloud network or data center network through the wireless communication network, rather than packet loss caused by network congestion, but in this case The sending end will also greatly reduce the congestion window, thereby greatly reducing the utilization of network bandwidth.
发明内容Contents of the invention
本申请提供了一种数据传输方法、装置、发送端与存储介质，能够在缓解随机丢包所带 来的影响的基础上，最大限度的保证了网络带宽的利用率。This application provides a data transmission method, device, sending end and storage medium, which can maximize the utilization of network bandwidth on the basis of alleviating the impact of random packet loss.
第一方面，本申请提供了一种数据传输方法，应用于发送端，方法包括：获取发送端所 发送的传输报文的往返时延的抖动，判断传输报文的往返时延的抖动是否大于拥塞判断阈值， 往返时延的抖动表示传输报文的往返时延偏离基准往返时延的幅度；若往返时延的抖动大于 拥塞判断阈值，则获取当前带宽，判断当前带宽的下降幅度是否超出下降阈值；若当前带宽 未下降或下降幅度未超出下降阈值，则按照第一预设微步长缩小拥塞窗口。In the first aspect, the application provides a data transmission method, which is applied to the sending end. The method includes: obtaining the jitter of the round-trip delay of the transmission message sent by the sending end, and judging whether the jitter of the round-trip delay of the transmission message is greater than Congestion judgment threshold, the jitter of the round-trip delay indicates the extent to which the round-trip delay of the transmitted message deviates from the reference round-trip delay; if the jitter of the round-trip delay is greater than the congestion judgment threshold, the current bandwidth is obtained, and it is judged whether the decrease in the current bandwidth exceeds the drop Threshold; if the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold, then reduce the congestion window according to the first preset micro-step.
在本发明实施例中，在判断往返时延的抖动大于拥塞判断阈值，但当前带宽未下降或下 降幅度未超出下降阈值时，判断数据传输出现的是随机丢包，并未发生网络拥塞，对拥塞窗 口进行微调。并未对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓 解随机丢包所带来的影响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络 带宽整体的利用率。In the embodiment of the present invention, when it is judged that the jitter of the round-trip delay is greater than the congestion judgment threshold, but the current bandwidth does not drop or the drop does not exceed the drop threshold, it is judged that random packet loss occurs in data transmission and no network congestion occurs. The congestion window is fine-tuned. The congestion window has not been greatly adjusted, which avoids misjudgment of network congestion, so that on the basis of mitigating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network performance. Overall bandwidth utilization.
根据第一方面，在第一方面的第一种可能中，上述数据传输方法还包括：若当前带宽的 下降幅度超出下降阈值，则获取网络丢包参数，判断网络丢包参数是否满足网络拥塞判断条 件，网络丢包参数记录了传输报文在传输过程中数据包的丢包信息，网络拥塞条件包括至少 一项条件；若网络丢包参数满足网络拥塞判断条件中任意一项，则按照第一预设粗步长缩小 拥塞窗口，第一预设粗步长大于第一预设微步长。According to the first aspect, in the first possibility of the first aspect, the above-mentioned data transmission method further includes: if the drop rate of the current bandwidth exceeds the drop threshold, then obtaining the network packet loss parameter, and judging whether the network packet loss parameter meets the network congestion judgment condition, the network packet loss parameter records the packet loss information of the data packet in the transmission process of the transmission message, and the network congestion condition includes at least one condition; if the network packet loss parameter meets any one of the network congestion judgment conditions, then according to the first The preset coarse step size reduces the congestion window, and the first preset coarse step size is larger than the first preset micro step size.
在经历了上述实施例中的往返时延的抖动和带宽的判定后，引入网络丢包参数，进一步 提高判定网络拥塞引起的丢包的准确性。当传输报文的往返时延的抖动大于拥塞判断阈值， 且当前带宽的下降幅度超出下降阈值，且网络丢包参数满足网络拥塞判断条件中任意一项时， 确定发生网络拥塞丢包，较大幅度缩小拥塞窗口，以缓解网络拥塞情况。After experiencing the jitter of the round-trip delay and the determination of the bandwidth in the above embodiments, the network packet loss parameter is introduced to further improve the accuracy of determining the packet loss caused by network congestion. When the jitter of the round-trip delay of the transmitted message is greater than the congestion judgment threshold, and the current bandwidth drop exceeds the drop threshold, and the network packet loss parameter meets any of the network congestion judgment conditions, it is determined that the network congestion packet loss occurs, and the larger Reduce the congestion window by a large amount to alleviate network congestion.
示例性的，网络丢包参数包括连续丢包数目，网络拥塞判断条件包括连续丢包数目大于 等于两个；和/或，网络丢包参数包括丢包率，网络拥塞判断条件包括丢包率大于预设的拥塞 丢包率阈值；和/或，网络丢包参数包括丢包重传报文数目，网络拥塞判断条件包括丢包重传 报文数目大于预设的拥塞丢包重传报文阈值。Exemplarily, the network packet loss parameter includes the number of consecutive packet loss, and the network congestion judgment condition includes that the number of consecutive packet loss is greater than or equal to two; and/or, the network packet loss parameter includes a packet loss rate, and the network congestion judgment condition includes a packet loss rate greater than or equal to Preset congestion packet loss rate threshold; and/or, network packet loss parameters include the number of packet loss and retransmission messages, and network congestion judgment conditions include the number of packet loss and retransmission messages being greater than the preset congestion packet loss and retransmission message threshold .
根据第一方面的第一种可能，在第一方面的第二种可能中，上述数据传输方法还包括： 若往返时延的抖动小于等于拥塞判断阈值，则判断往返时延的抖动是否处于非拥塞抖动范围 内，非拥塞抖动范围为传输报文在传输过程中未发生网络拥塞时往返时延的抖动的变化范围， 且拥塞判断阈值大于非拥塞抖动范围的上限值；若往返时延的抖动处于非拥塞抖动范围内， 保持拥塞窗口不变。According to the first possibility of the first aspect, in the second possibility of the first aspect, the above-mentioned data transmission method further includes: if the jitter of the round-trip delay is less than or equal to the congestion judgment threshold, judging whether the jitter of the round-trip delay is in the Within the congested jitter range, the non-congested jitter range is the change range of the round-trip delay jitter when no network congestion occurs during the transmission of the message, and the congestion judgment threshold is greater than the upper limit of the non-congested jitter range; if the round-trip delay The jitter is within the non-congested jitter range, keeping the congestion window unchanged.
确定未发生网络拥塞，且数据传输通畅，保持拥塞窗口不变，避免不必要的占用资源的 拥塞窗口调节。Make sure that there is no network congestion and data transmission is smooth, keep the congestion window unchanged, and avoid unnecessary resource-occupied congestion window adjustments.
根据第一方面的第二种可能，在第一方面的第三种可能中，上述数据传输方法还包括： 若往返时延的抖动小于非拥塞抖动范围的下限值，则按照第二预设微步长扩大拥塞窗口。According to the second possibility of the first aspect, in the third possibility of the first aspect, the above data transmission method further includes: if the jitter of the round-trip delay is less than the lower limit of the non-congested jitter range, according to the second preset Microstepping enlarges the congestion window.
确定数据传输过程中出现的是随机丢包，并未发生网络拥塞，对拥塞窗口进行微调。避 免对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓解随机丢包所带 来的影响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络带宽整体的利用 率。It is determined that random packet loss occurs during data transmission, and there is no network congestion, and fine-tune the congestion window. Avoid large adjustments to the congestion window and avoid misjudgment of network congestion. On the basis of alleviating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network bandwidth. overall utilization.
根据第一方面的第二种可能，在第一方面的第四种可能中，上述数据传输方法还包括： 若往返时延的抖动大于非拥塞抖动范围的上限值，则按照第三预设微步长缩小拥塞窗口。According to the second possibility of the first aspect, in the fourth possibility of the first aspect, the above data transmission method further includes: if the jitter of the round-trip delay is greater than the upper limit value of the non-congested jitter range, according to the third preset Microstepping shrinks the congestion window.
确定数据传输出现的是随机丢包，并未发生网络拥塞，对拥塞窗口进行微调。避免对拥 塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓解随机丢包所带来的影 响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络带宽整体的利用率。It is determined that random packet loss occurs during data transmission, and no network congestion occurs, and fine-tune the congestion window. Avoid large adjustments to the congestion window and avoid misjudgment of network congestion. On the basis of alleviating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network bandwidth. overall utilization.
根据第一方面的第一种可能，在第一方面的第五种可能中，上述数据传输方法还包括： 若网络丢包参数均未满足网络拥塞判断条件中任意一项，则按照第四预设微步长缩小拥塞窗 口。According to the first possibility of the first aspect, in the fifth possibility of the first aspect, the above-mentioned data transmission method further includes: If none of the network packet loss parameters meets any of the network congestion judgment conditions, then according to the fourth predetermined Set the micro-step size to reduce the congestion window.
确定数据传输出现的是随机丢包，并未发生网络拥塞，对拥塞窗口进行微调。避免对拥 塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓解随机丢包所带来的影 响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络带宽整体的利用率。It is determined that random packet loss occurs during data transmission, and no network congestion occurs, and fine-tune the congestion window. Avoid large adjustments to the congestion window and avoid misjudgment of network congestion. On the basis of alleviating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network bandwidth. overall utilization.
示例性的，上述数据传输方法还包括：若当前带宽未下降或下降幅度未超出下降阈值， 按照第一预定补偿步长，增加非拥塞抖动范围的上限值。Exemplarily, the above data transmission method further includes: if the current bandwidth does not drop or the drop range does not exceed the drop threshold, increasing the upper limit of the non-congested jitter range according to the first predetermined compensation step.
当传输报文的往返时延的抖动大于拥塞判断阈值，但当前带宽未下降或下降幅度未超出 下降阈值，表明未发生网络拥塞。为了后续的网络拥塞判断更加精准且效率更高，可适当增 加非拥塞抖动范围的上限值，拓宽非拥塞抖动范围，从而实现对非拥塞抖动范围的设定的优 化。When the jitter of the round-trip delay of transmitted packets is greater than the congestion judgment threshold, but the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold, it indicates that no network congestion occurs. In order to make the subsequent judgment of network congestion more accurate and efficient, the upper limit of the non-congested jitter range can be appropriately increased to widen the non-congested jitter range, thereby realizing the optimization of the setting of the non-congested jitter range.
示例性的，上述数据传输方法还包括：若网络丢包参数均未满足网络拥塞判断条件中任 意一项，按照第二预定补偿步长，增加非拥塞抖动范围的上限值。Exemplarily, the above data transmission method further includes: if none of the network packet loss parameters meets any of the network congestion judgment conditions, increasing the upper limit of the non-congested jitter range according to the second predetermined compensation step.
当传输报文的往返时延的抖动大于拥塞判断阈值，且当前带宽下降幅度超出下降阈值， 但网络丢包参数均未满足网络拥塞判断条件中任意一项，表明未发生网络拥塞。为了后续的 网络拥塞判断更加精准且效率更高，可适当增加非拥塞抖动范围的上限值，拓宽非拥塞抖动 范围，实现对非拥塞抖动范围的设定的优化。When the jitter of the round-trip delay of transmitted packets is greater than the congestion judgment threshold, and the current bandwidth drop exceeds the drop threshold, but none of the network packet loss parameters meets any of the network congestion judgment conditions, it indicates that no network congestion occurs. In order to make the subsequent judgment of network congestion more accurate and efficient, the upper limit of the non-congested jitter range can be appropriately increased to widen the non-congested jitter range and optimize the setting of the non-congested jitter range.
第二方面，本申请提供了一种数据传输方法，包括：获取发送的传输报文的往返时延的 抖动，判断传输报文的往返时延的抖动是否大于拥塞判断阈值，往返时延的抖动表征传输报 文的往返时延偏离基准往返时延的幅度；若往返时延的抖动大于拥塞判断阈值，则获取网络 丢包参数，判断网络丢包参数是否满足网络拥塞判断条件，网络丢包参数记录了传输报文在 传输过程中数据包的丢包信息；若网络丢包参数均未满足网络拥塞判断条件中任意一项，则 按照第一预设微步长缩小拥塞窗口，网络拥塞条件包括至少一项条件。In the second aspect, the present application provides a data transmission method, including: obtaining the jitter of the round-trip delay of the transmitted transmission message, judging whether the jitter of the round-trip delay of the transmission message is greater than the congestion judgment threshold, and the jitter of the round-trip delay It represents the magnitude of the round-trip delay of the transmitted message from the reference round-trip delay; if the jitter of the round-trip delay is greater than the congestion judgment threshold, the network packet loss parameter is obtained to determine whether the network packet loss parameter meets the network congestion judgment condition, and the network packet loss parameter Records the packet loss information of the data packet during the transmission of the transmission message; if the network packet loss parameters do not meet any of the network congestion judgment conditions, the congestion window will be reduced according to the first preset micro-step size. The network congestion conditions include at least one condition.
利用传输报文的往返时延的抖动和网络丢包参数，共同进行网络是否发生拥塞的判断。 在判断往返时延抖动大于拥塞判断阈值，但网络丢包参数均未满足网络拥塞判断条件中任意 一项时，判断数据传输出现的是随机丢包，并未发生网络拥塞，对拥塞窗口进行微调。并未 对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓解随机丢包所带来 的影响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络带宽整体的利用率。The jitter of the round-trip delay of the transmitted message and the network packet loss parameters are used to jointly judge whether the network is congested. When it is judged that the round-trip delay jitter is greater than the congestion judgment threshold, but the network packet loss parameters do not meet any of the network congestion judgment conditions, it is judged that random packet loss occurs in data transmission and no network congestion occurs, and the congestion window is fine-tuned . The congestion window has not been greatly adjusted, which avoids misjudgment of network congestion, so that on the basis of mitigating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network performance. Overall bandwidth utilization.
示例性的，网络丢包参数包括连续丢包数目，网络拥塞判断条件包括连续丢包数目大于 等于两个；和/或，网络丢包参数包括丢包率，网络拥塞判断条件包括丢包率大于预设的拥塞 丢包率阈值；和/或，网络丢包参数包括丢包重传报文数目，网络拥塞判断条件包括丢包重传 报文数目大于预设的拥塞丢包重传报文阈值。Exemplarily, the network packet loss parameter includes the number of consecutive packet loss, and the network congestion judgment condition includes that the number of consecutive packet loss is greater than or equal to two; and/or, the network packet loss parameter includes a packet loss rate, and the network congestion judgment condition includes a packet loss rate greater than or equal to Preset congestion packet loss rate threshold; and/or, network packet loss parameters include the number of packet loss and retransmission messages, and network congestion judgment conditions include the number of packet loss and retransmission messages being greater than the preset congestion packet loss and retransmission message threshold .
根据第二方面，在第二方面的第一种可能中，上述数据传输方法还包括：若网络丢包参 数满足网络拥塞判断条件中任意一项，则获取当前带宽，判断当前带宽的下降幅度是否超出 下降阈值；若当前带宽的下降幅度超出下降阈值，则按照第一预设粗步长缩小拥塞窗口，第 一预设粗步长大于第一预设微步长。According to the second aspect, in the first possibility of the second aspect, the above-mentioned data transmission method further includes: if the network packet loss parameter meets any one of the network congestion judgment conditions, then obtaining the current bandwidth, and judging whether the decrease in the current bandwidth is Exceeding the drop threshold; if the decrease of the current bandwidth exceeds the drop threshold, the congestion window is reduced according to the first preset coarse step size, and the first preset coarse step size is larger than the first preset micro step size.
结合往返时延的抖动、网络丢包参数和带宽三个方面的参数，判断是否发生网络拥塞丢 包，进一步提高了判断网络拥塞丢包的准确性。当传输报文的往返时延的抖动大于拥塞判断 阈值，且网络丢包参数满足网络拥塞判断条件中任意一项，且当前带宽的下降幅度超出下降 阈值时，确定发生网络拥塞丢包，较大幅度缩小拥塞窗口，以缓解网络拥塞情况。Combining the round-trip delay jitter, network packet loss parameters, and bandwidth parameters, it can judge whether network congestion packet loss occurs, which further improves the accuracy of judging network congestion packet loss. When the jitter of the round-trip delay of the transmitted message is greater than the congestion judgment threshold, and the network packet loss parameter meets any one of the network congestion judgment conditions, and the decrease of the current bandwidth exceeds the decrease threshold, it is determined that the packet loss due to network congestion occurs. Reduce the congestion window by a large amount to alleviate network congestion.
根据第二方面，在第二方面的第二种可能中，上述数据传输方法还包括：若往返时延的 抖动小于等于拥塞判断阈值，则判断往返时延的抖动是否处于非拥塞抖动范围内，非拥塞抖 动范围为传输报文在传输过程中未发生网络拥塞时往返时延的抖动的变化范围，且拥塞判断 阈值大于非拥塞抖动范围的上限值；若往返时延抖动处于非拥塞抖动范围内，保持拥塞窗口 不变。According to the second aspect, in the second possibility of the second aspect, the above data transmission method further includes: if the jitter of the round-trip delay is less than or equal to the congestion judgment threshold, then judging whether the jitter of the round-trip delay is within the non-congested jitter range, The non-congested jitter range is the variation range of the round-trip delay jitter when there is no network congestion during the transmission of the message, and the congestion judgment threshold is greater than the upper limit of the non-congested jitter range; if the round-trip jitter is in the non-congested jitter range Within, keep the congestion window unchanged.
确定未发生网络拥塞，且数据传输通畅，保持拥塞窗口不变，避免不必要的占用资源的 拥塞窗口调节。Make sure that there is no network congestion and data transmission is smooth, keep the congestion window unchanged, and avoid unnecessary resource-occupied congestion window adjustments.
根据第二方面的第二种可能，在第二方面的第三种可能中，上述数据传输方法还包括： 若往返时延的抖动小于非拥塞抖动范围的下限，则按照第二预设微步长扩大拥塞窗口。According to the second possibility of the second aspect, in the third possibility of the second aspect, the above data transmission method further includes: if the jitter of the round-trip delay is less than the lower limit of the non-congested jitter range, then follow the second preset microstep Extend the congestion window long.
往返时延变化较小，未发生网络拥塞，数据传输通畅，可平缓地适当扩大拥塞窗口,对拥 塞窗口进行微调，并未对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而 在缓解随机丢包所带来的影响的基础上，最大限度的保证了网络带宽的利用率，从而提高了 网络带宽整体的利用率。The round-trip delay changes little, there is no network congestion, and the data transmission is smooth. The congestion window can be expanded appropriately, and the congestion window can be fine-tuned. The congestion window has not been greatly adjusted, which avoids the misjudgment of network congestion. Therefore, on the basis of alleviating the impact caused by random packet loss, the utilization rate of the network bandwidth is guaranteed to the greatest extent, thereby improving the overall utilization rate of the network bandwidth.
根据第二方面的第二种可能，在第二方面的第四种可能中，上述数据传输方法还包括： 若往返时延的抖动大于非拥塞抖动范围的上限，则按照第三预设微步长缩小拥塞窗口。According to the second possibility of the second aspect, in the fourth possibility of the second aspect, the above data transmission method further includes: If the jitter of the round-trip delay is greater than the upper limit of the non-congested jitter range, then according to the third preset Long shrink the congestion window.
确定未发生网络拥塞，对拥塞窗口进行微调，并未对拥塞窗口进行大幅度调整，避免了 出现网络拥塞误判的情况，从而在缓解随机丢包所带来的影响的基础上，最大限度的保证了 网络带宽的利用率，从而提高了网络带宽整体的利用率。Make sure that no network congestion occurs, fine-tune the congestion window, and do not make a large adjustment to the congestion window, avoiding the misjudgment of network congestion, so as to alleviate the impact of random packet loss, maximize the The utilization rate of the network bandwidth is guaranteed, thereby improving the overall utilization rate of the network bandwidth.
根据第二方面的第一种可能，在第二方面的第五种可能中，上述数据传输方法还包括： 若当前带宽未下降或下降幅度未超出下降阈值，则按照第四预设微步长缩小拥塞窗口值。According to the first possibility of the second aspect, in the fifth possibility of the second aspect, the above-mentioned data transmission method further includes: if the current bandwidth does not drop or the drop range does not exceed the drop threshold, then according to the fourth preset micro-step Reduce the congestion window value.
确定未发生网络拥塞，对拥塞窗口进行微调，并未对拥塞窗口进行大幅度调整，避免了 出现网络拥塞误判的情况，从而在缓解随机丢包所带来的影响的基础上，最大限度的保证了 网络带宽的利用率，从而提高了网络带宽整体的利用率。Make sure that no network congestion occurs, fine-tune the congestion window, and do not make a large adjustment to the congestion window, avoiding the misjudgment of network congestion, so as to alleviate the impact of random packet loss, maximize the The utilization rate of the network bandwidth is guaranteed, thereby improving the overall utilization rate of the network bandwidth.
示例性的，上述数据传输方法还包括：若网络丢包参数均未满足网络拥塞判断条件中任 意一项，按照第一预定补偿步长，增加非拥塞抖动范围的上限值。Exemplarily, the above data transmission method further includes: if none of the network packet loss parameters meets any of the network congestion judgment conditions, increasing the upper limit of the non-congested jitter range according to the first predetermined compensation step.
当传输报文的往返时延抖动大于拥塞判断阈值，但网络丢包参数均未满足网络拥塞判断 条件中任意一项，表明未发生网络拥塞。为了后续的网络拥塞判断更加精准且效率更高，可 适当增加非拥塞抖动范围的上限值，拓宽非拥塞抖动范围，从而实现对非拥塞抖动范围的设 定的优化。When the round-trip delay jitter of transmitted packets is greater than the congestion judgment threshold, but none of the network packet loss parameters meets any of the network congestion judgment conditions, it indicates that no network congestion occurs. In order to make the subsequent judgment of network congestion more accurate and efficient, the upper limit of the non-congested jitter range can be appropriately increased to widen the non-congested jitter range, so as to optimize the setting of the non-congested jitter range.
