技术领域Technical Field
本发明属于无线通信技术领域，具体涉及一种基于SAGIN的任务卸载与资源分配联合优化方法及系统。The present invention belongs to the technical field of wireless communications, and in particular relates to a SAGIN-based task offloading and resource allocation joint optimization method and system.
背景技术Background Art
由于地面网络覆盖不足，地面与卫星的融合近年来受到越来越多的关注。利用空天地一体化网络(SAGIN)为远程物联网(IoT)设备的计算卸载提供无缝、灵活的网络服务。与地面不同，前者需要综合考虑信道条件、卫星计算资源承载能力等多种因素。因此，需要设计一个有效的卫星IoT多任务卸载框架，以确保可靠的卸载质量。如超高数据速率、低延迟、高可靠性等。此外，为了充分利用卫星上的可用资源，利用星间链路的快速性和可靠性来实现星间合作。它不仅增加了系统容量和覆盖范围，而且降低了单颗卫星的资源消耗，提高卫星网络的整体资源容量。基于此，卫星间的合作问题得到了广泛的讨论。尽管星间通信的研究已经取得了不错的结果，但仍然面临着一些挑战。Due to the insufficient coverage of ground networks, the integration of ground and satellite has received increasing attention in recent years. The Space-Ground Integrated Network (SAGIN) provides seamless and flexible network services for computing offloading of remote Internet of Things (IoT) devices. Unlike the ground, the former needs to consider a variety of factors such as channel conditions and satellite computing resource carrying capacity. Therefore, it is necessary to design an effective satellite IoT multi-task offloading framework to ensure reliable offloading quality. Such as ultra-high data rate, low latency, high reliability, etc. In addition, in order to make full use of the available resources on the satellite, the rapidity and reliability of the inter-satellite link are used to achieve inter-satellite cooperation. It not only increases the system capacity and coverage, but also reduces the resource consumption of a single satellite and improves the overall resource capacity of the satellite network. Based on this, the issue of cooperation between satellites has been widely discussed. Although the research on inter-satellite communication has achieved good results, it still faces some challenges.
现有的工作大多将卫星作为转发工具，然而卫星实现星载处理任务将是未来不可缺少的范式。此外，当将卫星卸载到遥远的地面云服务器时，它们之间的传输延迟是不可忽视的。而且由于任务到达和传输的动态性和随机性，卫星协同计算导致了巨大的状态和动作空间。队列稳定性的长期约束与短期卸载决策之间的耦合问题很难解决。Most existing works use satellites as forwarding tools, but satellites will be an indispensable paradigm for onboard processing tasks in the future. In addition, when offloading from satellites to distant ground cloud servers, the transmission delay between them cannot be ignored. And due to the dynamic and random nature of task arrival and transmission, satellite collaborative computing leads to a huge state and action space. The coupling problem between the long-term constraints of queue stability and the short-term offloading decision is difficult to solve.
发明内容Summary of the invention
本发明所要解决的技术问题在于针对上述现有技术中的不足，提供一种基于SAGIN的任务卸载与资源分配联合优化方法及系统，利用低轨道卫星集群网络来弥补每颗卫星容量的局限性，在保证网络稳定性和平均功率约束的条件下，以最大限度地提高网络数据处理能力，实现动态任务卸载、资源分配和关联控制，用于解决队列稳定性的长期约束与短期卸载决策之间的耦合问题的技术问题。The technical problem to be solved by the present invention is to provide a SAGIN-based joint optimization method and system for task offloading and resource allocation in response to the deficiencies in the above-mentioned prior art, which utilizes a low-orbit satellite cluster network to make up for the capacity limitations of each satellite, maximizes the network data processing capability while ensuring network stability and average power constraints, realizes dynamic task offloading, resource allocation and associated control, and solves the technical problem of the coupling problem between long-term constraints on queue stability and short-term offloading decisions.
本发明采用以下技术方案：The present invention adopts the following technical solutions:
一种基于SAGIN的任务卸载与资源分配联合优化方法，包括以下步骤：A SAGIN-based joint optimization method for task offloading and resource allocation includes the following steps:
S1、通过SAG-IoT系统收集任务及调度信息；S1. Collect task and scheduling information through the SAG-IoT system;
S2、利用步骤S1得到的任务数据及调度信息，基于匹配博弈算法实现UAV与IoT设备的最优关联，并对任务卸载做出最优决策，最大化SAG-IoT系统计算性能；S2. Using the task data and scheduling information obtained in step S1, the optimal association between UAV and IoT devices is achieved based on the matching game algorithm, and the optimal decision is made for task offloading to maximize the computing performance of the SAG-IoT system;
S3、基于步骤S2得到的最优关联、任务卸载最优决策以及最大化SAG-IoT系统计算性能，基于深度强化学习的在线卸载框架，构建从输入X(t)到最优动作x*(t)的低复杂度映射策略，基于不同模块与随机环境的重复交互，依次迭代和运行，实现任务卸载与资源分配联合优化。S3. Based on the optimal association obtained in step S2, the optimal decision for task offloading and the maximization of the computing performance of the SAG-IoT system, an online offloading framework based on deep reinforcement learning is constructed to construct a low-complexity mapping strategy from input X(t) to the optimal action x* (t). Based on the repeated interaction between different modules and the random environment, the strategies are iterated and run in sequence to achieve the joint optimization of task offloading and resource allocation.
优选地，步骤S1具体为：Preferably, step S1 specifically includes:
任务由IoT设备处理时，完成的任务量为Dil(t)，本地计算的延迟为Ti，当卸载至UAV，通过GS算法利用IoT设备和UAV的偏好列表实现多对一匹配，设UAV以固定速度v和飞行高度Hu飞行；上行链路采用正交传输，允许多个但有限的IoT设备同时卸载任务；When the task is processed by the IoT device, the amount of completed tasks is Dil (t), and the delay of local calculation is Ti . When offloaded to the UAV, the GS algorithm uses the preference lists of IoT devices and UAVs to achieve many-to-one matching. Assume that the UAV flies at a fixed speed v and flight altitude Hu ; the uplink adopts orthogonal transmission, allowing multiple but limited IoT devices to offload tasks simultaneously;
若卸载至卫星，接入卫星接收来自UAV的任务请求，并根据当前的工作负载状态进行决策，如果单颗卫星计算资源不足，通过卫星链路将任务传输给最近的CH，由CH决定计算或转交给CMs协同计算，设任务i的卸载路径为卫星网络中的总卸载延迟Tss包括传输延迟和传播延迟；If offloading to the satellite, the access satellite receives the task request from the UAV and makes a decision based on the current workload status. If the computing resources of a single satellite are insufficient, the task is transmitted to the nearest CH through the satellite link, and the CH decides whether to calculate or transfer it to the CMs for collaborative computing. Let the offloading path of task i be The total offloading delay Tss in the satellite network includes transmission delay and propagation delay;
系统总能耗包括本地计算能耗和