示例性的，上述数据传输方法还包括若当前带宽未下降或下降幅度未超出下降阈值，按 照第二预定补偿步长，增加非拥塞抖动范围的上限值。Exemplarily, the above data transmission method further includes increasing the upper limit of the non-congested jitter range according to the second predetermined compensation step if the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold.
当传输报文的往返时延抖动大于拥塞判断阈值，且当前带宽下降幅度超出下降阈值，但 网络丢包参数均未满足网络拥塞判断条件中任意一项，表明未发生网络拥塞。为了后续的网 络拥塞判断更加精准且效率更高，可适当增加非拥塞抖动范围的上限值，拓宽非拥塞抖动范 围，实现对非拥塞抖动范围的设定的优化。When the round-trip delay jitter of transmitted packets is greater than the congestion judgment threshold, and the current bandwidth drop exceeds the drop threshold, but none of the network packet loss parameters meets any of the network congestion judgment conditions, it indicates that no network congestion occurs. In order to make the subsequent judgment of network congestion more accurate and efficient, the upper limit of the non-congested jitter range can be appropriately increased to widen the non-congested jitter range and optimize the setting of the non-congested jitter range.
第三方面，本申请提供一种数据传输方法，包括：获取当前带宽，判断当前带宽的下降 幅度是否超出下降阈值；若当前带宽的下降幅度超出下降阈值，则获取网络丢包参数，并判 断网络丢包参数是否满足网络拥塞判断条件，网络丢包参数记录了传输报文在传输过程中数 据包的丢包信息；若网络丢包参数均未满足网络拥塞判断条件中任意一项，则按照第一预设 微步长缩小拥塞窗口。In the third aspect, the present application provides a data transmission method, including: obtaining the current bandwidth, and judging whether the drop rate of the current bandwidth exceeds the drop threshold; if the drop rate of the current bandwidth exceeds the drop threshold, obtaining network packet loss parameters, and judging the Whether the packet loss parameters meet the network congestion judgment conditions, the network packet loss parameters record the packet loss information of the data packets during the transmission of the transmission message; if the network packet loss parameters do not meet any of the network congestion judgment conditions, then follow the The congestion window is reduced by a preset microstep size.
在当前带宽的下降幅度超出下降阈值，但网络丢包参数均未满足网络拥塞判断条件中任 意一项时，确定数据传输出现的是随机丢包，并未发生网络拥塞，对拥塞窗口进行微调。避 免对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓解随机丢包所带 来的影响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络带宽整体的利用 率。When the decrease of the current bandwidth exceeds the drop threshold, but the network packet loss parameters do not meet any of the network congestion judgment conditions, it is determined that random packet loss occurs in data transmission and no network congestion occurs, and the congestion window is fine-tuned. Avoid large adjustments to the congestion window and avoid misjudgment of network congestion. On the basis of alleviating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network bandwidth. overall utilization.
示例性的，网络丢包参数包括连续丢包数目，网络拥塞判断条件包括连续丢包数目大于 等于两个；和/或，网络丢包参数包括丢包率，网络拥塞判断条件包括丢包率大于预设的拥塞 丢包率阈值；和/或，网络丢包参数包括丢包重传报文数目，网络拥塞判断条件包括丢包重传 报文数目大于预设的拥塞丢包重传报文阈值。Exemplarily, the network packet loss parameter includes the number of consecutive packet loss, and the network congestion judgment condition includes that the number of consecutive packet loss is greater than or equal to two; and/or, the network packet loss parameter includes a packet loss rate, and the network congestion judgment condition includes a packet loss rate greater than or equal to Preset congestion packet loss rate threshold; and/or, network packet loss parameters include the number of packet loss and retransmission messages, and network congestion judgment conditions include the number of packet loss and retransmission messages being greater than the preset congestion packet loss and retransmission message threshold .
根据第三方面，在第三方面的第一种可能中，上述数据传输方法还包括：若当前带宽未 下降或下降幅度未超出下降阈值，则按照第二预设微步长缩小拥塞窗口。According to the third aspect, in the first possibility of the third aspect, the above-mentioned data transmission method further includes: if the current bandwidth does not drop or the drop range does not exceed the drop threshold, shrinking the congestion window according to the second preset micro-step.
确定数据传输出现的是随机丢包，并未发生网络拥塞，对拥塞窗口进行微调。避免对拥 塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓解随机丢包所带来的影 响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络带宽整体的利用率。It is determined that random packet loss occurs during data transmission, and no network congestion occurs, and fine-tune the congestion window. Avoid large adjustments to the congestion window and avoid misjudgment of network congestion. On the basis of alleviating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network bandwidth. overall utilization.
根据第三方面，在第三方面的第二种可能中，上述数据传输方法还包括：若网络丢包参 数满足网络拥塞判断条件中任意一项，则获取传输报文的往返时延的抖动，判断往返时延的 抖动是否大于拥塞判断阈值；若往返时延的抖动小于等于拥塞判断阈值，则判断往返时延的 抖动是否处于非拥塞抖动范围内。According to the third aspect, in the second possibility of the third aspect, the above-mentioned data transmission method further includes: if the network packet loss parameter meets any one of the network congestion judgment conditions, obtaining the jitter of the round-trip delay of the transmitted message, It is judged whether the jitter of the round-trip delay is greater than the congestion judgment threshold; if the jitter of the round-trip delay is less than or equal to the congestion judgment threshold, it is judged whether the jitter of the round-trip delay is within the non-congested jitter range.
根据第三方面的第一种可能，在第三方面的第二种可能中，上述数据传输方法还包括： 若往返时延的抖动大于拥塞判断阈值，则按照第一预设粗步长缩小拥塞窗口，第一预设粗步 长大于第一预设微步长。According to the first possibility of the third aspect, in the second possibility of the third aspect, the above data transmission method further includes: if the jitter of the round-trip delay is greater than the congestion judgment threshold, reducing the congestion according to the first preset coarse step window, the first preset coarse step size is larger than the first preset micro step size.
在经历了上述实施例中的带宽和网络丢包参数的判定后，引入往返时延的抖动，进一步 提高判定网络拥塞引起的丢包的准确性。当当前带宽的下降幅度超出下降阈值，且网络丢包 参数满足网络拥塞判断条件中任意一项，且传输报文的往返时延的抖动大于拥塞判断阈值时， 确定发生网络拥塞丢包，较大幅度缩小拥塞窗口，以缓解网络拥塞情况。After the determination of bandwidth and network packet loss parameters in the above embodiments, the round-trip delay jitter is introduced to further improve the accuracy of determining packet loss caused by network congestion. When the decrease of the current bandwidth exceeds the decrease threshold, and the network packet loss parameters meet any one of the network congestion judgment conditions, and the jitter of the round-trip delay of the transmitted message is greater than the congestion judgment threshold, it is determined that the packet loss due to network congestion occurs. Reduce the congestion window by a large amount to alleviate network congestion.
根据第三方面的第一种可能，在第三方面的第三种可能中，上述数据传输方法还包括： 若往返时延的抖动大于非拥塞抖动范围的上限值，则按照第三预设微步长缩小拥塞窗口。According to the first possibility of the third aspect, in the third possibility of the third aspect, the above data transmission method further includes: if the round-trip delay jitter is greater than the upper limit value of the non-congested jitter range, according to the third preset Microstepping shrinks the congestion window.
确定数据传输出现的是随机丢包，并未发生网络拥塞，对拥塞窗口进行微调。避免对拥 塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓解随机丢包所带来的影 响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络带宽整体的利用率。It is determined that random packet loss occurs during data transmission, and no network congestion occurs, and fine-tune the congestion window. Avoid large adjustments to the congestion window and avoid misjudgment of network congestion. On the basis of alleviating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network bandwidth. overall utilization.
根据第三方面的第一种可能，在第三方面的第四种可能中，上述数据传输方法还包括： 若往返时延的抖动小于非拥塞抖动范围的下限值，则按照第四预设微步长缩小拥塞窗口。According to the first possibility of the third aspect, in the fourth possibility of the third aspect, the above data transmission method further includes: Microstepping shrinks the congestion window.
确定数据传输出现的是随机丢包，并未发生网络拥塞，对拥塞窗口进行微调。避免对拥 塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓解随机丢包所带来的影 响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络带宽整体的利用率。It is determined that random packet loss occurs during data transmission, and no network congestion occurs, and fine-tune the congestion window. Avoid large adjustments to the congestion window and avoid misjudgment of network congestion. On the basis of alleviating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network bandwidth. overall utilization.
根据第三方面的第一种可能，在第三方面的第五种可能中，上述数据传输方法还包括： 若往返时延的抖动处于非拥塞抖动范围内，则按照第五预设微步长缩小拥塞窗口。According to the first possibility of the third aspect, in the fifth possibility of the third aspect, the above data transmission method further includes: Reduce the congestion window.
确定数据传输出现的是随机丢包，并未发生网络拥塞，对拥塞窗口进行微调。避免对拥 塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓解随机丢包所带来的影 响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络带宽整体的利用率。It is determined that random packet loss occurs during data transmission, and no network congestion occurs, and fine-tune the congestion window. Avoid large adjustments to the congestion window and avoid misjudgment of network congestion. On the basis of alleviating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network bandwidth. overall utilization.
第四方面，本申请提供了一种数据传输装置，包括：第一判断单元，用于获取发送端所 发送的传输报文的往返时延的抖动，判断传输报文的往返时延的抖动是否大于拥塞判断阈值， 往返时延的抖动表示传输报文的往返时延偏离基准往返时延的幅度；第二判断单元，用于若 往返时延的抖动大于拥塞判断阈值，则获取当前带宽，判断当前带宽的下降幅度是否超出下 降阈值；第一处理单元，用于若当前带宽未下降或下降幅度未超出下降阈值，则按照第一预 设微步长缩小拥塞窗口。。In a fourth aspect, the present application provides a data transmission device, including: a first judging unit, configured to obtain the jitter of the round-trip delay of the transmission message sent by the sending end, and determine whether the jitter of the round-trip delay of the transmission message is Greater than the congestion judgment threshold, the jitter of the round-trip delay indicates the magnitude of the round-trip delay of the transmitted message deviating from the reference round-trip delay; the second judgment unit is used to obtain the current bandwidth if the jitter of the round-trip delay is greater than the congestion judgment threshold, and judge Whether the decrease of the current bandwidth exceeds the decrease threshold; the first processing unit is configured to reduce the congestion window according to the first preset micro-step if the current bandwidth does not decrease or the decrease does not exceed the decrease threshold. .
根据第四方面，在第四方面的第一种可能中，上述数据传输装置还包括：第三判断单元， 用于若当前带宽的下降幅度超出下降阈值，则获取网络丢包参数，判断网络丢包参数是否满 足网络拥塞判断条件，网络丢包参数记录了传输报文在传输过程中数据包的丢包信息，网络 拥塞条件包括至少一项条件；第二处理单元，用于若网络丢包参数满足网络拥塞判断条件中 任意一项，则按照第一预设粗步长缩小拥塞窗口，第一预设粗步长大于第一预设微步长。According to the fourth aspect, in the first possibility of the fourth aspect, the above-mentioned data transmission device further includes: a third judging unit, configured to obtain network packet loss parameters if the current bandwidth drop exceeds a drop threshold, and judge the network packet loss Whether the packet parameter meets the network congestion judgment condition, the network packet loss parameter records the packet loss information of the data packet in the transmission process of the transmission message, and the network congestion condition includes at least one condition; the second processing unit is used for if the network packet loss parameter If any one of the network congestion judgment conditions is satisfied, the congestion window is reduced according to a first preset coarse step size, and the first preset coarse step size is larger than the first preset micro step size.
示例性的，网络丢包参数包括连续丢包数目，网络拥塞判断条件包括连续丢包数目大于 等于两个；和/或，网络丢包参数包括丢包率，网络拥塞判断条件包括丢包率大于预设的拥塞 丢包率阈值；和/或，网络丢包参数包括丢包重传报文数目，网络拥塞判断条件包括丢包重传 报文数目大于预设的拥塞丢包重传报文阈值。Exemplarily, the network packet loss parameter includes the number of consecutive packet loss, and the network congestion judgment condition includes that the number of consecutive packet loss is greater than or equal to two; and/or, the network packet loss parameter includes a packet loss rate, and the network congestion judgment condition includes a packet loss rate greater than or equal to Preset congestion packet loss rate threshold; and/or, network packet loss parameters include the number of packet loss and retransmission messages, and network congestion judgment conditions include the number of packet loss and retransmission messages being greater than the preset congestion packet loss and retransmission message threshold .
根据第四方面的第一种可能，在第四方面的第二种可能中，第四判断单元，用于若往返 时延的抖动小于等于拥塞判断阈值，则判断往返时延的抖动是否处于非拥塞抖动范围内，非 拥塞抖动范围为传输报文在传输过程中未发生网络拥塞时往返时延的抖动的变化范围，且拥 塞判断阈值大于非拥塞抖动范围的上限值；第三处理单元，用于若往返时延的抖动处于非拥 塞抖动范围内，保持拥塞窗口不变。According to the first possibility of the fourth aspect, in the second possibility of the fourth aspect, the fourth judging unit is configured to judge whether the jitter of the round-trip delay is less than or equal to the congestion judging threshold Within the congestion jitter range, the non-congestion jitter range is the variation range of the round-trip delay jitter when no network congestion occurs during the transmission of the transmission message, and the congestion judgment threshold is greater than the upper limit of the non-congestion jitter range; the third processing unit, It is used to keep the congestion window unchanged if the jitter of the round-trip delay is within the non-congested jitter range.
根据第四方面的第二种可能，在第四方面的第三种可能中，上述数据传输装置还包括： 第四处理单元，用于若往返时延的抖动小于非拥塞抖动范围的下限值，则按照第二预设微步 长扩大拥塞窗口。According to the second possibility of the fourth aspect, in the third possibility of the fourth aspect, the above-mentioned data transmission device further includes: a fourth processing unit, configured to if the jitter of the round-trip delay is less than the lower limit value of the non-congested jitter range , the congestion window is expanded according to the second preset micro-step size.
根据第四方面的第二种可能，在第四方面的第四种可能中，上述数据传输装置还包括： 第五处理单元，用于若往返时延的抖动大于非拥塞抖动范围的上限值，则按照第三预设微步 长缩小拥塞窗口。According to the second possibility of the fourth aspect, in the fourth possibility of the fourth aspect, the above-mentioned data transmission device further includes: a fifth processing unit, configured to if the jitter of the round-trip delay is greater than the upper limit of the non-congested jitter range , then reduce the congestion window according to the third preset micro-step size.
根据第四方面的第一种可能，在第四方面的第五种可能中，上述数据传输装置还包括： 第六处理单元，用于若网络丢包参数均未满足网络拥塞判断条件中任意一项，则按照第四预 设微步长缩小拥塞窗口。According to the first possibility of the fourth aspect, in the fifth possibility of the fourth aspect, the above-mentioned data transmission device further includes: a sixth processing unit, configured to, if none of the network packet loss parameters meet any of the network congestion judgment conditions item, the congestion window is reduced according to the fourth preset micro-step size.
示例性的，上述数据传输装置还包括：第一调整单元，被配置为若当前带宽未下降或下 降幅度未超出下降阈值，按照第一预定补偿步长，增加非拥塞抖动范围的上限值。Exemplarily, the above data transmission device further includes: a first adjustment unit configured to increase the upper limit of the non-congested jitter range according to a first predetermined compensation step if the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold.
示例性的，上述数据传输装置还包括：第二调整单元，被配置为若网络丢包参数均未满 足网络拥塞判断条件中任意一项，按照第二预定补偿步长，增加非拥塞抖动范围的上限值。Exemplarily, the above data transmission device further includes: a second adjustment unit configured to increase the non-congested jitter range according to the second predetermined compensation step size if the network packet loss parameters do not meet any of the network congestion judgment conditions. Upper limit.
第五方面，本申请提供了一种数据传输装置，包括：第一判断单元，用于获取发送的传 输报文的往返时延的抖动，判断传输报文的往返时延的抖动是否大于拥塞判断阈值，往返时 延的抖动表征传输报文的往返时延偏离基准往返时延的幅度；第二判断单元，用于若往返时 延的抖动大于拥塞判断阈值，则获取网络丢包参数，判断网络丢包参数是否满足网络拥塞判 断条件，网络丢包参数记录了传输报文在传输过程中数据包的丢包信息；第一处理单元，用 于若网络丢包参数均未满足网络拥塞判断条件中任意一项，则按照第一预设微步长缩小拥塞 窗口，网络拥塞条件包括至少一项条件。In a fifth aspect, the present application provides a data transmission device, including: a first judging unit, configured to obtain the jitter of the round-trip delay of the transmitted transmission message, and determine whether the jitter of the round-trip delay of the transmission message is greater than the congestion judgment Threshold, the jitter of the round-trip delay represents the magnitude of the round-trip delay of the transmitted message from the reference round-trip delay; the second judgment unit is used to obtain network packet loss parameters if the jitter of the round-trip delay is greater than the congestion judgment threshold, and judge the network Whether the packet loss parameter meets the network congestion judgment condition, the network packet loss parameter records the packet loss information of the data packet in the transmission process of the transmission message; the first processing unit is used if the network packet loss parameter does not meet the network congestion judgment condition Any item, the congestion window is reduced according to the first preset micro-step size, and the network congestion condition includes at least one condition.
示例性的，网络丢包参数包括连续丢包数目，网络拥塞判断条件包括连续丢包数目大于 等于两个；和/或，网络丢包参数包括丢包率，网络拥塞判断条件包括丢包率大于预设的拥塞 丢包率阈值；和/或，网络丢包参数包括丢包重传报文数目，网络拥塞判断条件包括丢包重传 报文数目大于预设的拥塞丢包重传报文阈值。Exemplarily, the network packet loss parameter includes the number of consecutive packet loss, and the network congestion judgment condition includes that the number of consecutive packet loss is greater than or equal to two; and/or, the network packet loss parameter includes a packet loss rate, and the network congestion judgment condition includes a packet loss rate greater than or equal to Preset congestion packet loss rate threshold; and/or, network packet loss parameters include the number of packet loss and retransmission messages, and network congestion judgment conditions include the number of packet loss and retransmission messages being greater than the preset congestion packet loss and retransmission message threshold .
根据第五方面，在第五方面的第一种可能中，上述数据传输装置还包括：第三判断单元， 用于若网络丢包参数满足网络拥塞判断条件中任意一项，则获取当前带宽，判断当前带宽的 下降幅度是否超出下降阈值；第二处理单元，用于若当前带宽的下降幅度超出下降阈值，则 按照第一预设粗步长缩小拥塞窗口，第一预设粗步长大于第一预设微步长。According to the fifth aspect, in the first possibility of the fifth aspect, the above-mentioned data transmission device further includes: a third judging unit, configured to acquire the current bandwidth if the network packet loss parameter satisfies any one of the network congestion judging conditions, Judging whether the drop rate of the current bandwidth exceeds the drop threshold; the second processing unit is configured to reduce the congestion window according to the first preset coarse step size if the drop rate of the current bandwidth exceeds the drop threshold value, and the first preset coarse step size is greater than the first preset coarse step size A preset microstep size.
根据第五方面，在第五方面的第二种可能中，上述数据传输装置还包括：第四判断单元， 用于若往返时延的抖动小于等于拥塞判断阈值，则判断往返时延的抖动是否处于非拥塞抖动 范围内，非拥塞抖动范围为传输报文在传输过程中未发生网络拥塞时往返时延的抖动的变化 范围，且拥塞判断阈值大于非拥塞抖动范围的上限值；第三处理单元，用于若往返时延抖动 处于非拥塞抖动范围内，保持拥塞窗口不变。According to the fifth aspect, in the second possibility of the fifth aspect, the above-mentioned data transmission device further includes: a fourth judging unit, configured to judge whether the jitter of the round-trip delay is less than or equal to the congestion judging threshold Be in the non-congested jitter range, the non-congested jitter range is the variation range of the round-trip delay jitter when there is no network congestion in the transmission process of the transmission message, and the congestion judgment threshold is greater than the upper limit of the non-congested jitter range; the third processing The unit is used to keep the congestion window unchanged if the round-trip delay jitter is within the non-congested jitter range.
根据第五方面的第二种可能，在第五方面的第三种可能中，上述数据传输装置还包括： 第四处理单元，用于若往返时延的抖动小于非拥塞抖动范围的下限，则按照第二预设微步长 扩大拥塞窗口。According to the second possibility of the fifth aspect, in the third possibility of the fifth aspect, the above-mentioned data transmission device further includes: a fourth processing unit configured to: if the round-trip delay jitter is less than the lower limit of the non-congested jitter range, then The congestion window is expanded according to a second preset micro-step size.
根据第五方面的第二种可能，在第五方面的第四种可能中，上述数据传输装置还包括： 第五处理单元，用于若往返时延的抖动大于非拥塞抖动范围的上限，则按照第三预设微步长 缩小拥塞窗口。According to the second possibility of the fifth aspect, in the fourth possibility of the fifth aspect, the above-mentioned data transmission device further includes: a fifth processing unit, configured to: if the round-trip delay jitter is greater than the upper limit of the non-congested jitter range, then The congestion window is reduced according to a third preset micro-step size.
根据第五方面的第一种可能，在第五方面的第五种可能中，上述数据传输装置还包括： 第六处理单元，用于若当前带宽未下降或下降幅度未超出下降阈值，则按照第四预设微步长 缩小拥塞窗口值。According to the first possibility of the fifth aspect, in the fifth possibility of the fifth aspect, the above-mentioned data transmission device further includes: a sixth processing unit, configured to: The fourth preset micro-step reduces the congestion window value.
示例性的，上述数据传输装置还包括：第一调整单元，被配置为若网络丢包参数均未满 足网络拥塞判断条件中任意一项，按照第一预定补偿步长，增加非拥塞抖动范围的上限值。Exemplarily, the above data transmission device further includes: a first adjustment unit configured to increase the non-congested jitter range according to the first predetermined compensation step size if the network packet loss parameters do not meet any of the network congestion judgment conditions. Upper limit.
示例性的，上述数据传输装置还包括：第二调整单元，被配置为若当前带宽未下降或下 降幅度未超出下降阈值，按照第二预定补偿步长，增加非拥塞抖动范围的上限值。Exemplarily, the above data transmission device further includes: a second adjustment unit configured to increase the upper limit of the non-congested jitter range according to a second predetermined compensation step if the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold.
第六方面，本申请提供一种数据传输装置，包括：第一判断单元，用于获取当前带宽， 判断当前带宽的下降幅度是否超出下降阈值；第二判断单元，用于若当前带宽的下降幅度超 出下降阈值，则获取网络丢包参数，并判断网络丢包参数是否满足网络拥塞判断条件，网络 丢包参数记录了传输报文在传输过程中数据包的丢包信息；第一处理单元，用于若网络丢包 参数均未满足网络拥塞判断条件中任意一项，则按照第一预设微步长缩小拥塞窗口。In a sixth aspect, the present application provides a data transmission device, including: a first judging unit, configured to obtain the current bandwidth, and judge whether the drop rate of the current bandwidth exceeds a drop threshold; a second judging unit, used to determine if the drop rate of the current bandwidth Exceeding the falling threshold, then obtain the network packet loss parameter, and judge whether the network packet loss parameter meets the network congestion judgment condition, the network packet loss parameter records the packet loss information of the data packet in the transmission process of the transmission message; the first processing unit uses If none of the network packet loss parameters meets any one of the network congestion judgment conditions, the congestion window is reduced according to the first preset micro-step size.
示例性的，网络丢包参数包括连续丢包数目，网络拥塞判断条件包括连续丢包数目大于 等于两个；和/或，网络丢包参数包括丢包率，网络拥塞判断条件包括丢包率大于预设的拥塞 丢包率阈值；和/或，网络丢包参数包括丢包重传报文数目，网络拥塞判断条件包括丢包重传 报文数目大于预设的拥塞丢包重传报文阈值。Exemplarily, the network packet loss parameter includes the number of consecutive packet loss, and the network congestion judgment condition includes that the number of consecutive packet loss is greater than or equal to two; and/or, the network packet loss parameter includes a packet loss rate, and the network congestion judgment condition includes a packet loss rate greater than or equal to Preset congestion packet loss rate threshold; and/or, network packet loss parameters include the number of packet loss and retransmission messages, and network congestion judgment conditions include the number of packet loss and retransmission messages being greater than the preset congestion packet loss and retransmission message threshold .
根据第六方面，在第六方面的第一种可能中，上述数据传输装置还包括：第三判断单元， 用于若网络丢包参数满足网络拥塞判断条件中任意一项，则获取传输报文的往返时延的抖动， 判断往返时延的抖动是否大于拥塞判断阈值；第二处理单元，用于若往返时延的抖动大于拥 塞判断阈值，则按照第一预设粗步长缩小拥塞窗口。According to the sixth aspect, in the first possibility of the sixth aspect, the above-mentioned data transmission device further includes: a third judging unit, configured to acquire the transmission message if the network packet loss parameter satisfies any one of the network congestion judging conditions The jitter of the round-trip delay determines whether the jitter of the round-trip delay is greater than the congestion judgment threshold; the second processing unit is configured to shrink the congestion window according to the first preset coarse step size if the jitter of the round-trip delay is greater than the congestion judgment threshold.
根据第六方面的第一种可能，在第六方面的第二种可能中，上述数据传输装置还包括： 第三处理单元，用于若当前带宽未下降或下降幅度未超出下降阈值，则按照第二预设微步长 缩小拥塞窗口。According to the first possibility of the sixth aspect, in the second possibility of the sixth aspect, the above-mentioned data transmission device further includes: a third processing unit, configured to: The second preset micro-step size reduces the congestion window.
根据第六方面，在第六方面的第三种可能中，上述数据传输装置还包括：第四判断单元， 用于若往返时延的抖动小于等于拥塞判断阈值，则判断往返时延的抖动是否处于非拥塞抖动 范围内；第六处理单元，用于若往返时延的抖动处于非拥塞抖动范围内，则按照第五预设微 步长缩小拥塞窗口。According to the sixth aspect, in the third possibility of the sixth aspect, the above-mentioned data transmission device further includes: a fourth judging unit, configured to judge whether the jitter of the round-trip delay is less than or equal to the congestion judging threshold within the range of non-congested jitter; the sixth processing unit is configured to, if the jitter of the round-trip delay is within the range of non-congested jitter, reduce the congestion window according to the fifth preset micro-step.
根据第六方面的第三种可能，在第六方面的第四种可能中，上述数据传输装置还包括： 第四处理单元，用于若往返时延的抖动大于非拥塞抖动范围的上限值，则按照第三预设微步 长缩小拥塞窗口。According to the third possibility of the sixth aspect, in the fourth possibility of the sixth aspect, the above-mentioned data transmission device further includes: a fourth processing unit, configured to if the jitter of the round-trip delay is greater than the upper limit value of the non-congested jitter range , then reduce the congestion window according to the third preset micro-step size.
根据第六方面的第三种可能，在第六方面的第五种可能中，上述数据传输装置还包括： 第五处理单元609，用于若往返时延的抖动小于非拥塞抖动范围的下限值，则按照第四预设 微步长缩小拥塞窗口。According to the third possibility of the sixth aspect, in the fifth possibility of the sixth aspect, the above-mentioned data transmission device further includes: a fifth processing unit 609, configured to if the jitter of the round-trip delay is less than the lower limit of the non-congested jitter range value, the congestion window is reduced according to the fourth preset micro-step size.
第七方面，本发明实施例提供了一种数据传输设备，包括存储器、处理器及存储在存 储器上并可在处理器上运行的程序，处理器执行程序时实现如上述技术方案中的数据传输方 法。In the seventh aspect, the embodiment of the present invention provides a data transmission device, including a memory, a processor, and a program stored in the memory and operable on the processor. When the processor executes the program, the data transmission in the above technical solution is realized. method.
第八方面，本发明实施例提供了一种存储介质，该存储介质上存储有程序，该程序被处 理器执行时实现如上述技术方案中的数据传输方法。In an eighth aspect, an embodiment of the present invention provides a storage medium, on which a program is stored, and when the program is executed by a processor, the data transmission method in the above technical solution is implemented.
本申请提供了一种数据传输方法、装置、发送端与存储介质，可利用传输报文的往返时 延的抖动和网络的带宽，共同进行网络是否发生拥塞的判断。在判断往返时延的抖动大于拥 塞判断阈值，但当前带宽未下降或下降幅度未超出下降阈值时，判断数据传输出现的是随机 丢包，并未发生网络拥塞，对拥塞窗口进行微调。或者，利用传输报文的往返时延的抖动和 网络丢包参数，共同进行网络是否发生拥塞的判断。在判断往返时延的抖动大于拥塞判断阈 值，但网络丢包参数均未满足网络拥塞判断条件中任意一项时，判断数据传输出现的是随机 丢包，并未发生网络拥塞，对拥塞窗口进行微调。并未对拥塞窗口进行大幅度调整，避免了 出现网络拥塞误判的情况，从而在缓解随机丢包所带来的影响的基础上，最大限度的保证了 网络带宽的利用率，从而提高了网络带宽整体的利用率。This application provides a data transmission method, device, sending end and storage medium, which can use the jitter of the round-trip delay of the transmitted message and the bandwidth of the network to jointly judge whether the network is congested. When it is judged that the jitter of the round-trip delay is greater than the congestion judgment threshold, but the current bandwidth does not drop or the drop rate does not exceed the drop threshold, it is judged that random packet loss occurs in data transmission and there is no network congestion, and the congestion window is fine-tuned. Or, use the jitter of the round-trip delay of the transmitted message and the network packet loss parameter to jointly judge whether the network is congested. When it is judged that the jitter of the round-trip delay is greater than the congestion judgment threshold, but the network packet loss parameters do not meet any of the network congestion judgment conditions, it is judged that random packet loss occurs in data transmission, and no network congestion occurs, and the congestion window is checked. fine-tuning. The congestion window has not been greatly adjusted, which avoids misjudgment of network congestion, so that on the basis of mitigating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network performance. Overall bandwidth utilization.
附图说明Description of drawings
图1为本发明实施例的数据传输系统的架构示意图；FIG. 1 is a schematic diagram of the architecture of a data transmission system according to an embodiment of the present invention;
图2为本发明一实施例中一种数据传输的方法的流程图；FIG. 2 is a flowchart of a data transmission method in an embodiment of the present invention;
图3为本发明另一实施例中一种数据传输方法的流程图；Fig. 3 is a flowchart of a data transmission method in another embodiment of the present invention;
图4为本发明又一实施例中一种数据传输方法的流程图；4 is a flowchart of a data transmission method in another embodiment of the present invention;
图5为本发明一实施例中一种数据传输装置的结构示意图；5 is a schematic structural diagram of a data transmission device in an embodiment of the present invention;
图6为本发明另一实施例中一种数据传输装置的结构示意图；6 is a schematic structural diagram of a data transmission device in another embodiment of the present invention;
图7为本发明又一实施例中一种数据传输装置的结构示意图；7 is a schematic structural diagram of a data transmission device in another embodiment of the present invention;
图8为本发明一实施例中的一种发送端的硬件结构示意图。FIG. 8 is a schematic diagram of a hardware structure of a sending end in an embodiment of the present invention.
具体实施方式Detailed ways
本发明实施例提供了一种数据传输方法、装置、发送端与存储介质，可应用于利用传输 控制协议(TCP，Transmission Control Protocol)传输数据的过程。图1为本发明实施例 的数据传输系统的架构示意图。如图1所示，数据传输系统包括发送端、无线通信网络和接 收端。发送端和接收端通过无线通信网络进行数据传输。具体的，发送端发送的数据可通过 无线通信网络传输给接收端，接收端接收发送端发送来的数据。其中，发送端可发送报文， 报文包括至少一个数据包，数据包中包括数据，从而实现数据的传输。在一个示例中，发送 端可以为电脑、手机、平板电脑等用户终端，接收端可为公有云服务器或私有云服务器，无 线通信网络可包括无线网络、无线保真(WiFi，Wireless-Fidelity)或公网等。在另外一个 示例中，发送端可为一数据中心，接收端可为另一数据中心，接收端的数据中心可作为发送 端的数据中心的容灾备份数据中心。或者，发送端和接收端可跨广域网访问。在发送端和接 收端通过无线通信网络进行数据传输的过程中，可能存在随机丢包的场景。在本发明实施例 中，可区分随机丢包和网络拥塞丢包，从而针对不同原因引起的丢包，采取不同的相应措施， 以达到最大限度利用网络带宽，并在传输报文的传输过程中尽量避免网络拥塞的目的。The embodiment of the present invention provides a data transmission method, device, sending end and storage medium, which can be applied to the process of transmitting data using Transmission Control Protocol (TCP, Transmission Control Protocol). Fig. 1 is a schematic diagram of the architecture of a data transmission system according to an embodiment of the present invention. As shown in Figure 1, the data transmission system includes a sending end, a wireless communication network and a receiving end. The sending end and the receiving end carry out data transmission through the wireless communication network. Specifically, the data sent by the sending end can be transmitted to the receiving end through the wireless communication network, and the receiving end receives the data sent by the sending end. Wherein, the sending end may send a message, the message includes at least one data packet, and the data packet includes data, so as to realize data transmission. In one example, the sending end may be a user terminal such as a computer, a mobile phone, or a tablet computer, the receiving end may be a public cloud server or a private cloud server, and the wireless communication network may include a wireless network, wireless fidelity (WiFi, Wireless-Fidelity) or public network, etc. In another example, the sending end may be a data center, the receiving end may be another data center, and the data center at the receiving end may serve as a disaster recovery backup data center for the data center at the sending end. Alternatively, the sender and receiver can be accessed across a wide area network. During the data transmission process between the sending end and the receiving end through the wireless communication network, there may be random packet loss scenarios. In the embodiment of the present invention, it is possible to distinguish between random packet loss and network congestion packet loss, so that different corresponding measures are taken for packet loss caused by different reasons, so as to maximize the use of network bandwidth, and during the transmission of packets The purpose of avoiding network congestion as much as possible.
图2为本发明一实施例中一种数据传输方法的流程图。该数据传输方法可应用于数据传 输过程中的发送端，如图1所示，数据传输方法可包括步骤S101至步骤S112。FIG. 2 is a flowchart of a data transmission method in an embodiment of the present invention. The data transmission method can be applied to the sending end in the data transmission process. As shown in FIG. 1, the data transmission method can include steps S101 to S112.
在步骤S101中，获取发送的传输报文的往返时延的抖动，判断传输报文的往返时延的抖 动是否大于拥塞判断阈值。In step S101, the jitter of the round-trip delay of the transmitted transmission message is obtained, and it is judged whether the jitter of the round-trip delay of the transmission message is greater than the congestion judgment threshold.
传输报文包括至少一个数据包。数据包的往返时延(RTT，Round Trip Time)为从发送 端发送该数据包开始计时，到接收到接收端返回的该数据包的回复确认消息(Acknowledgement，ACK)为止的时长。传输报文的往返时延可为传输报文中任意一个数据包的往返时延，或者为传输报文中多个数据包的往返时延的平均值或中值。The transmission message includes at least one data packet. The round-trip delay (RTT, Round Trip Time) of a data packet is the time period from when the sender sends the data packet to when the reply acknowledgment message (Acknowledgment, ACK) of the data packet returned by the receiving end is received. The round-trip delay of the transmitted message may be the round-trip delay of any data packet in the transmitted message, or the average or median value of the round-trip delays of multiple data packets in the transmitted message.
往返时延的抖动表征传输报文的往返时延偏离基准往返时延的幅度。基准往返时延为在 非网络拥塞的场景下的传输报文的往返时延。具体的，基准往返时延可在非网络拥塞的场景 下经过多次测算得到。The jitter of the round-trip delay represents the extent to which the round-trip delay of the transmitted message deviates from the reference round-trip delay. The base round-trip delay is the round-trip delay of transmitting packets in a non-network congestion scenario. Specifically, the reference round-trip delay can be obtained through multiple calculations in a non-network congestion scenario.
拥塞判断阈值为判断网络是否发生拥塞的判断条件之一。若传输报文的往返时延的抖动 大于拥塞判断阈值，则判断数据传输可能发生网络拥塞，需要通过进一步的网络拥塞判定， 确定是否确实发生网络拥塞。The congestion judgment threshold is one of the judgment conditions for judging whether the network is congested. If the jitter of the round-trip delay of the transmitted message is greater than the congestion judgment threshold, it is judged that network congestion may occur in data transmission, and further network congestion judgment is required to determine whether network congestion does occur.
在步骤S102中，若往返时延抖动大于拥塞判断阈值，则获取当前带宽，判断当前带宽的 下降幅度是否超出下降阈值。In step S102, if the round-trip delay jitter is greater than the congestion judgment threshold, the current bandwidth is obtained, and it is judged whether the decrease in the current bandwidth exceeds the decrease threshold.
进一步的网络拥塞判定可利用带宽进行。数据传输装置实时采集网络的带宽。为了可及 时获取得到当前带宽，在一个示例中，可利用数据传输装置采集的历史带宽，得到带宽变化 趋势，并利用上一次采集到的网络的实际带宽，结合带宽变化趋势，估算得到当前带宽。利 用当前带宽进行判断，具体的，可判断当前带宽是否下降以及当前带宽的下降幅度是否超出 下降阈值。当前带宽的下降幅度为当前带宽与上一次带宽的差值。下降阈值可根据工作场景、 工作需求或工作经验设置，在此并不限定。在往返时延的抖动大于拥塞判断阈值的基础上， 增加带宽判定，进一步确定数据传输是否发生网络拥塞。Further network congestion determinations can be made using bandwidth. The data transmission device collects the bandwidth of the network in real time. In order to obtain the current bandwidth in time, in an example, the historical bandwidth collected by the data transmission device can be used to obtain the bandwidth change trend, and the actual bandwidth of the network collected last time can be combined with the bandwidth change trend to estimate the current bandwidth. The current bandwidth is used for judgment. Specifically, it can be judged whether the current bandwidth has decreased and whether the decrease of the current bandwidth exceeds the decrease threshold. The decrease range of the current bandwidth is the difference between the current bandwidth and the previous bandwidth. The drop threshold can be set according to work scenarios, work requirements or work experience, and is not limited here. On the basis that the jitter of the round-trip delay is greater than the congestion judgment threshold, bandwidth judgment is added to further determine whether network congestion occurs in data transmission.
在步骤S103中，若当前带宽未下降或下降幅度未超出下降阈值，则按照第一预设微步长 缩小拥塞窗口。In step S103, if the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold, the congestion window is reduced according to the first preset micro-step.
若当前带宽未下降或者当前带宽的下降幅度未超过下降阈值，则表明带宽未发生明显的 网络拥塞变化。但由于往返时延的抖动大于拥塞判断阈值，传输报文的往返时延较长，需要 对拥塞窗口进行细微平缓的调节。具体的，可按照第一预设微步长缩小拥塞串口。第一预设 微步长为细微调节使用的步长。在本方案中，步长为每次调节所增大或减小的拥塞窗口的值。 第一预设微步长可根据工作场景、工作需求或工作经验设置，在此并不限定。在一个示例中， 可以设置调节周期，每经过一个调节周期，按照第一预设微步长缩小一次拥塞窗口。比如， 每经过一次传输报文的往返时延的时长，则将拥塞窗口的值减一。If the current bandwidth does not decrease or the decrease of the current bandwidth does not exceed the decrease threshold, it indicates that there is no obvious change in network congestion in the bandwidth. However, since the jitter of the round-trip delay is greater than the congestion judgment threshold, the round-trip delay of transmitted packets is relatively long, and the congestion window needs to be finely and gently adjusted. Specifically, the congested serial port may be reduced according to the first preset micro-step size. The first preset Microstep is the step size used for fine adjustments. In this solution, the step size is the value of the congestion window that is increased or decreased by each adjustment. The first preset micro-step size can be set according to the work scene, work requirement or work experience, and is not limited here. In an example, an adjustment period may be set, and the congestion window is reduced according to a first preset micro-step every time an adjustment period passes. For example, the value of the congestion window is decremented by one each time the length of the round-trip delay for transmitting a packet passes.
在一个示例中，在发现数据传输过程中出现丢包，不管是随机丢包还是网络拥塞丢包， 均需要将丢失的数据包重新进行传输。In an example, it is found that packet loss occurs during data transmission, whether it is random packet loss or network congestion packet loss, it is necessary to retransmit the lost data packet.
在本发明实施例中，在判断往返时延的抖动大于拥塞判断阈值，但当前带宽未下降或下 降幅度未超出下降阈值时，判断数据传输出现的是随机丢包，并未发生网络拥塞，对拥塞窗 口进行微调。并未对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓 解随机丢包所带来的影响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络 带宽整体的利用率。In the embodiment of the present invention, when it is judged that the jitter of the round-trip delay is greater than the congestion judgment threshold, but the current bandwidth does not drop or the drop does not exceed the drop threshold, it is judged that random packet loss occurs in data transmission and no network congestion occurs. The congestion window is fine-tuned. The congestion window has not been greatly adjusted, which avoids misjudgment of network congestion, so that on the basis of mitigating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network performance. Overall bandwidth utilization.
尤其是在带宽和时延均较大的网络中，若一旦发现丢包，就大幅度减小拥塞窗口，这样， 如果只是发生随机丢包，则会对网络带宽的利用率造成不必要的影响。本发明实施例能够区 别随机丢包和网络拥塞丢包，在带宽和时延均较大的网络中能够及时有效的优化网络的拥塞 判断机制。Especially in a network with large bandwidth and delay, if packet loss is found, the congestion window will be greatly reduced. In this way, if only random packet loss occurs, it will have an unnecessary impact on the utilization of network bandwidth . The embodiment of the present invention can distinguish between random packet loss and network congestion packet loss, and can timely and effectively optimize the congestion judgment mechanism of the network in a network with large bandwidth and time delay.
在步骤S104中，若当前带宽的下降幅度超出下降阈值，则获取网络丢包参数，判断网络 丢包参数是否满足网络拥塞判断条件。In step S104, if the drop rate of the current bandwidth exceeds the drop threshold, the network packet loss parameter is obtained, and it is judged whether the network packet loss parameter satisfies the network congestion judgment condition.
若当前带宽的下降幅度超出下降阈值，则表明网络的带宽大幅度下降，很有可能发生了 网络拥塞。可进行进一步的网络拥塞判定。具体的，可通过网络丢包参数判定。网络丢包参 数表示所述传输报文在传输过程中数据包的丢包信息，比如，网络丢包参数可包括连续丢包 数目、丢包率、丢包重传报文数目等参数中的一项或多项。If the decrease in the current bandwidth exceeds the drop threshold, it indicates that the network bandwidth has dropped significantly, and network congestion is likely to occur. Further network congestion determinations may be made. Specifically, it can be determined through network packet loss parameters. The network packet loss parameter represents the packet loss information of the data packet in the transmission process of the transmission message. For example, the network packet loss parameter can include one of the parameters such as the number of consecutive packet loss, the packet loss rate, and the number of packets lost and retransmitted. item or items.
网络拥塞条件包括至少一项条件，具体的，网络拥塞判断条件可以根据网络丢包参数对 应设定。网络拥塞判断条件用于针对网络丢包参数进行的网络拥塞判断。比如，若网络丢包 参数包括连续丢包数目，则网络拥塞判断条件包括连续丢包数目大于等于两个。也就是说， 若发生连续丢包，表示网络丢包参数满足网络拥塞判断条件。若网络丢包参数包括丢包率， 则网络拥塞判断条件包括丢包率大于预设的拥塞丢包率阈值。预设的拥塞丢包率阈值可根据 工作场景、工作需求和工作经验设定，在此并不限定。例如，预设的拥塞丢包率阈值可为0.1％。 若网络丢包参数包括丢包重传报文数目，在网络拥塞判断条件包括在发送的固定数目的传输 报文中丢包重传报文数目大于预设的拥塞丢包重传报文阈值。在出现丢包情况时，数据传输 装置会将丢失的数据包通过报文重新传输。预设的拥塞丢包重传报文阈值可根据工作场景、 工作需求和工作经验设定，在此并不限定。例如，在发送1000条传输报文的条件下，预设的 拥塞丢包重传报文阈值可为3。The network congestion condition includes at least one condition. Specifically, the network congestion judgment condition can be set correspondingly according to the network packet loss parameter. The network congestion judgment condition is used for network congestion judgment based on network packet loss parameters. For example, if the network packet loss parameter includes the number of consecutive packet loss, the network congestion judgment condition includes the number of consecutive packet loss being greater than or equal to two. That is to say, if continuous packet loss occurs, it means that the network packet loss parameter meets the network congestion judgment condition. If the network packet loss parameter includes a packet loss rate, the network congestion judgment condition includes that the packet loss rate is greater than a preset congestion packet loss rate threshold. The preset congestion packet loss rate threshold can be set according to work scenarios, work requirements and work experience, and is not limited here. For example, the preset congestion packet loss rate threshold may be 0.1%. If the network packet loss parameter includes the number of packet loss and retransmission packets, the network congestion judgment condition includes that the number of packet loss and retransmission packets in the fixed number of transmission packets sent is greater than the preset congestion packet loss and retransmission packet threshold. When a packet loss occurs, the data transmission device will retransmit the lost data packet through a message. The preset threshold for packet loss and retransmission due to congestion can be set according to work scenarios, work requirements and work experience, and is not limited here. For example, under the condition of sending 1000 transmission packets, the preset packet loss and retransmission packet threshold can be 3.
在本实施例中，在经历了上述实施例中的往返时延的抖动和带宽的判定后，引入网络丢 包参数，进一步提高判定网络拥塞引起的丢包的准确性。In this embodiment, after experiencing the round-trip delay jitter and bandwidth determination in the above embodiments, the network packet loss parameter is introduced to further improve the accuracy of determining packet loss caused by network congestion.
在步骤S105中，若网络丢包参数满足网络拥塞判断条件中任意一项，则按照第一预设粗 步长缩小拥塞窗口。In step S105, if the network packet loss parameter meets any one of the network congestion judgment conditions, the congestion window is reduced according to the first preset coarse step size.
网络拥塞判断条件可包括一项条件，也可包括多项判断项条件。若网络拥塞判断条件包 括一项条件，则网络丢包参数满足这一项条件，则可按照第一预设粗步长缩小拥塞窗口。若 网络拥塞判断条件包括多项条件，则网络丢包参数满足多项条件中的一项判断条件，则可按 照第一预设粗步长缩小拥塞窗口。The network congestion judging condition may include one condition or multiple judging item conditions. If the network congestion judgment condition includes a condition, then the network packet loss parameter meets this condition, and then the congestion window can be reduced according to the first preset coarse step size. If the network congestion judging condition includes multiple conditions, then the network packet loss parameter meets one of the judging conditions in the multiple conditions, and then the congestion window can be reduced according to the first preset coarse step size.
其中，第一预设粗步长大于第一预设微步长。当传输报文的往返时延的抖动大于拥塞判 断阈值，且当前带宽的下降幅度超出下降阈值，且网络丢包参数满足网络拥塞判断条件中任 意一项时，判定网络发生网络拥塞，出现网络拥塞丢包，需要较大幅度缩小拥塞窗口。比如， 遵循和式增加，积式减小(AIMD，Additive Increase MultiplicativeDecrease)原则，设 置第一预设粗步长为原拥塞窗口的值的一半，将拥塞窗口的值降低一半。Wherein, the first preset coarse step size is greater than the first preset micro step size. When the jitter of the round-trip delay of transmitted packets is greater than the congestion judgment threshold, and the current bandwidth drop exceeds the drop threshold, and the network packet loss parameter meets any of the network congestion judgment conditions, it is determined that the network is congested, and network congestion occurs For packet loss, the congestion window needs to be greatly reduced. For example, following the AIMD (Additive Increase Multiplicative Decrease) principle, set the first preset coarse step size to half the value of the original congestion window, and reduce the value of the congestion window by half.
在本发明实施例中，结合往返时延的抖动、带宽和网络丢包参数三个方面的参数，判断 是否发生网络拥塞丢包，进一步提高了判断网络拥塞丢包的准确性。当传输报文的往返时延 的抖动大于拥塞判断阈值，且当前带宽的下降幅度超出下降阈值，且网络丢包参数满足网络 拥塞判断条件中任意一项时，确定发生网络拥塞丢包，较大幅度缩小拥塞窗口，以缓解网络 拥塞情况。In the embodiment of the present invention, combined with the round-trip delay jitter, bandwidth, and network packet loss parameters, it is judged whether network congestion packet loss occurs, which further improves the accuracy of judging network congestion packet loss. When the jitter of the round-trip delay of transmitted packets is greater than the congestion judgment threshold, and the current bandwidth drop exceeds the drop threshold, and the network packet loss parameters meet any of the network congestion judgment conditions, it is determined that network congestion and packet loss have occurred. Reduce the congestion window by a large amount to alleviate network congestion.
在步骤S106中，若网络丢包参数均未满足网络拥塞判断条件中任意一项，则按照第四预 设微步长缩小拥塞窗口。In step S106, if the network packet loss parameters do not meet any one of the network congestion judgment conditions, then reduce the congestion window according to the fourth preset micro-step size.
网络丢包参数均未满足网络拥塞判断条件中任意一项，即网络丢包参数全部不满足网络 拥塞判断条件。若传输报文的往返时延抖动大于拥塞判断阈值，且当前带宽的下降幅度超出 下降阈值，但网络丢包参数均未满足网络拥塞判断条件中任意一项，表明未发生网络拥塞， 可平缓地适当缩小拥塞窗口。第四预设微步长为细微调节使用的步长。第四预设微步长可根 据工作场景、工作需求和工作经验设定，在此并不限定。在一个示例中，可以设置调节周期， 每经过一个调节周期，按照第四预设微步长缩小一次拥塞窗口。比如，每经过一次传输报文 的往返时延，则将拥塞窗口的值减一。None of the network packet loss parameters meet any of the network congestion judgment conditions, that is, all network packet loss parameters do not meet the network congestion judgment conditions. If the round-trip delay jitter of transmitted packets is greater than the congestion judgment threshold, and the current bandwidth drop exceeds the drop threshold, but none of the network packet loss parameters meets any of the network congestion judgment conditions, it means that no network congestion occurs, and the network can be smoothly Appropriately reduce the congestion window. The fourth preset micro-step size is a step size used for fine-tuning. The fourth preset micro-step size can be set according to the work scene, work requirements and work experience, and is not limited here. In an example, an adjustment period may be set, and after each adjustment period passes, the congestion window is reduced according to a fourth preset micro-step. For example, the value of the congestion window is reduced by one each time the round-trip delay of a transmitted message passes.
在步骤107中，若往返时延的抖动小于等于拥塞判断阈值，则判断往返时延的抖动是否 处于非拥塞抖动范围内。In step 107, if the jitter of the round-trip delay is less than or equal to the congestion judgment threshold, it is judged whether the jitter of the round-trip delay is within the non-congested jitter range.
其中，且拥塞判断阈值大于非拥塞抖动范围的上限值。具体的，可以利用当前的往返时 延与基准往返时延的差值来表示往返时延抖动，也可以利用当前的往返时延与基准往返时延 的差值占基准往返时延的比例，或对当前的往返时延与基准往返时延的差值占基准往返时延 的比例进行一定处理，利用得到的处理结果表示往返时延抖动。比如，diff＝100× (rtt-base_rtt)/base_rtt。其中，diff为传输报文的往返时延的抖动，rtt为传输报文当 前的往返时延，base_rtt为传输报文的基准往返时延。根据往返时延的抖动可得知往返时延 的变化幅度。当前的往返时延可以根据发送传输报文的时间和接收到该传输报文的回复确认 消息(ACK，Acknowledgement)的时间计算得到。Wherein, the congestion judgment threshold is greater than the upper limit of the non-congested jitter range. Specifically, the difference between the current round-trip delay and the reference round-trip delay may be used to represent the round-trip delay jitter, or the ratio of the difference between the current round-trip delay and the reference round-trip delay to the reference round-trip delay may be used, or A certain processing is performed on the ratio of the difference between the current round-trip delay and the reference round-trip delay to the reference round-trip delay, and the obtained processing result is used to represent the round-trip delay jitter. For example, diff=100×(rtt-base_rtt)/base_rtt. Among them, diff is the jitter of the round-trip delay of the transmitted message, rtt is the current round-trip delay of the transmitted message, and base_rtt is the base round-trip delay of the transmitted message. According to the jitter of the round-trip delay, the variation range of the round-trip delay can be obtained. The current round-trip delay can be calculated according to the time of sending the transmission message and the time of receiving the reply acknowledgment message (ACK, Acknowledgment) of the transmission message.
非拥塞抖动范围为传输报文在传输过程中未发生网络拥塞时往返时延的抖动的变化范 围，可通过事先针对未发生网络拥塞的网络环境进行网络勘测，从而总结得到非拥塞抖动范 围。比如，若往返时延抖动计算方法为上述的diff＝100×(rtt-base_rtt)/base_rtt，在 一个示例中，非拥塞抖动范围可以为10至50之间，包括10和50。本发明实施例涉及到的 与非拥塞抖动范围相关的参数包括非拥塞抖动范围的下限值alpha、非拥塞抖动范围的上限 值beta和拥塞判断阈值gamma。具体的，若往返时延的抖动大于拥塞判断阈值，则判断数据 传输可能发生网络拥塞，需要进行进一步的判定。其中，拥塞判断阈值gamma大于非拥塞抖 动范围的上限值beta。在一个示例中，可设置参数blow，blow为非拥塞抖动范围的上限值 beta与非拥塞抖动范围的下限值alpha之间的最小距离，也为非拥塞抖动范围的上限值beta 与拥塞判断阈值gamma之间的最小距离。也就是说，alpha+blow≤beta≤gamma-blow，从而 限定非拥塞抖动范围的上限值beta的大小处于非拥塞抖动范围的下限值alpha与拥塞判断阈 值gamma之间。在本发明实施例中，非拥塞抖动范围可随着网络环境的变化而改变，可通过 调整参数blow来调节beta和gamma。非拥塞抖动范围的上限值beta与非拥塞抖动范围的下 限值alpha之间的距离表示往返时延的抖动的稳定性，如果非拥塞抖动范围的上限值beta与 非拥塞抖动范围的下限值alpha之间的距离增大，则表示往返时延的变化范围增大。The non-congested jitter range refers to the variation range of the round-trip delay jitter when there is no network congestion during the transmission of packets. The non-congested jitter range can be obtained by conducting a network survey in advance for a network environment without network congestion. For example, if the calculation method of the round-trip delay jitter is the above-mentioned diff=100×(rtt-base_rtt)/base_rtt, in an example, the non-congested jitter range may be between 10 and 50, including 10 and 50. The parameters related to the non-congested jitter range involved in the embodiment of the present invention include the lower limit alpha of the non-congested jitter range, the upper limit beta of the non-congested jitter range, and the congestion judgment threshold gamma. Specifically, if the jitter of the round-trip delay is greater than the congestion judgment threshold, it is judged that network congestion may occur during data transmission, and further judgment is required. Wherein, the congestion judgment threshold gamma is greater than the upper limit beta of the non-congested jitter range. In an example, the parameter blow can be set, and blow is the minimum distance between the upper limit value beta of the non-congested jitter range and the lower limit value alpha of the non-congested jitter range, and is also the distance between the upper limit value beta of the non-congested jitter range and the congested The minimum distance between judging threshold gamma. That is to say, alpha+blow≤beta≤gamma-blow, so that the upper limit beta of the non-congested jitter range is defined to be between the lower limit alpha of the non-congested jitter range and the congestion judgment threshold gamma. In the embodiment of the present invention, the non-congested jitter range can be changed as the network environment changes, and beta and gamma can be adjusted by adjusting the parameter blow. The distance between the upper limit value beta of the non-congested jitter range and the lower limit value alpha of the non-congested jitter range indicates the stability of the jitter of the round-trip delay. If the upper limit value beta of the non-congested jitter range and the lower limit value alpha As the distance between the limit values alpha increases, it means that the variation range of the round-trip delay increases.
在步骤S108中，若往返时延的抖动小于非拥塞抖动范围的下限值，则按照第二预设微步 长扩大拥塞窗口。In step S108, if the jitter of the round-trip delay is less than the lower limit of the non-congested jitter range, the congestion window is expanded according to the second preset micro-step.
往返时延的抖动小于非拥塞抖动范围的下限值，表明往返时延的变化幅度较小，未发生 网络拥塞，且往返时延较小，数据传输通畅，可平缓地适当扩大拥塞窗口。第二预设微步长 为细微调节使用的步长。第二预设微步长可根据工作场景、工作需求和工作经验设定，在此 并不限定。在一个示例中，可以设置调节周期，每经过一个调节周期，按照第二预设微步长 扩大一次拥塞窗口。比如，每经过一次传输报文的往返时延的时长，则将拥塞窗口的值加一。The jitter of the round-trip delay is smaller than the lower limit of the non-congested jitter range, which indicates that the change range of the round-trip delay is small, there is no network congestion, and the round-trip delay is small, the data transmission is smooth, and the congestion window can be expanded appropriately. The second preset micro step size is the step size used for fine adjustment. The second preset micro-step size can be set according to the work scene, work requirements and work experience, and is not limited here. In an example, an adjustment cycle can be set, and the congestion window is expanded according to the second preset micro-step every time an adjustment cycle passes. For example, the value of the congestion window is increased by one every time the round-trip delay of a packet is transmitted.
在步骤S109中，若往返时延的抖动大于非拥塞抖动范围的上限值，则按照第三预设微步 长缩小拥塞窗口。In step S109, if the jitter of the round-trip delay is greater than the upper limit of the non-congested jitter range, the congestion window is reduced according to the third preset micro-step size.
往返时延抖动大于非拥塞抖动范围的上限值且小于或等于拥塞判断阈值，表明往返时延 变化较大，但未发生网络拥塞，可平缓地适当缩小拥塞窗口。第三预设微步长为细微调节使 用的步长。第三预设微步长可根据工作场景、工作需求和工作经验设定，在此并不限定。在 一个示例中，可以设置调节周期，每经过一个调节周期，按照第三预设微步长缩小一次拥塞 窗口。比如，每经过一次传输报文的往返时延的时长，则将拥塞窗口的值减一。If the round-trip delay jitter is greater than the upper limit of the non-congested jitter range and less than or equal to the congestion judgment threshold, it indicates that the round-trip delay varies greatly, but there is no network congestion, and the congestion window can be appropriately narrowed down. The third preset micro-step size is the step size used for fine adjustment. The third preset micro-step length can be set according to the work scene, work requirements and work experience, and is not limited here. In an example, an adjustment period can be set, and the congestion window is reduced according to the third preset micro-step every time an adjustment period passes. For example, the value of the congestion window is decremented by one each time the round-trip delay of a packet is transmitted.
在步骤S110中，若往返时延的抖动处于非拥塞抖动范围内，保持拥塞窗口不变。In step S110, if the jitter of the round-trip delay is within the non-congested jitter range, keep the congestion window unchanged.
往返时延的抖动处于非拥塞抖动范围内，表明往返时延基本未发生变化，未发生网络拥 塞，可保持现状，即保持拥塞窗口不变。The jitter of the round-trip delay is within the range of non-congested jitter, indicating that the round-trip delay basically does not change, there is no network congestion, and the status quo can be maintained, that is, the congestion window remains unchanged.
需要说明的是，上述第一预设微步长、第二预设微步长、第三预设微步长和第四预设微 步长可以相同，也可以不同。第一预设粗步长均大于第一预设微步长、第二预设微步长、第 三预设微步长和第四预设微步长。It should be noted that, the first preset micro-step size, the second preset micro-step size, the third preset micro-step size and the fourth preset micro-step size may be the same or different. The first preset coarse step size is larger than the first preset micro step size, the second preset micro step size, the third preset micro step size and the fourth preset micro step size.
需要说明的是，在进行网络拥塞判断的往返时延的抖动判断、带宽判断、网络丢包参数 判断的三个判断中，若三个判断均如上述实施例中满足各自的拥塞判断条件，则可判断发生 网络拥塞。若三个判断中有至少一个判断未满足上述实施例中各自的拥塞判断条件，则可判 断未发生网络拥塞，可能只是发生了随机丢包或错帧等问题。It should be noted that, in the three judgments of the round-trip delay jitter judgment, bandwidth judgment, and network packet loss parameter judgment for judging network congestion, if the three judgments all meet their respective congestion judgment conditions as in the above-mentioned embodiment, then It can be judged that network congestion occurs. If at least one of the three judgments does not meet the respective congestion judgment conditions in the above embodiments, it can be judged that there is no network congestion, and there may be only random packet loss or frame error.
在本发明实施例中，判定未发生网络拥塞，则平缓适当调整拥塞窗口，可最大限度的保 证了网络带宽的利用率，从而提高了网络带宽整体的利用率。In the embodiment of the present invention, if it is determined that no network congestion occurs, then the congestion window is adjusted gently and appropriately, which can ensure the utilization rate of the network bandwidth to the greatest extent, thereby improving the overall utilization rate of the network bandwidth.
在步骤S111(图1中未标出)中，若当前带宽未下降或下降幅度未超出下降阈值，按照 第一预定补偿步长，增加非拥塞抖动范围的上限值。In step S111 (not marked in FIG. 1 ), if the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold, increase the upper limit of the non-congested jitter range according to the first predetermined compensation step.
当传输报文的往返时延的抖动大于拥塞判断阈值，但当前带宽未下降或下降幅度未超出 下降阈值，表明未发生网络拥塞。为了后续的网络拥塞判断更加精准且效率更高，可适当增 加非拥塞抖动范围的上限值，拓宽非拥塞抖动范围，从而实现对非拥塞抖动范围的设定的优 化。When the jitter of the round-trip delay of transmitted packets is greater than the congestion judgment threshold, but the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold, it indicates that no network congestion occurs. In order to make the subsequent judgment of network congestion more accurate and efficient, the upper limit of the non-congested jitter range can be appropriately increased to widen the non-congested jitter range, thereby realizing the optimization of the setting of the non-congested jitter range.
第一预定补偿步长为实现对非拥塞抖动范围的补偿的步长。第一预定补偿步长可以根据 工作场景、工作需求和工作经验设定，在此并不限定。The first predetermined compensation step size is a step size for realizing compensation for the non-congested jitter range. The first predetermined compensation step size can be set according to the work scene, work demand and work experience, and is not limited here.
在步骤S112(图1中未标出)中，若网络丢包参数均未满足网络拥塞判断条件中任意一 项，按照第二预定补偿步长，增加非拥塞抖动范围的上限值。In step S112 (not marked in Figure 1), if the network packet loss parameters do not meet any one of the network congestion judgment conditions, according to the second predetermined compensation step, increase the upper limit of the non-congestion jitter range.
当传输报文的往返时延的抖动大于拥塞判断阈值，且当前带宽下降幅度超出下降阈值， 但网络丢包参数均未满足网络拥塞判断条件中任意一项，表明未发生网络拥塞。为了后续的 网络拥塞判断更加精准且效率更高，可适当增加非拥塞抖动范围的上限值，拓宽非拥塞抖动 范围，实现对非拥塞抖动范围的设定的优化。When the jitter of the round-trip delay of transmitted packets is greater than the congestion judgment threshold, and the current bandwidth drop exceeds the drop threshold, but none of the network packet loss parameters meets any of the network congestion judgment conditions, it indicates that no network congestion occurs. In order to make the subsequent judgment of network congestion more accurate and efficient, the upper limit of the non-congested jitter range can be appropriately increased to widen the non-congested jitter range and optimize the setting of the non-congested jitter range.
第二预定补偿步长为实现对非拥塞抖动范围的补偿的步长。第二预定补偿步长可以根据 工作场景、工作需求和工作经验设定，在此并不限定。The second predetermined compensation step size is a step size for realizing compensation for the non-congested jitter range. The second predetermined compensation step can be set according to the work scene, work demand and work experience, and is not limited here.
在本发明实施例中，可以根据往返时延的抖动判断、带宽判断、网络丢包参数判断的三 个判断所反馈的内容，一方面，调节拥塞窗口，以实现在避免或缓解网络拥塞的情况下最大 限度保证网络带宽的利用率的目的；另一方面，也可根据三个判断的反馈，对非拥塞抖动范 围进行优化更新，从而提高网络拥塞判断的精准度和效率。In the embodiment of the present invention, according to the content fed back by the three judgments of round-trip delay jitter judgment, bandwidth judgment, and network packet loss parameter judgment, on the one hand, the congestion window is adjusted to avoid or alleviate network congestion. On the other hand, according to the feedback of the three judgments, the non-congested jitter range can be optimized and updated, so as to improve the accuracy and efficiency of network congestion judgment.
在一个示例中，可将数据传输方法与状态机相结合，状态机为协调相关信号动作、完成 特定操作的控制中心，在本发明实施例中，数据传输装置可利用状态机实现，从而利用状态 机的状态的转换，以及实现网络拥塞判断后，对于不同判断结果所指定的不同处理策略。比 如：在进行网络拥塞的三个判断之前，具体的，可将初始状态设为状态机的初始状态(open 状态)；若网络拥塞的三个判断中有至少一个判断未满足上述实施例中各自的拥塞判断条件， 可将状态机的状态转换为乱序状态(disorder状态)；若网络拥塞的三个判断均满足上述实 施例中各自的拥塞判断条件，可将状态机的状态转换为拥塞状态(loss状态)。In one example, the data transmission method can be combined with a state machine, and the state machine is a control center that coordinates related signal actions and completes specific operations. In the embodiment of the present invention, the data transmission device can be implemented using a state The transition of the state of the machine, and the different processing strategies specified for different judgment results after the network congestion judgment is realized. For example: before performing the three judgments of network congestion, specifically, the initial state can be set to the initial state (open state) of the state machine; if at least one of the three judgments of network congestion does not satisfy the respective The congestion judgment condition of the state machine can be converted into a disordered state (disorder state); if the three judgments of network congestion all meet the respective congestion judgment conditions in the above-mentioned embodiments, the state of the state machine can be converted into a congestion state (loss state).
图3为本发明另一实施例中一种数据传输方法的流程图。该数据传输方法可应用于数据 传输过程中的发送端，如图3所示，数据传输方法包括步骤S201至步骤S212。Fig. 3 is a flowchart of a data transmission method in another embodiment of the present invention. The data transmission method can be applied to the sending end in the data transmission process. As shown in FIG. 3, the data transmission method includes steps S201 to S212.
在步骤S201中，获取发送的传输报文的往返时延的抖动，判断往返时延的抖动是否大于 拥塞判断阈值。In step S201, obtain the jitter of the round-trip delay of the transmitted transmission message, and judge whether the jitter of the round-trip delay is greater than the congestion judgment threshold.
其中，关于往返时延的抖动和拥塞判断阈值的说明请参见上述实施例中步骤S101中的相 关说明。Wherein, for the description of the jitter of the round-trip delay and the congestion judgment threshold, please refer to the relevant description in step S101 in the above embodiment.
在步骤S202中，若往返时延的抖动大于拥塞判断阈值，则获取网络丢包参数，判断网络 丢包参数是否满足网络拥塞判断条件。In step S202, if the jitter of the round-trip delay is greater than the congestion judgment threshold, the network packet loss parameter is obtained, and it is judged whether the network packet loss parameter meets the network congestion judgment condition.
其中，拥塞判断阈值的说明请参见上述实施例中步骤S101中的相关说明。网络丢包参数 和网络拥塞判断条件的说明请参见上述实施例中步骤S104至步骤S106中的相关说明。For the description of the congestion judgment threshold, please refer to the relevant description in step S101 in the above embodiment. For descriptions of network packet loss parameters and network congestion judgment conditions, please refer to the relevant descriptions in step S104 to step S106 in the above embodiment.
在步骤S203中，若网络丢包参数均未满足网络拥塞判断条件中任意一项，则按照第一预 设微步长缩小拥塞窗口。In step S203, if the network packet loss parameters do not meet any one of the network congestion judgment conditions, the congestion window is reduced according to the first preset micro-step size.
其中，网络丢包参数满足以及未满足网络拥塞判断条件中任意一项的说明，请参见上述 实施例中的步骤S105、步骤S106和步骤S112的相关说明。Wherein, for the description of whether the network packet loss parameter meets or fails to meet any one of the network congestion judgment conditions, please refer to the relevant descriptions of step S105, step S106 and step S112 in the above-mentioned embodiment.
第一预设微步长为细微调节使用的步长，可根据工作场景、工作需求或工作经验设置， 在此并不限定。在一个示例中，可以设置调节周期，每经过一个调节周期，按照第一预设微 步长缩小一次拥塞窗口。比如，每经过一次传输报文的往返时延的时长，则将拥塞窗口的值 减一。The first preset micro-step size is a step size used for fine-tuning, which can be set according to work scenarios, work requirements or work experience, and is not limited here. In an example, an adjustment cycle can be set, and the congestion window is reduced according to the first preset micro-step every time an adjustment cycle passes. For example, the value of the congestion window is reduced by one every time the round-trip delay of a packet is transmitted.
在本发明实施例中，利用传输报文的往返时延的抖动和网络丢包参数，共同进行网络是 否发生拥塞的判断。在判断往返时延的抖动大于拥塞判断阈值，但网络丢包参数均未满足网 络拥塞判断条件中任意一项时，判断数据传输出现的是随机丢包，并未发生网络拥塞，对拥 塞窗口进行微调。并未对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而 在缓解随机丢包所带来的影响的基础上，最大限度的保证了网络带宽的利用率，从而提高了 网络带宽整体的利用率。In the embodiment of the present invention, the jitter of the round-trip delay of the transmitted message and the network packet loss parameter are used to jointly judge whether the network is congested. When it is judged that the jitter of the round-trip delay is greater than the congestion judgment threshold, but the network packet loss parameters do not meet any of the network congestion judgment conditions, it is judged that random packet loss occurs in data transmission, and no network congestion occurs, and the congestion window is checked. fine-tuning. The congestion window has not been greatly adjusted, which avoids misjudgment of network congestion, so that on the basis of mitigating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network performance. Overall bandwidth utilization.
尤其是在带宽和时延均较大的网络中，若单纯的发现丢包，则大幅度减小拥塞窗口，则 会对网络带宽的利用率造成更大影响。本发明实施例能够区别随机丢包和网络拥塞丢包，在 带宽和时延均较大的网络中能够及时有效的优化网络的拥塞判断机制。Especially in a network with large bandwidth and delay, if packet loss is simply found, the congestion window will be greatly reduced, which will have a greater impact on the utilization of network bandwidth. The embodiment of the present invention can distinguish between random packet loss and network congestion packet loss, and can timely and effectively optimize the congestion judgment mechanism of the network in a network with large bandwidth and delay.
在步骤S204中，若网络丢包参数满足网络拥塞判断条件中任意一项，则获取当前带宽， 判断当前带宽的下降幅度是否超出下降阈值。In step S204, if the network packet loss parameter satisfies any one of the network congestion judging conditions, the current bandwidth is obtained, and it is judged whether the decrease of the current bandwidth exceeds the decrease threshold.
其中，网络丢包参数满足网络拥塞判断条件中任意一项的说明可参见上述实施例中的步 骤S105的相关说明。带宽和当前带宽的下降幅度是否超过下降阈值的说明可参见上述实施例 中的步骤102和步骤103的相关说明。Wherein, the description that the network packet loss parameter satisfies any one of the network congestion judgment conditions can refer to the relevant description of step S105 in the above-mentioned embodiment. For the description of whether the decrease range of the bandwidth and the current bandwidth exceeds the decrease threshold, refer to the relevant descriptions of step 102 and step 103 in the above-mentioned embodiment.
在步骤S205中，若当前带宽的下降幅度超出下降阈值，则按照第一预设粗步长缩小拥塞 窗口。In step S205, if the decrease of the current bandwidth exceeds the decrease threshold, the congestion window is reduced according to the first preset coarse step.
其中，第一预设粗步长大于第一预设微步长。第一预设粗步长可根据工作场景、工作需 求或工作经验设置。第一预设粗步长与上一实施例中的第一预设粗步长可以相同，也可以不 同。当往返时延的抖动大于拥塞判断阈值，且网络丢包参数满足网络拥塞判断条件中任意一 项，且当前带宽的下降幅度超出下降阈值时，判定网络发生网络拥塞，出现网络拥塞丢包， 需要较大幅度缩小拥塞窗口。比如，遵循和式增加，积式减小原则，设置第一预设粗步长为 原拥塞窗口的值的一半，将拥塞窗口的值降低一半。Wherein, the first preset coarse step size is greater than the first preset micro step size. The first preset coarse step length can be set according to the work scene, work requirements or work experience. The first preset coarse step size may be the same as or different from the first preset coarse step size in the previous embodiment. When the jitter of the round-trip delay is greater than the congestion judgment threshold, and the network packet loss parameter meets any of the network congestion judgment conditions, and the current bandwidth decline exceeds the decline threshold, it is determined that the network is congested and packet loss due to network congestion occurs. Reduce the congestion window considerably. For example, following the principle of increasing the sum and decreasing the product, set the first preset coarse step to half the value of the original congestion window, and reduce the value of the congestion window by half.
在本发明实施例中，结合往返时延的抖动、网络丢包参数和带宽三个方向的参数，判断 是否发生网络拥塞丢包，进一步提高了判断网络拥塞丢包的准确性。当传输报文的往返时延 的抖动大于拥塞判断阈值，且网络丢包参数满足网络拥塞判断条件中任意一项，且当前带宽 的下降幅度超出下降阈值时，确定发生网络拥塞丢包，较大幅度缩小拥塞窗口，以缓解网络 拥塞情况。In the embodiment of the present invention, combined with the round-trip delay jitter, network packet loss parameters and parameters in three directions of bandwidth, it is judged whether network congestion packet loss occurs, which further improves the accuracy of judging network congestion packet loss. When the jitter of the round-trip delay of the transmitted message is greater than the congestion judgment threshold, and the network packet loss parameter meets any one of the network congestion judgment conditions, and the decrease of the current bandwidth exceeds the decrease threshold, it is determined that the packet loss due to network congestion occurs. Reduce the congestion window by a large amount to alleviate network congestion.
在步骤S206中，若往返时延的抖动小于等于拥塞判断阈值，则判断往返时延的抖动是否 处于非拥塞抖动范围内。In step S206, if the jitter of the round-trip delay is less than or equal to the congestion judgment threshold, it is judged whether the jitter of the round-trip delay is within the non-congested jitter range.
其中，非拥塞抖动范围的说明可参见上述实施例中的步骤S107的相关说明。For the description of the non-congested jitter range, refer to the related description of step S107 in the above embodiment.
在步骤S207中，若往返时延的抖动小于非拥塞抖动范围的下限，则按照第二预设微步 长扩大拥塞窗口。In step S207, if the jitter of the round-trip delay is less than the lower limit of the non-congested jitter range, the congestion window is expanded according to the second preset micro-step size.
往返时延的抖动小于非拥塞抖动范围的下限值，表明往返时延变化幅度较小，未发生网 络拥塞，数据传输通畅，可平缓地适当扩大拥塞窗口。第二预设微步长为细微调节使用的步 长，可根据工作场景、工作需求和工作经验设定，在此并不限定。在一个示例中，可以设置 调节周期，每经过一个调节周期，按照第二预设微步长扩大一次拥塞窗口。比如，每经过一 次传输报文的往返时延的时长，则将拥塞窗口的值加一。The jitter of the round-trip delay is less than the lower limit of the non-congested jitter range, indicating that the round-trip delay has a small variation, no network congestion occurs, data transmission is smooth, and the congestion window can be appropriately expanded gently. The second preset micro-step size is the step size used for fine-tuning, which can be set according to the work scene, work requirements and work experience, and is not limited here. In one example, an adjustment period can be set, and the congestion window is enlarged according to the second preset micro-step length every time an adjustment period passes. For example, the value of the congestion window is increased by one every time the round-trip delay of a packet is transmitted.
在步骤S208中，若往返时延的抖动大于非拥塞抖动范围的上限，则按照第三预设微步长 缩小拥塞窗口。In step S208, if the jitter of the round-trip delay is greater than the upper limit of the non-congested jitter range, the congestion window is reduced according to the third preset micro-step size.
往返时延的抖动大于非拥塞抖动范围的上限值且小于或等于拥塞判断阈值，表明往返时 延变化幅度较大，但未发生网络拥塞，可平缓地适当缩小拥塞窗口。第三预设微步长为细微 调节使用的步长，可根据工作场景、工作需求和工作经验设定，在此并不限定。在一个示例 中，可以设置调节周期，每经过一个调节周期，按照第三预设微步长缩小一次拥塞窗口。比 如，每经过一次传输报文的往返时延的时长，则将拥塞窗口的值减一。If the round-trip delay jitter is greater than the upper limit of the non-congested jitter range and less than or equal to the congestion judgment threshold, it indicates that the round-trip delay varies greatly, but there is no network congestion, and the congestion window can be appropriately narrowed down. The third preset micro-step size is the step size used for fine adjustment, which can be set according to the work scene, work requirements and work experience, and is not limited here. In an example, an adjustment cycle can be set, and the congestion window is reduced according to the third preset micro-step every time an adjustment cycle passes. For example, the value of the congestion window is decremented by one every time the round-trip delay of a packet is transmitted.
在步骤S209中，若往返时延的抖动处于非拥塞抖动范围内，保持拥塞窗口不变。In step S209, if the jitter of the round-trip delay is within the non-congested jitter range, keep the congestion window unchanged.
往返时延的抖动处于非拥塞抖动范围内，表明往返时延基本未发生变化，未发生网络拥 塞，可保持现状，即保持拥塞窗口不变。The jitter of the round-trip delay is within the range of non-congested jitter, indicating that the round-trip delay basically does not change, there is no network congestion, and the status quo can be maintained, that is, the congestion window remains unchanged.
在步骤S210中，若当前带宽未下降或下降幅度未超出下降阈值，则按照第四预设微步长 缩小拥塞窗口值。In step S210, if the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold, the congestion window value is reduced according to the fourth preset micro-step size.
若传输报文的往返时延的抖动大于拥塞判断阈值，且当前带宽的下降幅度超出下降阈值， 但网络丢包参数均未满足网络拥塞判断条件中任意一项，表明未发生网络拥塞，可平缓地适 当缩小拥塞窗口。第四预设微步长为细微调节使用的步长，可根据工作场景、工作需求和工 作经验设定，在此并不限定。在一个示例中，可以设置调节周期，每经过一个调节周期，按 照第四预设微步长缩小一次拥塞窗口。比如，每经过一次传输报文的往返时延的时长，则将 拥塞窗口的值减一。If the jitter of the round-trip delay of transmitted packets is greater than the congestion judgment threshold, and the current bandwidth drop exceeds the drop threshold, but none of the network packet loss parameters meets any of the network congestion judgment conditions, it means that no network congestion occurs and can be eased Properly reduce the congestion window. The fourth preset micro-step size is the step size used for fine-tuning, which can be set according to the work scene, work requirements and work experience, and is not limited here. In an example, an adjustment cycle can be set, and the congestion window is reduced according to a fourth preset micro-step every time an adjustment cycle passes. For example, the value of the congestion window is reduced by one every time the round-trip delay of a packet is transmitted.
需要说明的是，上述第一预设微步长、第二预设微步长、第三预设微步长和第四预设微 步长可以相同，也可以不同。在一个示例中，第一预设粗步长均大于第一预设微步长、第二 预设微步长、第三预设微步长和第四预设微步长。It should be noted that, the first preset micro-step size, the second preset micro-step size, the third preset micro-step size and the fourth preset micro-step size may be the same or different. In one example, the first preset coarse step size is larger than the first preset micro step size, the second preset micro step size, the third preset micro step size and the fourth preset micro step size.
在本发明实施例中，判定未发生网络拥塞，则平缓适当调整拥塞窗口，可最大限度的保 证了网络带宽的利用率，从而提高了网络带宽整体的利用率。In the embodiment of the present invention, if it is determined that no network congestion occurs, then the congestion window is adjusted gently and appropriately, which can ensure the utilization rate of the network bandwidth to the greatest extent, thereby improving the overall utilization rate of the network bandwidth.
在步骤S211(图2中未示出)中，若网络丢包参数均未满足网络拥塞判断条件中任意一 项，按照第一预定补偿步长，增加非拥塞抖动范围的上限值。In step S211 (not shown in Figure 2), if the network packet loss parameters do not meet any one of the network congestion judgment conditions, according to the first predetermined compensation step, increase the upper limit of the non-congestion jitter range.
当传输报文的往返时延的抖动大于拥塞判断阈值，但网络丢包参数均未满足网络拥塞判 断条件中任意一项，表明未发生网络拥塞。为了后续的网络拥塞判断更加精准且效率更高， 可适当增加非拥塞抖动范围的上限值，拓宽非拥塞抖动范围，从而实现对非拥塞抖动范围的 设定的优化。When the round-trip delay jitter of transmitted packets is greater than the congestion judgment threshold, but none of the network packet loss parameters meets any of the network congestion judgment conditions, it indicates that no network congestion occurs. In order to make the subsequent judgment of network congestion more accurate and more efficient, the upper limit of the non-congested jitter range can be appropriately increased to widen the non-congested jitter range, so as to optimize the setting of the non-congested jitter range.
第一预定补偿步长为实现对非拥塞抖动范围的补偿的步长，可以根据工作场景、工作需 求和工作经验设定，在此并不限定。The first predetermined compensation step size is the step size for realizing the compensation of the non-congested jitter range, which can be set according to the working scene, work requirements and work experience, and is not limited here.
在步骤S212(图2中未示出)中，若当前带宽未下降或下降幅度未超出下降阈值，按照 第二预定补偿步长，增加非拥塞抖动范围的上限值。In step S212 (not shown in FIG. 2 ), if the current bandwidth does not drop or the drop range does not exceed the drop threshold, increase the upper limit of the non-congested jitter range according to the second predetermined compensation step.
当传输报文的往返时延的抖动大于拥塞判断阈值，且当前带宽下降幅度超出下降阈值， 但网络丢包参数均未满足网络拥塞判断条件中任意一项，表明未发生网络拥塞。为了后续的 网络拥塞判断更加精准且效率更高，可适当增加非拥塞抖动范围的上限值，拓宽非拥塞抖动 范围，实现对非拥塞抖动范围的设定的优化。When the jitter of the round-trip delay of transmitted packets is greater than the congestion judgment threshold, and the current bandwidth drop exceeds the drop threshold, but none of the network packet loss parameters meets any of the network congestion judgment conditions, it indicates that no network congestion occurs. In order to make the subsequent judgment of network congestion more accurate and efficient, the upper limit of the non-congested jitter range can be appropriately increased to widen the non-congested jitter range and optimize the setting of the non-congested jitter range.
第二预定补偿步长为实现对非拥塞抖动范围的补偿的步长，可以根据工作场景、工作需 求和工作经验设定，在此并不限定。The second predetermined compensation step size is the step size for realizing the compensation of the non-congested jitter range, which can be set according to the work scene, work requirements and work experience, and is not limited here.
在本发明实施例中，可以根据往返时延的抖动判断、网络丢包参数、带宽判断判断的三 个判断的反馈，一方面，调节拥塞窗口，以实现在避免或缓解网络拥塞的情况下最大限度保 证网络带宽的利用率的目的；另一方面，也可根据三个判断的反馈，对非拥塞抖动范围进行 优化更新，从而提高网络拥塞判断的精准度和效率。In the embodiment of the present invention, according to the feedback of the three judgments of the round-trip delay jitter judgment, network packet loss parameter, and bandwidth judgment, on the one hand, the congestion window is adjusted to achieve the maximum while avoiding or alleviating network congestion. The purpose of ensuring the maximum utilization of network bandwidth; on the other hand, it can also optimize and update the non-congested jitter range according to the feedback of the three judgments, so as to improve the accuracy and efficiency of network congestion judgment.
在一个示例中，可将数据传输方法与状态机相结合。具体结合方式可参照上述实施例中 的说明内容。In one example, the data transfer method can be combined with a state machine. For specific combination methods, please refer to the descriptions in the above-mentioned embodiments.
图4为本发明又一实施例中一种数据传输方法的流程图。该数据传输方法可应用于数据 传输过程中的发送端，如图4所示，数据传输方法可包括步骤S301至步骤S310。Fig. 4 is a flowchart of a data transmission method in another embodiment of the present invention. The data transmission method can be applied to the sending end in the data transmission process. As shown in FIG. 4, the data transmission method can include steps S301 to S310.
在步骤S301中，获取当前带宽，判断当前带宽的下降幅度是否超出下降阈值。In step S301, the current bandwidth is acquired, and it is judged whether the decrease range of the current bandwidth exceeds the decrease threshold.
其中，当前带宽、下降幅度和下降阈值的说明可参照上述实施例中的步骤S102和步骤 S103的相关说明。Wherein, for the description of the current bandwidth, the drop range and the drop threshold, reference may be made to the related descriptions of step S102 and step S103 in the above embodiment.
在步骤S302中，若当前带宽的下降幅度超出下降阈值，则获取网络丢包参数，并判断网 络丢包参数是否满足网络拥塞判断条件。In step S302, if the drop rate of the current bandwidth exceeds the drop threshold, the network packet loss parameter is obtained, and it is judged whether the network packet loss parameter satisfies the network congestion judgment condition.
其中，网络丢包参数和网络拥塞判断条件的说明可参见上述实施例中的步骤S104、步骤 S105和步骤S106中的相关说明。For the description of network packet loss parameters and network congestion judgment conditions, please refer to the related descriptions in step S104, step S105 and step S106 in the above embodiment.
在一个示例中，网络丢包参数包括连续丢包数目，网络拥塞判断条件包括连续丢包数目 大于等于两个；和/或，网络丢包参数包括丢包率，网络拥塞判断条件包括丢包率大于预设的 拥塞丢包率阈值；和/或，网络丢包参数包括丢包重传报文数目，网络拥塞判断条件包括丢包 重传报文数目大于预设的拥塞丢包重传报文阈值。In one example, the network packet loss parameter includes the number of consecutive packet loss, and the network congestion judgment condition includes that the number of consecutive packet loss is greater than or equal to two; and/or, the network packet loss parameter includes a packet loss rate, and the network congestion judgment condition includes a packet loss rate greater than the preset congestion packet loss rate threshold; and/or, the network packet loss parameters include the number of packet loss and retransmission packets, and the network congestion judgment condition includes the number of packet loss and retransmission packets being greater than the preset congestion packet loss and retransmission packets threshold.
在步骤S303中，若网络丢包参数均未满足网络拥塞判断条件中任意一项，则按照第一预 设微步长缩小拥塞窗口。In step S303, if the network packet loss parameters do not meet any one of the network congestion judgment conditions, the congestion window is reduced according to the first preset micro-step size.
其中，网络丢包参数均未满足网络拥塞判断条件中任意一项的说明可参见上述实施例中 的步骤S106的相关说明。Wherein, the description that none of the network packet loss parameters meet any of the network congestion judgment conditions can refer to the relevant description of step S106 in the above-mentioned embodiment.
第一预设微步长为细微调节使用的步长，可根据工作场景、工作需求和工作经验设定， 在此并不限定。在一个示例中，可以设置调节周期，每经过一个调节周期，按照第一预设微 步长缩小一次拥塞窗口。比如，每经过一次传输报文的往返时延的时长，则将拥塞窗口的值 减一。The first preset micro-step size is a step size used for fine-tuning, which can be set according to work scenarios, work requirements and work experience, and is not limited here. In an example, an adjustment cycle can be set, and the congestion window is reduced according to the first preset micro-step every time an adjustment cycle passes. For example, the value of the congestion window is reduced by one every time the round-trip delay of a packet is transmitted.
在本发明实施例中，在判断当前带宽的下降幅度超出下降阈值，但网络丢包参数均未满 足网络拥塞判断条件中任意一项时，判断数据传输出现的是随机丢包，并未发生网络拥塞， 对拥塞窗口进行微调。并未对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况， 从而在缓解随机丢包所带来的影响的基础上，最大限度的保证了网络带宽的利用率，从而提 高了网络带宽整体的利用率。In the embodiment of the present invention, when it is judged that the drop rate of the current bandwidth exceeds the drop threshold, but none of the network packet loss parameters satisfies any of the network congestion judgment conditions, it is judged that random packet loss occurs in data transmission, and no network congestion occurs. Congestion, fine-tuning the congestion window. The congestion window has not been greatly adjusted, which avoids misjudgment of network congestion, thereby mitigating the impact of random packet loss, maximizing the utilization of network bandwidth, and improving network performance. Overall bandwidth utilization.
在步骤S304中，若网络丢包参数满足网络拥塞判断条件中任意一项，则获取传输报文的 往返时延的抖动，判断往返时延的抖动是否大于拥塞判断阈值。In step S304, if the network packet loss parameter meets any one of the network congestion judgment conditions, then obtain the jitter of the round-trip delay of the transmitted message, and judge whether the jitter of the round-trip delay is greater than the congestion judgment threshold.
其中，往返时延的抖动和非拥塞抖动范围的说明可参见上述实施例中的步骤S101和步骤 S107的相关说明。For the description of the round-trip delay jitter and non-congested jitter range, refer to the related descriptions of step S101 and step S107 in the above-mentioned embodiment.
在步骤S305中，若往返时延的抖动大于拥塞判断阈值，则按照第一预设粗步长缩小拥塞 窗口。In step S305, if the jitter of the round-trip delay is greater than the congestion judgment threshold, the congestion window is reduced according to the first preset coarse step.
其中，拥塞判断阈值的说明可参见上述实施例中的步骤S101的相关说明。For the description of the congestion judgment threshold, refer to the related description of step S101 in the above embodiment.
第一预设粗步长大于第一预设微步长。第一预设粗步长可根据工作场景、工作需求或工 作经验设置。第一预设粗步长与上述两个实施例中的第一预设粗步长可以相同，也可以不同。 当当前带宽的下降幅度超出下降阈值，且网络丢包参数满足网络拥塞判断条件中任意一项， 且往返时延的抖动大于拥塞判断阈值时，判定网络发生网络拥塞，出现网络拥塞丢包，需要 较大幅度缩小拥塞窗口。比如，遵循和式增加，积式减小原则，设置第一预设粗步长为原拥 塞窗口的值的一半，将拥塞窗口的值降低一半。The first preset coarse step size is greater than the first preset micro step size. The first preset coarse step length can be set according to work scenarios, work requirements or work experience. The first preset coarse step size may be the same as or different from the first preset coarse step size in the above two embodiments. When the decline of the current bandwidth exceeds the drop threshold, and the network packet loss parameters meet any of the network congestion judgment conditions, and the round-trip delay jitter is greater than the congestion judgment threshold, it is determined that the network is congested and packet loss due to network congestion occurs. Reduce the congestion window considerably. For example, following the principle of increasing the sum and decreasing the product, set the first preset coarse step to half the value of the original congestion window, and reduce the value of the congestion window by half.
在本发明实施例中，结合带宽、网络丢包参数和往返时延的抖动三个方向的参数，判断 是否发生网络拥塞丢包，进一步提高了判断网络拥塞丢包的准确性。当当前带宽的下降幅度 超出下降阈值，且网络丢包参数满足网络拥塞判断条件中任意一项，且传输报文的往返时延 的抖动大于拥塞判断阈值时，确定发生网络拥塞丢包，较大幅度缩小拥塞窗口，以缓解网络 拥塞情况。In the embodiment of the present invention, it is judged whether packet loss due to network congestion occurs in combination with parameters in three directions of jitter of bandwidth, network packet loss parameter and round-trip delay, which further improves the accuracy of judging network congestion and packet loss. When the decline of the current bandwidth exceeds the drop threshold, and the network packet loss parameters meet any of the network congestion judgment conditions, and the jitter of the round-trip delay of the transmitted message is greater than the congestion judgment threshold, it is determined that the packet loss due to network congestion occurs. Reduce the congestion window by a large amount to alleviate network congestion.
在步骤S306中，若当前带宽未下降或下降幅度未超出下降阈值，则按照第二预设微步长 缩小拥塞窗口。In step S306, if the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold, the congestion window is reduced according to the second preset micro-step.
若当前带宽的下降幅度未超出下降阈值，表明未发生网络拥塞，可平缓地适当缩小拥塞 窗口。第二预设微步长为细微调节使用的步长，可根据工作场景、工作需求和工作经验设定， 在此并不限定。在一个示例中，可以设置调节周期，每经过一个调节周期，按照第二预设微 步长缩小一次拥塞窗口。比如，每经过一次传输报文的往返时延的时长，则将拥塞窗口的值 减一。If the decrease of the current bandwidth does not exceed the decrease threshold, it indicates that there is no network congestion, and the congestion window can be appropriately reduced gently. The second preset micro-step size is a step size used for fine-tuning, which can be set according to work scenarios, work requirements and work experience, and is not limited here. In an example, an adjustment cycle can be set, and the congestion window is reduced according to the second preset micro-step every time an adjustment cycle passes. For example, the value of the congestion window is reduced by one every time the round-trip delay of a packet is transmitted.
在步骤S307中，若往返时延的抖动小于等于拥塞判断阈值，则判断往返时延的抖动是否 处于非拥塞抖动范围内。In step S307, if the jitter of the round-trip delay is less than or equal to the congestion judgment threshold, it is judged whether the jitter of the round-trip delay is within the non-congested jitter range.
在步骤S308中，若往返时延的抖动大于非拥塞抖动范围的上限值，则按照第三预设微步 长缩小拥塞窗口。In step S308, if the jitter of the round-trip delay is greater than the upper limit of the non-congested jitter range, the congestion window is reduced according to the third preset micro-step size.
往返时延的抖动大于非拥塞抖动范围的上限值且小于或等于拥塞判断阈值，表明往返时 延变化幅度较大，但未发生网络拥塞，可平缓地适当缩小拥塞窗口。第三预设微步长为细微 调节使用的步长，可根据工作场景、工作需求和工作经验设定，在此并不限定。在一个示例 中，可以设置调节周期，每经过一个调节周期，按照第三预设微步长缩小一次拥塞窗口。比 如，每经过一次传输报文的往返时延的时长，则将拥塞窗口的值减一。If the round-trip delay jitter is greater than the upper limit of the non-congested jitter range and less than or equal to the congestion judgment threshold, it indicates that the round-trip delay varies greatly, but there is no network congestion, and the congestion window can be appropriately narrowed down. The third preset micro-step size is the step size used for fine adjustment, which can be set according to the work scene, work requirements and work experience, and is not limited here. In an example, an adjustment cycle can be set, and the congestion window is reduced according to the third preset micro-step every time an adjustment cycle passes. For example, the value of the congestion window is decremented by one every time the round-trip delay of a packet is transmitted.
在步骤S309中，若往返时延的抖动小于非拥塞抖动范围的下限值，则按照第四预设微步 长缩小拥塞窗口。In step S309, if the jitter of the round-trip delay is less than the lower limit of the non-congested jitter range, the congestion window is reduced according to the fourth preset micro-step size.
往返时延的抖动小于非拥塞抖动范围的下限，表明未发生网络拥塞，可平缓地适当缩小 拥塞窗口。第四预设微步长为细微调节使用的步长，可根据工作场景、工作需求和工作经验 设定，在此并不限定。在一个示例中，可以设置调节周期，每经过一个调节周期，按照第四 预设微步长缩小一次拥塞窗口。比如，每经过一次传输报文的往返时延的时长，则将拥塞窗 口的值减一。The jitter of the round-trip delay is less than the lower limit of the non-congested jitter range, indicating that no network congestion occurs, and the congestion window can be appropriately reduced smoothly. The fourth preset micro-step size is the step size used for fine-tuning, which can be set according to the work scene, work requirements and work experience, and is not limited here. In an example, an adjustment cycle can be set, and the congestion window is reduced according to the fourth preset micro-step length every time an adjustment cycle passes. For example, the value of the congestion window is decremented by one each time the round-trip delay of a packet is transmitted.
在步骤S310中，若往返时延的抖动处于非拥塞抖动范围内，则按照第五预设微步长缩小 拥塞窗口。In step S310, if the jitter of the round-trip delay is within the non-congested jitter range, the congestion window is reduced according to the fifth preset micro-step size.
往返时延的抖动处于非拥塞抖动范围内，表明未发生网络拥塞，可平缓地适当缩小拥塞 窗口。第五预设微步长为细微调节使用的步长，可根据工作场景、工作需求和工作经验设定， 在此并不限定。在一个示例中，可以设置调节周期，每经过一个调节周期，按照第五预设微 步长缩小一次拥塞窗口。比如，每经过一次传输报文的往返时延的时长，则将拥塞窗口的值 减一。The jitter of the round-trip delay is within the non-congested jitter range, indicating that there is no network congestion, and the congestion window can be appropriately reduced smoothly. The fifth preset micro-step size is a step size used for fine-tuning, which can be set according to work scenarios, work requirements and work experience, and is not limited here. In an example, an adjustment cycle can be set, and the congestion window is reduced according to the fifth preset micro-step every time an adjustment cycle passes. For example, the value of the congestion window is reduced by one every time the round-trip delay of a packet is transmitted.
需要说明的是，上述第一预设微步长、第二预设微步长、第三预设微步长、第四预设微 步长和第五预设微步长可以相同，也可以不同。在一个示例中，第一预设粗步长均大于第一 预设微步长、第二预设微步长、第三预设微步长、第四预设微步长和第五预设微步长。It should be noted that the above-mentioned first preset micro-step size, second preset micro-step size, third preset micro-step size, fourth preset micro-step size and fifth preset micro-step size can be the same, or can be different. In one example, the first preset coarse step size is larger than the first preset micro step size, the second preset micro step size, the third preset micro step size, the fourth preset micro step size and the fifth preset micro step size microstep.
在本发明实施例中，判定未发生网络拥塞，则平缓适当调整拥塞窗口，可最大限度的保 证了网络带宽的利用率，从而提高了网络带宽整体的利用率。In the embodiment of the present invention, if it is determined that no network congestion occurs, then the congestion window is adjusted gently and appropriately, which can ensure the utilization rate of the network bandwidth to the greatest extent, thereby improving the overall utilization rate of the network bandwidth.
在一个示例中，也可先进行网络丢包参数的判断；在网络丢包参数满足网络拥塞判断条 件中任意一项的条件下，再进行带宽的判断；在当前带宽的下降幅度超出下降阈值的条件下， 再进行往返时延的抖动的判断。In an example, the judgment of the network packet loss parameter may also be performed first; under the condition that the network packet loss parameter satisfies any one of the network congestion judgment conditions, then the bandwidth judgment is performed; Under the condition, the jitter of the round-trip delay is judged.
在另一个示例中，也可按照网络丢包参数的判断、往返时延的抖动的判断和带宽的判断 的顺序实现网络拥塞的判断。在又一个示例中，也可按照带宽的判断、往返时延的抖动的判 断和网络丢包参数的判断的顺序实现网络拥塞的判断。在再一个示例中，也可按照网络丢包 参数的判断、带宽的判断和往返时延的抖动的判断的顺序实现网络拥塞的判断。In another example, the judgment of network congestion can also be realized in the order of the judgment of the network packet loss parameter, the judgment of the jitter of the round-trip delay, and the judgment of the bandwidth. In yet another example, the judgment of network congestion may also be realized in the order of bandwidth judgment, round-trip delay jitter judgment, and network packet loss parameter judgment. In yet another example, network congestion can also be determined in the order of network packet loss parameter determination, bandwidth determination, and round-trip delay jitter determination.
图5为本发明一实施例中一种数据传输装置400的结构示意图。如图5所示，数据传输 装置400可包括第一判断单元401、第二判断单元402、第一处理单元403、第三判断单元404、 第二处理单元405、第四判断单元406、第三处理单元407、第四处理单元408、第五处理单 元409、第六处理单元410、第一调整单元411和第二调整单元412。FIG. 5 is a schematic structural diagram of a data transmission device 400 in an embodiment of the present invention. As shown in Figure 5, the data transmission device 400 may include a first judging unit 401, a second judging unit 402, a first processing unit 403, a third judging unit 404, a second processing unit 405, a fourth judging unit 406, a third A processing unit 407 , a fourth processing unit 408 , a fifth processing unit 409 , a sixth processing unit 410 , a first adjustment unit 411 and a second adjustment unit 412 .
第一判断单元401，用于获取发送端所发送的传输报文的往返时延的抖动，判断传输报 文的往返时延的抖动是否大于拥塞判断阈值，往返时延的抖动表示传输报文的往返时延偏离 基准往返时延的幅度。The first judging unit 401 is used to obtain the jitter of the round-trip delay of the transmission message sent by the sending end, and judge whether the jitter of the round-trip delay of the transmission message is greater than the congestion judgment threshold, and the jitter of the round-trip delay indicates the jitter of the transmission message. The magnitude by which the round-trip delay deviates from the baseline round-trip delay.
其中，第一判断单元401的相关说明可参见上述实施例中的步骤S101的相关说明部分。For the relevant description of the first judging unit 401, refer to the related description of step S101 in the above embodiment.
第二判断单元402，用于若往返时延的抖动大于拥塞判断阈值，则获取当前带宽，判断 当前带宽的下降幅度是否超出下降阈值。The second judging unit 402 is configured to acquire the current bandwidth if the jitter of the round-trip delay is greater than the congestion judging threshold, and judge whether the drop rate of the current bandwidth exceeds the dropping threshold.
其中，第二判断单元402的相关说明可参见上述实施例中的步骤S102的相关说明部分。For the relevant description of the second judging unit 402, refer to the related description of step S102 in the above embodiment.
第一处理单元403，用于若当前带宽未下降或下降幅度未超出下降阈值，则按照第一预 设微步长缩小拥塞窗口。The first processing unit 403 is configured to reduce the congestion window according to the first preset micro-step if the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold.
其中，第一处理单元403的相关说明可参见上述实施例中的步骤S103的相关说明部分。For relevant descriptions of the first processing unit 403, reference may be made to relevant descriptions of step S103 in the foregoing embodiments.
在本发明实施例中，可利用传输报文的往返时延的抖动和网络的带宽，共同进行网络是 否发生拥塞的判断。在判断往返时延的抖动大于拥塞判断阈值，但当前带宽未下降或下降幅 度未超出下降阈值时，判断数据传输出现的是随机丢包，并未发生网络拥塞，对拥塞窗口进 行微调。并未对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓解随 机丢包所带来的影响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络带宽 整体的利用率。In the embodiment of the present invention, the jitter of the round-trip delay of the transmitted message and the bandwidth of the network can be used to jointly judge whether the network is congested. When it is judged that the round-trip delay jitter is greater than the congestion judgment threshold, but the current bandwidth does not drop or the drop does not exceed the drop threshold, it is judged that random packet loss occurs in data transmission and no network congestion occurs, and the congestion window is fine-tuned. The congestion window has not been greatly adjusted, which avoids misjudgment of network congestion, so that on the basis of mitigating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network performance. Overall bandwidth utilization.
第三判断单元404，用于若当前带宽的下降幅度超出下降阈值，则获取网络丢包参数， 判断网络丢包参数是否满足网络拥塞判断条件，网络丢包参数记录了传输报文在传输过程中 数据包的丢包信息，网络拥塞条件包括至少一项条件。The third judging unit 404 is used to obtain the network packet loss parameter if the decrease in the current bandwidth exceeds the drop threshold, and judge whether the network packet loss parameter meets the network congestion judgment condition. The network packet loss parameter records the transmission message during transmission. The packet loss information of the data packet, the network congestion condition includes at least one condition.
其中，第三判断单元404的相关说明可参见上述实施例中的步骤S104的相关说明部分。For the relevant description of the third judging unit 404, please refer to the related description of step S104 in the above embodiment.
第二处理单元405，用于若网络丢包参数满足网络拥塞判断条件中任意一项，则按照第 一预设粗步长缩小拥塞窗口，第一预设粗步长大于第一预设微步长。The second processing unit 405 is configured to reduce the congestion window according to the first preset coarse step if the network packet loss parameter meets any one of the network congestion judgment conditions, and the first preset coarse step is larger than the first preset microstep long.
其中，第二处理单元405的相关说明可参见上述实施例中的步骤S105的相关说明部分。For relevant descriptions of the second processing unit 405, reference may be made to relevant descriptions of step S105 in the foregoing embodiments.
在一个示例中，网络丢包参数包括连续丢包数目，网络拥塞判断条件包括连续丢包数目 大于等于两个；和/或，网络丢包参数包括丢包率，网络拥塞判断条件包括丢包率大于预设的 拥塞丢包率阈值；和/或，网络丢包参数包括丢包重传报文数目，网络拥塞判断条件包括丢包 重传报文数目大于预设的拥塞丢包重传报文阈值。In one example, the network packet loss parameter includes the number of consecutive packet loss, and the network congestion judgment condition includes that the number of consecutive packet loss is greater than or equal to two; and/or, the network packet loss parameter includes a packet loss rate, and the network congestion judgment condition includes a packet loss rate greater than the preset congestion packet loss rate threshold; and/or, the network packet loss parameters include the number of packet loss and retransmission packets, and the network congestion judgment condition includes the number of packet loss and retransmission packets being greater than the preset congestion packet loss and retransmission packets threshold.
在本发明实施例中，结合往返时延抖动、带宽和网络丢包参数三个方向的参数，判断是 否发生网络拥塞丢包，进一步提高了判断网络拥塞丢包的准确性。当传输报文的往返时延抖 动大于拥塞判断阈值，且当前带宽的下降幅度超出下降阈值，且网络丢包参数满足网络拥塞 判断条件中任意一项时，确定发生网络拥塞丢包，较大幅度缩小拥塞窗口，以缓解网络拥塞 情况。In the embodiment of the present invention, combined with the parameters of the three directions of round-trip delay jitter, bandwidth and network packet loss parameters, it is judged whether network congestion packet loss occurs, which further improves the accuracy of judging network congestion packet loss. When the round-trip delay jitter of transmitted packets is greater than the congestion judgment threshold, and the current bandwidth drop exceeds the drop threshold, and the network packet loss parameters meet any of the network congestion judgment conditions, it is determined that network congestion packet loss occurs, with a relatively large Reduce the congestion window to alleviate network congestion.
第四判断单元406，用于若往返时延的抖动小于等于拥塞判断阈值，则判断往返时延的 抖动是否处于非拥塞抖动范围内，非拥塞抖动范围为传输报文在传输过程中未发生网络拥塞 时往返时延的抖动的变化范围，且拥塞判断阈值大于非拥塞抖动范围的上限值。The fourth judging unit 406 is configured to judge whether the round-trip delay jitter is within the range of non-congested jitter if the jitter of the round-trip delay is less than or equal to the congestion judgment threshold. The variation range of the round-trip delay jitter during congestion, and the congestion judgment threshold is greater than the upper limit of the non-congested jitter range.
其中，第四判断单元406的相关说明可参见上述实施例中的步骤S107的相关说明部分。For the relevant description of the fourth judging unit 406, please refer to the related description of step S107 in the above embodiment.
第三处理单元407，用于若往返时延的抖动处于非拥塞抖动范围内，保持拥塞窗口不变。The third processing unit 407 is configured to keep the congestion window unchanged if the jitter of the round-trip delay is within the non-congested jitter range.
其中，第三处理单元407的相关说明可参见上述实施例中的步骤S110的相关说明部分。For the relevant description of the third processing unit 407, reference may be made to the relevant description of step S110 in the foregoing embodiment.
第四处理单元408，用于若往返时延的抖动小于非拥塞抖动范围的下限值，则按照第二 预设微步长扩大拥塞窗口。The fourth processing unit 408 is configured to expand the congestion window according to the second preset micro-step size if the jitter of the round-trip delay is less than the lower limit of the non-congested jitter range.
其中，第四处理单元408的相关说明可参见上述实施例中的步骤S108的相关说明部分。Wherein, for relevant descriptions of the fourth processing unit 408, reference may be made to relevant descriptions of step S108 in the foregoing embodiment.
第五处理单元409，用于若往返时延的抖动大于非拥塞抖动范围的上限值，则按照第三 预设微步长缩小拥塞窗口。The fifth processing unit 409 is configured to reduce the congestion window according to a third preset micro-step if the jitter of the round-trip delay is greater than the upper limit of the non-congested jitter range.
其中，第五处理单元409的相关说明可参见上述实施例中的步骤S109的相关说明部分。Wherein, for relevant descriptions of the fifth processing unit 409, reference may be made to relevant descriptions of step S109 in the foregoing embodiment.
第六处理单元410，用于若网络丢包参数均未满足网络拥塞判断条件中任意一项，则按 照第四预设微步长缩小拥塞窗口。The sixth processing unit 410 is configured to reduce the congestion window according to the fourth preset micro-step size if the network packet loss parameters do not meet any of the network congestion judgment conditions.
其中，第六处理单元410的相关说明可参见上述实施例中的步骤S106的相关说明部分。Wherein, for relevant descriptions of the sixth processing unit 410, reference may be made to relevant descriptions of step S106 in the foregoing embodiments.
在本发明实施例中，判定未发生网络拥塞，则平缓适当调整拥塞窗口，可最大限度的保 证了网络带宽的利用率，从而提高了网络带宽整体的利用率。In the embodiment of the present invention, if it is determined that no network congestion occurs, then the congestion window is adjusted gently and appropriately, which can ensure the utilization rate of the network bandwidth to the greatest extent, thereby improving the overall utilization rate of the network bandwidth.
第一调整单元411，被配置为若当前带宽未下降或下降幅度未超出下降阈值，按照第一 预定补偿步长，增加非拥塞抖动范围的上限值。The first adjustment unit 411 is configured to increase the upper limit of the non-congested jitter range according to the first predetermined compensation step if the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold.
其中，第一调整单元411的相关说明可参见上述实施例中的步骤S111的相关说明部分。For the relevant description of the first adjustment unit 411, refer to the related description of step S111 in the above embodiment.
第二调整单元412，被配置为若网络丢包参数均未满足网络拥塞判断条件中任意一项， 按照第二预定补偿步长，增加非拥塞抖动范围的上限值。The second adjustment unit 412 is configured to increase the upper limit of the non-congested jitter range according to the second predetermined compensation step size if none of the network packet loss parameters meets any one of the network congestion judgment conditions.
其中，第二调整单元412的相关说明可参见上述实施例中的步骤S112的相关说明部分。For the relevant description of the second adjusting unit 412, refer to the related description of step S112 in the above embodiment.
在本发明实施例中，可以根据往返时延的抖动判断、带宽判断、网络丢包参数判断的三 个判断所反馈的内容，一方面，调节拥塞窗口，以实现在避免或缓解网络拥塞的情况下最大 限度保证网络带宽的利用率的目的；另一方面，也可根据三个判断的反馈，对非拥塞抖动范 围进行优化更新，从而提高网络拥塞判断的精准度和效率。In the embodiment of the present invention, according to the content fed back by the three judgments of round-trip delay jitter judgment, bandwidth judgment, and network packet loss parameter judgment, on the one hand, the congestion window is adjusted to avoid or alleviate network congestion. On the other hand, according to the feedback of the three judgments, the non-congested jitter range can be optimized and updated, so as to improve the accuracy and efficiency of network congestion judgment.
图6为本发明另一实施例中一种数据传输装置500的结构示意图。如图6所示，数据传 输装置500可包括第一判断单元501、第二判断单元502、第一处理单元503、第三判断单元 504、第二处理单元505、第四判断单元506、第三处理单元507、第四处理单元508、第五处理单元509、第六处理单元510、第一调整单元511和第二调整单元512。FIG. 6 is a schematic structural diagram of a data transmission device 500 in another embodiment of the present invention. As shown in Figure 6, the data transmission device 500 may include a first judging unit 501, a second judging unit 502, a first processing unit 503, a third judging unit 504, a second processing unit 505, a fourth judging unit 506, a third A processing unit 507 , a fourth processing unit 508 , a fifth processing unit 509 , a sixth processing unit 510 , a first adjustment unit 511 and a second adjustment unit 512 .
第一判断单元501，用于获取发送的传输报文的往返时延的抖动，判断传输报文的往返 时延的抖动是否大于拥塞判断阈值，往返时延的抖动表征传输报文的往返时延偏离基准往返 时延的幅度。The first judgment unit 501 is used to obtain the jitter of the round-trip delay of the transmitted transmission message, and judge whether the jitter of the round-trip delay of the transmission message is greater than the congestion judgment threshold, and the jitter of the round-trip delay represents the round-trip delay of the transmission message The magnitude of the deviation from the baseline round-trip delay.
其中，第一判断单元501的相关说明可参见上述实施例中的步骤S201的相关说明部分。For the relevant description of the first judging unit 501, refer to the related description of step S201 in the above embodiment.
第二判断单元502，用于若往返时延的抖动大于拥塞判断阈值，则获取网络丢包参数， 判断网络丢包参数是否满足网络拥塞判断条件，网络丢包参数记录了传输报文在传输过程中 数据包的丢包信息。The second judging unit 502 is used to obtain the network packet loss parameter if the jitter of the round-trip delay is greater than the congestion judgment threshold, and judge whether the network packet loss parameter meets the network congestion judgment condition. Packet loss information of the data packets in.
其中，第二判断单元502的相关说明可参见上述实施例中的步骤S202的相关说明部分。For the relevant description of the second judging unit 502, please refer to the related description of step S202 in the above embodiment.
第一处理单元503，用于若网络丢包参数均未满足网络拥塞判断条件中任意一项，则按 照第一预设微步长缩小拥塞窗口，网络拥塞条件包括至少一项条件。The first processing unit 503 is configured to reduce the congestion window according to the first preset micro-step size if the network packet loss parameters do not meet any of the network congestion judgment conditions, and the network congestion conditions include at least one condition.
其中，第一处理单元503的相关说明可参见上述实施例中的步骤S203的相关说明部分。For relevant descriptions of the first processing unit 503, reference may be made to relevant descriptions of step S203 in the foregoing embodiments.
在一个示例中，网络丢包参数包括连续丢包数目，网络拥塞判断条件包括连续丢包数目 大于等于两个；和/或，网络丢包参数包括丢包率，网络拥塞判断条件包括丢包率大于预设的 拥塞丢包率阈值；和/或，网络丢包参数包括丢包重传报文数目，网络拥塞判断条件包括丢包 重传报文数目大于预设的拥塞丢包重传报文阈值。In one example, the network packet loss parameter includes the number of consecutive packet loss, and the network congestion judgment condition includes that the number of consecutive packet loss is greater than or equal to two; and/or, the network packet loss parameter includes a packet loss rate, and the network congestion judgment condition includes a packet loss rate greater than the preset congestion packet loss rate threshold; and/or, the network packet loss parameters include the number of packet loss and retransmission packets, and the network congestion judgment condition includes the number of packet loss and retransmission packets being greater than the preset congestion packet loss and retransmission packets threshold.
在本发明实施例中，利用传输报文的往返时延的抖动和网络丢包参数，共同进行网络是 否发生拥塞的判断。在判断往返时延的抖动大于拥塞判断阈值，但网络丢包参数均未满足网 络拥塞判断条件中任意一项时，判断数据传输出现的是随机丢包，并未发生网络拥塞，对拥 塞窗口进行微调。并未对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而 在缓解随机丢包所带来的影响的基础上，最大限度的保证了网络带宽的利用率，从而提高了 网络带宽整体的利用率。In the embodiment of the present invention, the jitter of the round-trip delay of the transmitted message and the network packet loss parameter are used to jointly judge whether the network is congested. When it is judged that the jitter of the round-trip delay is greater than the congestion judgment threshold, but the network packet loss parameters do not meet any of the network congestion judgment conditions, it is judged that random packet loss occurs in data transmission, and no network congestion occurs, and the congestion window is checked. fine-tuning. The congestion window has not been greatly adjusted, which avoids misjudgment of network congestion, so that on the basis of mitigating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network performance. Overall bandwidth utilization.
第三判断单元504，用于若网络丢包参数满足网络拥塞判断条件中任意一项，则获取当前 带宽，判断当前带宽的下降幅度是否超出下降阈值。The third judging unit 504 is used to obtain the current bandwidth if the network packet loss parameter satisfies any one of the network congestion judging conditions, and judge whether the drop rate of the current bandwidth exceeds the drop threshold.
其中，第三判断单元504的相关说明可参见上述实施例中的步骤S204的相关说明部分。For the relevant description of the third judging unit 504, please refer to the related description of step S204 in the above embodiment.
第二处理单元505，用于若当前带宽的下降幅度超出下降阈值，则按照第一预设粗步长 缩小拥塞窗口，第一预设粗步长大于第一预设微步长。The second processing unit 505 is configured to reduce the congestion window according to a first preset coarse step size if the drop rate of the current bandwidth exceeds the drop threshold, and the first preset coarse step size is larger than the first preset micro step size.
其中，第二处理单元505的相关说明可参见上述实施例中的步骤S205的相关说明部分。For relevant descriptions of the second processing unit 505, reference may be made to relevant descriptions of step S205 in the foregoing embodiments.
在本发明实施例中，结合往返时延的抖动、网络丢包参数和带宽三个方向的参数，判断 是否发生网络拥塞丢包，进一步提高了判断网络拥塞丢包的准确性。当传输报文的往返时延 的抖动大于拥塞判断阈值，且网络丢包参数满足网络拥塞判断条件中任意一项，且当前带宽 的下降幅度超出下降阈值时，确定发生网络拥塞丢包，较大幅度缩小拥塞窗口，以缓解网络 拥塞情况。In the embodiment of the present invention, combined with the round-trip delay jitter, network packet loss parameters and parameters in three directions of bandwidth, it is judged whether network congestion packet loss occurs, which further improves the accuracy of judging network congestion packet loss. When the jitter of the round-trip delay of the transmitted message is greater than the congestion judgment threshold, and the network packet loss parameter meets any one of the network congestion judgment conditions, and the decrease of the current bandwidth exceeds the decrease threshold, it is determined that the packet loss due to network congestion occurs. Reduce the congestion window significantly to alleviate network congestion.
第四判断单元506，用于若往返时延的抖动小于等于拥塞判断阈值，则判断往返时延的 抖动是否处于非拥塞抖动范围内，非拥塞抖动范围为传输报文在传输过程中未发生网络拥塞 时往返时延的抖动的变化范围，且拥塞判断阈值大于非拥塞抖动范围的上限值。The fourth judging unit 506 is configured to determine whether the round-trip delay jitter is within the range of non-congested jitter if the jitter of the round-trip delay is less than or equal to the congestion judgment threshold. The variation range of the round-trip delay jitter during congestion, and the congestion judgment threshold is greater than the upper limit of the non-congested jitter range.
其中，第四判断单元506的相关说明可参见上述实施例中的步骤S206的相关说明部分。For the relevant description of the fourth judging unit 506, please refer to the related description of step S206 in the above embodiment.
第三处理单元507，用于若往返时延抖动处于非拥塞抖动范围内，保持拥塞窗口不变。The third processing unit 507 is configured to keep the congestion window unchanged if the round-trip delay jitter is within the non-congested jitter range.
其中，第三处理单元507的相关说明可参见上述实施例中的步骤S209的相关说明部分。For the relevant description of the third processing unit 507, refer to the related description of step S209 in the above embodiment.
第四处理单元508，用于若往返时延的抖动小于非拥塞抖动范围的下限，则按照第二预 设微步长扩大拥塞窗口。The fourth processing unit 508 is configured to expand the congestion window according to the second preset micro-step size if the jitter of the round-trip delay is less than the lower limit of the non-congested jitter range.
其中，第四处理单元508的相关说明可参见上述实施例中的步骤S207的相关说明部分。Wherein, for relevant descriptions of the fourth processing unit 508, reference may be made to relevant descriptions of step S207 in the foregoing embodiment.
第五处理单元509，用于若往返时延的抖动大于非拥塞抖动范围的上限，则按照第三预 设微步长缩小拥塞窗口。The fifth processing unit 509 is configured to reduce the congestion window according to a third preset micro-step if the jitter of the round-trip delay is greater than the upper limit of the non-congested jitter range.
其中，第五处理单元509的相关说明可参见上述实施例中的步骤S208的相关说明部分。For relevant descriptions of the fifth processing unit 509, reference may be made to relevant descriptions of step S208 in the foregoing embodiments.
第六处理单元510，用于若当前带宽未下降或下降幅度未超出下降阈值，则按照第四预 设微步长缩小拥塞窗口值The sixth processing unit 510 is configured to reduce the congestion window value according to the fourth preset micro-step size if the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold
其中，第六处理单元510的相关说明可参见上述实施例中的步骤S210的相关说明部分。For relevant descriptions of the sixth processing unit 510, reference may be made to relevant descriptions of step S210 in the foregoing embodiments.
在本发明实施例中，确定是随机丢包，并未发生网络拥塞，数据传输通畅，可平缓地适 当扩大拥塞窗口,对拥塞窗口进行微调。或者，平缓地适当缩小拥塞窗口，对拥塞窗口进行微 调。并未对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓解随机丢 包所带来的影响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络带宽整体 的利用率。或者确定未发生网络拥塞，且数据传输通畅，保持拥塞窗口不变，避免不必要的 占用资源的拥塞窗口调节。In the embodiment of the present invention, it is determined that there is random packet loss, there is no network congestion, and the data transmission is smooth, so the congestion window can be appropriately expanded gently, and the congestion window can be fine-tuned. Or, reduce the congestion window gradually and appropriately, and fine-tune the congestion window. The congestion window has not been greatly adjusted, which avoids misjudgment of network congestion, so that on the basis of mitigating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network performance. Overall bandwidth utilization. Or determine that there is no network congestion and the data transmission is smooth, keep the congestion window unchanged, and avoid unnecessary resource-occupied congestion window adjustments.
第一调整单元511，被配置为若网络丢包参数均未满足网络拥塞判断条件中任意一项， 按照第一预定补偿步长，增加非拥塞抖动范围的上限值。The first adjustment unit 511 is configured to increase the upper limit of the non-congested jitter range according to the first predetermined compensation step size if none of the network packet loss parameters meets any of the network congestion judgment conditions.
其中，第一调整单元511的相关说明可参见上述实施例中的步骤S211的相关说明部分。For the relevant description of the first adjusting unit 511, refer to the related description of step S211 in the above embodiment.
第二调整单元512，被配置为若当前带宽未下降或下降幅度未超出下降阈值，按照第二 预定补偿步长，增加非拥塞抖动范围的上限值。The second adjustment unit 512 is configured to increase the upper limit of the non-congested jitter range according to the second predetermined compensation step if the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold.
其中，第二调整单元512的相关说明可参见上述实施例中的步骤S212的相关说明部分。Wherein, for the relevant description of the second adjusting unit 512, refer to the related description of step S212 in the above embodiment.
在本发明实施例中，当传输报文的往返时延的抖动大于拥塞判断阈值，但网络丢包参数 均未满足网络拥塞判断条件中任意一项，表明未发生网络拥塞。或者，当传输报文的往返时 延的抖动大于拥塞判断阈值，且当前带宽下降幅度超出下降阈值，但网络丢包参数均未满足 网络拥塞判断条件中任意一项，表明未发生网络拥塞。为了后续的网络拥塞判断更加精准且 效率更高，可适当增加非拥塞抖动范围的上限值，拓宽非拥塞抖动范围，从而实现对非拥塞 抖动范围的设定的优化。In the embodiment of the present invention, when the jitter of the round-trip delay of the transmitted message is greater than the congestion judgment threshold, but the network packet loss parameters do not meet any one of the network congestion judgment conditions, it indicates that no network congestion occurs. Or, when the jitter of the round-trip delay of transmitted packets is greater than the congestion judgment threshold, and the current bandwidth drop exceeds the drop threshold, but none of the network packet loss parameters meets any of the network congestion judgment conditions, it indicates that no network congestion occurs. In order to make the subsequent judgment of network congestion more accurate and efficient, the upper limit of the non-congested jitter range can be appropriately increased to widen the non-congested jitter range, so as to optimize the setting of the non-congested jitter range.
图7为本发明又一实施例中一种数据传输装置600的结构示意图。如图7所示，数据传 输装置600可包括第一判断单元601、第二判断单元602、第一处理单元603、第三判断单元 604、第二处理单元605、第三处理单元606、第四判断单元607、第四处理单元608、第五处理单元609和第六处理单元610。FIG. 7 is a schematic structural diagram of a data transmission device 600 in another embodiment of the present invention. As shown in Figure 7, the data transmission device 600 may include a first judging unit 601, a second judging unit 602, a first processing unit 603, a third judging unit 604, a second processing unit 605, a third processing unit 606, a fourth A judging unit 607 , a fourth processing unit 608 , a fifth processing unit 609 and a sixth processing unit 610 .
第一判断单元601，用于获取当前带宽，判断当前带宽的下降幅度是否超出下降阈值。The first judging unit 601 is configured to acquire the current bandwidth, and judge whether the drop range of the current bandwidth exceeds a drop threshold.
其中，第一判断单元601的相关说明可参见上述实施例中的步骤S301的相关说明部分。For the relevant description of the first judging unit 601, refer to the related description of step S301 in the above embodiment.
第二判断单元602，用于若当前带宽的下降幅度超出下降阈值，则获取网络丢包参数， 并判断网络丢包参数是否满足网络拥塞判断条件，网络丢包参数记录了传输报文在传输过程 中数据包的丢包信息。The second judging unit 602 is used to obtain the network packet loss parameter if the decline rate of the current bandwidth exceeds the drop threshold, and judge whether the network packet loss parameter meets the network congestion judgment condition. The network packet loss parameter records the transmission process of the transmission message Packet loss information of the data packets in.
其中，第二判断单元602的相关说明可参见上述实施例中的步骤S302的相关说明部分。For the relevant description of the second judging unit 602, refer to the related description of step S302 in the above embodiment.
第一处理单元603，用于若网络丢包参数均未满足网络拥塞判断条件中任意一项，则按 照第一预设微步长缩小拥塞窗口。The first processing unit 603 is configured to reduce the congestion window according to the first preset micro-step size if the network packet loss parameters do not meet any of the network congestion judgment conditions.
其中，第一处理单元603的相关说明可参见上述实施例中的步骤S303的相关说明部分。For relevant descriptions of the first processing unit 603, reference may be made to relevant descriptions of step S303 in the foregoing embodiments.
在一个示例中，网络丢包参数包括连续丢包数目，网络拥塞判断条件包括连续丢包数目 大于等于两个；和/或，网络丢包参数包括丢包率，网络拥塞判断条件包括丢包率大于预设的 拥塞丢包率阈值；和/或，网络丢包参数包括丢包重传报文数目，网络拥塞判断条件包括丢包 重传报文数目大于预设的拥塞丢包重传报文阈值。In one example, the network packet loss parameter includes the number of consecutive packet loss, and the network congestion judgment condition includes that the number of consecutive packet loss is greater than or equal to two; and/or, the network packet loss parameter includes a packet loss rate, and the network congestion judgment condition includes a packet loss rate greater than the preset congestion packet loss rate threshold; and/or, the network packet loss parameters include the number of packet loss and retransmission packets, and the network congestion judgment condition includes the number of packet loss and retransmission packets being greater than the preset congestion packet loss and retransmission packets threshold.
在本发明实施例中，在当前带宽的下降幅度超出下降阈值，但网络丢包参数均未满足网 络拥塞判断条件中任意一项时，确定数据传输出现的是随机丢包，并未发生网络拥塞，对拥 塞窗口进行微调。避免对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而 在缓解随机丢包所带来的影响的基础上，最大限度的保证了网络带宽的利用率，从而提高了 网络带宽整体的利用率。In the embodiment of the present invention, when the drop rate of the current bandwidth exceeds the drop threshold, but none of the network packet loss parameters meets any of the network congestion judgment conditions, it is determined that random packet loss occurs in data transmission, and no network congestion occurs , fine-tuning the congestion window. Avoid large adjustments to the congestion window and avoid misjudgment of network congestion. On the basis of alleviating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network bandwidth. overall utilization.
第三判断单元604，用于若网络丢包参数满足网络拥塞判断条件中任意一项，则获取传 输报文的往返时延的抖动，判断往返时延的抖动是否大于拥塞判断阈值。The third judging unit 604 is used to obtain the jitter of the round-trip delay of the transmitted message if the network packet loss parameter meets any one of the network congestion judgment conditions, and judge whether the jitter of the round-trip delay is greater than the congestion judgment threshold.
其中，第三判断单元604的相关说明可参见上述实施例中的步骤S304的相关说明部分。For the relevant description of the third judging unit 604, refer to the related description of step S304 in the above embodiment.
第二处理单元605，用于若往返时延的抖动大于拥塞判断阈值，则按照第一预设粗步长 缩小拥塞窗口。The second processing unit 605 is configured to reduce the congestion window according to a first preset coarse step size if the jitter of the round-trip delay is greater than the congestion judgment threshold.
其中，第二处理单元605的相关说明可参见上述实施例中的步骤S305的相关说明部分。Wherein, for relevant descriptions of the second processing unit 605, reference may be made to relevant descriptions of step S305 in the foregoing embodiments.
第三处理单元606，用于若当前带宽未下降或下降幅度未超出下降阈值，则按照第二预 设微步长缩小拥塞窗口。The third processing unit 606 is configured to reduce the congestion window according to the second preset micro-step size if the current bandwidth does not decrease or the decrease range does not exceed the decrease threshold.
其中，第三处理单元606的相关说明可参见上述实施例中的步骤S306的相关说明部分。For relevant descriptions of the third processing unit 606, reference may be made to relevant descriptions of step S306 in the foregoing embodiments.
在本发明实施例中，在经历了上述实施例中的带宽和网络丢包参数的判定后，引入往返 时延抖动，进一步提高判定网络拥塞引起的丢包的准确性。当当前带宽的下降幅度超出下降 阈值，且网络丢包参数满足网络拥塞判断条件中任意一项，且传输报文的往返时延的抖动大 于拥塞判断阈值时，确定发生网络拥塞丢包，较大幅度缩小拥塞窗口，以缓解网络拥塞情况。In the embodiment of the present invention, after the determination of the bandwidth and network packet loss parameters in the above embodiments, the round-trip delay jitter is introduced to further improve the accuracy of determining the packet loss caused by network congestion. When the decline of the current bandwidth exceeds the drop threshold, and the network packet loss parameters meet any of the network congestion judgment conditions, and the jitter of the round-trip delay of the transmitted message is greater than the congestion judgment threshold, it is determined that the packet loss due to network congestion occurs. Reduce the congestion window by a large amount to alleviate network congestion.
第四判断单元607，用于若往返时延的抖动小于等于拥塞判断阈值，则判断往返时延的 抖动是否处于非拥塞抖动范围内。The fourth judging unit 607 is configured to judge whether the jitter of the round-trip delay is within the non-congested jitter range if the jitter of the round-trip delay is less than or equal to the congestion judgment threshold.
其中，第四判断单元607的相关说明可参见上述实施例中的步骤S307的相关说明部分。Wherein, for relevant descriptions of the fourth judging unit 607, reference may be made to relevant descriptions of step S307 in the foregoing embodiments.
第四处理单元608，用于若往返时延的抖动大于非拥塞抖动范围的上限值，则按照第三 预设微步长缩小拥塞窗口。The fourth processing unit 608 is configured to reduce the congestion window according to the third preset micro-step size if the jitter of the round-trip delay is greater than the upper limit of the non-congested jitter range.
其中，第四处理单元608的相关说明可参见上述实施例中的步骤S308的相关说明部分。Wherein, for relevant descriptions of the fourth processing unit 608, reference may be made to relevant descriptions of step S308 in the foregoing embodiment.
第五处理单元609，用于若往返时延的抖动小于非拥塞抖动范围的下限值，则按照第四 预设微步长缩小拥塞窗口。The fifth processing unit 609 is configured to reduce the congestion window according to a fourth preset micro-step size if the jitter of the round-trip delay is less than the lower limit of the non-congested jitter range.
其中，第五处理单元609的相关说明可参见上述实施例中的步骤S309的相关说明部分。For relevant descriptions of the fifth processing unit 609, reference may be made to relevant descriptions of step S309 in the foregoing embodiments.
第六处理单元610，用于若往返时延的抖动处于非拥塞抖动范围内，则按照第五预设微 步长缩小拥塞窗口。The sixth processing unit 610 is configured to reduce the congestion window according to the fifth preset micro-step if the jitter of the round-trip delay is within the non-congested jitter range.
其中，第六处理单元610的相关说明可参见上述实施例中的步骤S310的相关说明部分。Wherein, for relevant descriptions of the sixth processing unit 610, reference may be made to relevant descriptions of step S310 in the foregoing embodiments.
在本发明实施例中，确定数据传输出现的是随机丢包，并未发生网络拥塞，对拥塞窗口 进行微调。避免对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况，从而在缓解 随机丢包所带来的影响的基础上，最大限度的保证了网络带宽的利用率，从而提高了网络带 宽整体的利用率。In the embodiment of the present invention, it is determined that random packet loss occurs in data transmission, and no network congestion occurs, and the congestion window is fine-tuned. Avoid large adjustments to the congestion window and avoid misjudgment of network congestion. On the basis of alleviating the impact of random packet loss, the utilization rate of network bandwidth is guaranteed to the maximum extent, thereby improving network bandwidth. overall utilization.
结合图1至图7描述的根据本发明实施例的数据传输方法和装置可以由发送端700来实现。图8为发明一实施例中的一种发送端700的硬件结构示意图。The data transmission method and apparatus according to the embodiments of the present invention described in conjunction with FIGS. 1 to 7 may be implemented by the sending end 700 . FIG. 8 is a schematic diagram of a hardware structure of a sending end 700 in an embodiment of the invention.
发送端700包括存储器701、处理器702及存储在存储器701上并可在处理器702 上运行的计算机程序。The sending end 700 includes a memory 701 , a processor 702 and a computer program stored in the memory 701 and operable on the processor 702 .
在一个示例中，上述处理器702可以包括中央处理器(CPU)，或者特定集成电路(ASIC)，或者可以被配置成实施本发明实施例的一个或多个集成电路。In one example, the above-mentioned processor 702 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present invention.
存储器701可以包括用于数据或指令的大容量存储器。举例来说而非限制，存储器701可包括HDD、软盘驱动器、闪存、光盘、磁光盘、磁带或通用串行总线(USB)驱动器 或者两个或更多个以上这些的组合。在合适的情况下，存储器701可包括可移除或不可 移除(或固定)的介质。在合适的情况下，存储器701可在发送端700的内部或外部。在 特定实施例中，存储器701是非易失性固态存储器。在特定实施例中，存储器701包括 只读存储器(ROM)。在合适的情况下，该ROM可以是掩模编程的ROM、可编程ROM(PROM)、 可擦除PROM(EPROM)、电可擦除PROM(EEPROM)、电可改写ROM(EAROM)或闪存或者两个或 更多个以上这些的组合。Memory 701 may include mass storage for data or instructions. By way of example and not limitation, memory 701 may include a HDD, floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or universal serial bus (USB) drive, or a combination of two or more of these. Storage 701 may comprise removable or non-removable (or fixed) media, where appropriate. Where appropriate, the memory 701 may be inside or outside the sending end 700 . In a particular embodiment, memory 701 is a non-volatile solid-state memory. In particular embodiments, memory 701 includes read-only memory (ROM). Where appropriate, the ROM may be mask programmed ROM, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), electrically rewritable ROM (EAROM), or flash memory or A combination of two or more of the above.
处理器702通过读取存储器701中存储的可执行程序代码来运行与可执行程序代码 对应的程序，以用于执行上述各个实施例中的数据传输方法。The processor 702 runs the program corresponding to the executable program code by reading the executable program code stored in the memory 701, so as to execute the data transmission method in each of the foregoing embodiments.
在一个示例中，发送端700还可包括通信接口703和总线704。其中，如图8所示， 存储器701、处理器702、通信接口703通过总线704连接并完成相互间的通信。In an example, the sending end 700 may further include a communication interface 703 and a bus 704 . Wherein, as shown in FIG. 8 , the memory 701 , the processor 702 , and the communication interface 703 are connected through a bus 704 to complete mutual communication.
通信接口703，主要用于实现本发明实施例中各模块、装置、单元和/或设备之间的通信。也可通信接口703接入输入设备和/或输出设备。The communication interface 703 is mainly used to realize the communication between various modules, devices, units and/or devices in the embodiments of the present invention. The communication interface 703 can also be connected to input devices and/or output devices.
总线704包括硬件、软件或两者，将发送端700的部件彼此耦接在一起。举例来说而非限制，总线704可包括加速图形端口(AGP)或其他图形总线、增强工业标准架构(EISA)总线、前端总线(FSB)、超传输(HT)互连、工业标准架构(ISA)总线、无限带宽互连、低 引脚数(LPC)总线、存储器总线、微信道架构(MCA)总线、外围组件互连(PCI)总线、 PCI-Express(PCI-X)总线、串行高级技术附件(SATA)总线、视频电子标准协会局部(VLB) 总线或其他合适的总线或者两个或更多个以上这些的组合。在合适的情况下，总线704 可包括一个或多个总线。尽管本发明实施例描述和示出了特定的总线，但本发明考虑任 何合适的总线或互连。The bus 704 includes hardware, software or both, and couples the components of the sending end 700 to each other. By way of example and not limitation, bus 704 may include Accelerated Graphics Port (AGP) or other graphics bus, Enhanced Industry Standard Architecture (EISA) bus, Front Side Bus (FSB), HyperTransport (HT) interconnect, Industry Standard Architecture (ISA ) bus, InfiniBand Interconnect, Low Pin Count (LPC) bus, memory bus, Micro Channel Architecture (MCA) bus, Peripheral Component Interconnect (PCI) bus, PCI-Express (PCI-X) bus, Serial Advanced A Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable bus or a combination of two or more of these. Bus 704 may comprise one or more buses, where appropriate. Although the embodiments of the invention describe and illustrate a particular bus, the invention contemplates any suitable bus or interconnect.
本发明一实施例还提供一种存储介质，该存储介质上存储有程序，该程序被处理器执行 时可实现上述各个实施例中的数据传输方法。An embodiment of the present invention also provides a storage medium, on which a program is stored, and when the program is executed by a processor, the data transmission method in each of the above-mentioned embodiments can be realized.
本发明实施例提供的数据传输设备和存储介质，可利用传输报文的往返时延的抖动和网 络的带宽，共同进行网络是否发生拥塞的判断。在判断往返时延的抖动大于拥塞判断阈值， 但当前带宽未下降或下降幅度未超出下降阈值时，判断数据传输出现的是随机丢包，并未发 生网络拥塞，对拥塞窗口进行微调。或者，利用传输报文的往返时延的抖动和网络丢包参数， 共同进行网络是否发生拥塞的判断。在判断往返时延的抖动大于拥塞判断阈值，但网络丢包 参数均未满足网络拥塞判断条件中任意一项时，判断数据传输出现的是随机丢包，并未发生 网络拥塞，对拥塞窗口进行微调。或者，可利用网络的带宽和网络丢包参数，共同进行网络 是否发生拥塞的判断。在判断网络的带宽的下降幅度超出下降阈值，但网络丢包参数均未满 足网络拥塞判断条件中任意一项时，判断数据传输出现的是随机丢包，并未发生网络拥塞， 对拥塞窗口进行微调。并未对拥塞窗口进行大幅度调整，避免了出现网络拥塞误判的情况， 从而在缓解随机丢包所带来的影响的基础上，最大限度的保证了网络带宽的利用率，从而提 高了网络带宽整体的利用率。The data transmission device and the storage medium provided by the embodiment of the present invention can use the jitter of the round-trip delay of the transmitted message and the bandwidth of the network to jointly judge whether the network is congested. When it is judged that the jitter of the round-trip delay is greater than the congestion judgment threshold, but the current bandwidth does not drop or the drop rate does not exceed the drop threshold, it is judged that random packet loss occurs in data transmission and no network congestion occurs, and the congestion window is fine-tuned. Or, the jitter of the round-trip delay of the transmitted message and the network packet loss parameter are used to jointly judge whether the network is congested. When it is judged that the jitter of the round-trip delay is greater than the congestion judgment threshold, but the network packet loss parameters do not meet any of the network congestion judgment conditions, it is judged that random packet loss occurs in data transmission, and no network congestion occurs, and the congestion window is checked. fine-tuning. Alternatively, network bandwidth and network packet loss parameters can be used to jointly determine whether network congestion occurs. When it is judged that the drop in network bandwidth exceeds the drop threshold, but none of the network packet loss parameters meets any of the network congestion judgment conditions, it is judged that random packet loss occurs in data transmission and no network congestion occurs, and the congestion window is checked. fine-tuning. The congestion window has not been greatly adjusted, which avoids misjudgment of network congestion, thereby mitigating the impact of random packet loss, maximizing the utilization of network bandwidth, and improving network performance. Overall bandwidth utilization.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711140319.4ACN108075988A (en) | 2017-11-16 | 2017-11-16 | Data transmission method and device |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711140319.4ACN108075988A (en) | 2017-11-16 | 2017-11-16 | Data transmission method and device |
| Publication Number | Publication Date |
|---|---|
| CN108075988Atrue CN108075988A (en) | 2018-05-25 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711140319.4APendingCN108075988A (en) | 2017-11-16 | 2017-11-16 | Data transmission method and device |
| Country | Link |
|---|---|
| CN (1) | CN108075988A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109039728A (en)* | 2018-07-24 | 2018-12-18 | 烽火通信科技股份有限公司 | Flow congestion detection method and system based on BFD |
| CN109474538A (en)* | 2018-12-29 | 2019-03-15 | 北京达佳互联信息技术有限公司 | A kind of data transmission method, device, terminal device and storage medium |
| CN110417669A (en)* | 2018-07-09 | 2019-11-05 | 腾讯科技（深圳）有限公司 | Data packet retransmission control, network congestion detection method and device |
| CN110896368A (en)* | 2019-11-01 | 2020-03-20 | 咪咕文化科技有限公司 | Network quality monitoring method and device |
| CN112702230A (en)* | 2020-12-21 | 2021-04-23 | 厦门亿联网络技术股份有限公司 | Network congestion detection method, system, terminal device and readable storage medium |
| CN112887140A (en)* | 2021-01-25 | 2021-06-01 | 中国科学院沈阳自动化研究所 | Multi-cloud collaborative network-cut-off disaster recovery method |
| CN113179220A (en)* | 2021-05-12 | 2021-07-27 | 中移智行网络科技有限公司 | Data transmission method, data transmission device and computer readable storage medium |
| CN113747498A (en)* | 2021-08-03 | 2021-12-03 | 上海交通大学 | Configurable low-delay congestion control method and system for mobile cellular network |
| CN115426689A (en)* | 2022-08-31 | 2022-12-02 | 北京奇艺世纪科技有限公司 | Message transmission method and device, electronic equipment and storage medium |
| CN115714746A (en)* | 2022-10-13 | 2023-02-24 | 上海赛连信息科技有限公司 | Congestion control method and system based on delay and packet loss |
| CN116170376A (en)* | 2021-11-25 | 2023-05-26 | 中移在线服务有限公司 | Data transmission control method, device, equipment and computer storage medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102388584A (en)* | 2011-09-22 | 2012-03-21 | 华为技术有限公司 | Congestion control method and device |
| CN102468941A (en)* | 2010-11-18 | 2012-05-23 | 华为技术有限公司 | Network packet loss processing method and device |
| CN103152278A (en)* | 2013-01-31 | 2013-06-12 | 北京星网锐捷网络技术有限公司 | Congestion determination method, congestion determination device and congestion determination network equipment |
| CN103929370A (en)* | 2013-01-11 | 2014-07-16 | 中国科学院声学研究所 | A TCP Congestion Control Method for Bandwidth Reservation Network |
| CN105049369A (en)* | 2015-08-14 | 2015-11-11 | 浙江大学 | Video transmission congestion control method based on MPTCP in heterogeneous wireless network |
| CN105991462A (en)* | 2015-03-02 | 2016-10-05 | 华为技术有限公司 | Transmission control protocol (TCP) data packet transmission method, transmission device and system |
| CN107800642A (en)* | 2016-08-31 | 2018-03-13 | 北京金山云网络技术有限公司 | A kind of jamming control method and device |
| US20180270157A1 (en)* | 2015-10-21 | 2018-09-20 | Citrix Systems, Inc. | System and method of providing improved throughput control under delay-based congestion situation in a network |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102468941A (en)* | 2010-11-18 | 2012-05-23 | 华为技术有限公司 | Network packet loss processing method and device |
| CN102388584A (en)* | 2011-09-22 | 2012-03-21 | 华为技术有限公司 | Congestion control method and device |
| CN103929370A (en)* | 2013-01-11 | 2014-07-16 | 中国科学院声学研究所 | A TCP Congestion Control Method for Bandwidth Reservation Network |
| CN103152278A (en)* | 2013-01-31 | 2013-06-12 | 北京星网锐捷网络技术有限公司 | Congestion determination method, congestion determination device and congestion determination network equipment |
| CN105991462A (en)* | 2015-03-02 | 2016-10-05 | 华为技术有限公司 | Transmission control protocol (TCP) data packet transmission method, transmission device and system |
| CN105049369A (en)* | 2015-08-14 | 2015-11-11 | 浙江大学 | Video transmission congestion control method based on MPTCP in heterogeneous wireless network |
| US20180270157A1 (en)* | 2015-10-21 | 2018-09-20 | Citrix Systems, Inc. | System and method of providing improved throughput control under delay-based congestion situation in a network |
| CN107800642A (en)* | 2016-08-31 | 2018-03-13 | 北京金山云网络技术有限公司 | A kind of jamming control method and device |
| Title |
|---|
| 丁囡: "基于丢包区分的TCP拥塞控制算法研究", 《中国优秀硕士学位论文全文数据库》* |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110417669A (en)* | 2018-07-09 | 2019-11-05 | 腾讯科技（深圳）有限公司 | Data packet retransmission control, network congestion detection method and device |
| CN110417669B (en)* | 2018-07-09 | 2022-08-26 | 腾讯科技（深圳）有限公司 | Data packet retransmission control method and device and network congestion detection method and device |
| CN109039728B (en)* | 2018-07-24 | 2021-08-03 | 烽火通信科技股份有限公司 | BFD-based flow congestion detection method and system |
| CN109039728A (en)* | 2018-07-24 | 2018-12-18 | 烽火通信科技股份有限公司 | Flow congestion detection method and system based on BFD |
| CN109474538B (en)* | 2018-12-29 | 2021-07-30 | 北京达佳互联信息技术有限公司 | Data transmission method and device, terminal equipment and storage medium |
| CN109474538A (en)* | 2018-12-29 | 2019-03-15 | 北京达佳互联信息技术有限公司 | A kind of data transmission method, device, terminal device and storage medium |
| CN110896368A (en)* | 2019-11-01 | 2020-03-20 | 咪咕文化科技有限公司 | Network quality monitoring method and device |
| CN112702230B (en)* | 2020-12-21 | 2022-07-08 | 厦门亿联网络技术股份有限公司 | Network congestion detection method, system, terminal device and readable storage medium |
| CN112702230A (en)* | 2020-12-21 | 2021-04-23 | 厦门亿联网络技术股份有限公司 | Network congestion detection method, system, terminal device and readable storage medium |
| CN112887140A (en)* | 2021-01-25 | 2021-06-01 | 中国科学院沈阳自动化研究所 | Multi-cloud collaborative network-cut-off disaster recovery method |
| CN112887140B (en)* | 2021-01-25 | 2022-02-11 | 中国科学院沈阳自动化研究所 | Multi-cloud collaborative network-cut-off disaster recovery method |
| CN113179220A (en)* | 2021-05-12 | 2021-07-27 | 中移智行网络科技有限公司 | Data transmission method, data transmission device and computer readable storage medium |
| CN113747498A (en)* | 2021-08-03 | 2021-12-03 | 上海交通大学 | Configurable low-delay congestion control method and system for mobile cellular network |
| CN113747498B (en)* | 2021-08-03 | 2023-11-28 | 上海交通大学 | Configurable low-delay congestion control method and system for mobile cellular network |
| CN116170376A (en)* | 2021-11-25 | 2023-05-26 | 中移在线服务有限公司 | Data transmission control method, device, equipment and computer storage medium |
| CN115426689A (en)* | 2022-08-31 | 2022-12-02 | 北京奇艺世纪科技有限公司 | Message transmission method and device, electronic equipment and storage medium |
| CN115714746A (en)* | 2022-10-13 | 2023-02-24 | 上海赛连信息科技有限公司 | Congestion control method and system based on delay and packet loss |
| Publication | Publication Date | Title |
|---|---|---|
| CN108075988A (en) | Data transmission method and device | |
| US9276866B2 (en) | Tuning congestion notification for data center networks | |
| US11838208B2 (en) | Congestion control method and apparatus, device, and storage medium | |
| CN110661723B (en) | Data transmission method, computing device, network device and data transmission system | |
| US8854992B2 (en) | Artificial delay inflation and jitter reduction to improve TCP throughputs | |
| CN113726671B (en) | Network congestion control method and related products | |
| CN103986548B (en) | A kind of method and terminal for determining packet loss reason | |
| US11258717B2 (en) | Method for sending service packet, network device, and system | |
| EP3525406A1 (en) | Method and apparatus for determining tcp congestion window | |
| CN102204182A (en) | Method and apparatus for controlling congestion of data transmission | |
| EP4145780A1 (en) | Method for controlling network congestion and network device | |
| JP5832335B2 (en) | Communication apparatus and communication system | |
| CN106789718A (en) | The jamming control method of data transfer, equipment, server and programmable device | |
| CN107800638B (en) | Congestion control method and device | |
| CN106856457B (en) | A data transmission method, sending device and receiving device | |
| WO2020192397A1 (en) | Adjustment method for transmitting apparatus and communication device | |
| WO2017114231A1 (en) | Packet sending method, tcp proxy, and tcp client | |
| US9882820B2 (en) | Communication apparatus | |
| CN115022227B (en) | Data transmission method and system based on circulation or rerouting in data center network | |
| US20170346749A1 (en) | Method and system for upload optimization | |
| WO2018137158A1 (en) | Data transmission method and apparatus, and customer premises equipment | |
| CN116266826A (en) | A distributed machine learning network optimization system, method and electronic equipment | |
| EP4117252A1 (en) | Traffic control method and apparatus | |
| CN115866095A (en) | Data transmission method and device, electronic equipment and storage medium | |
| CN117014379A (en) | Code rate control method, device, electronic equipment and storage medium |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20180525 | |
| RJ01 | Rejection of invention patent application after publication |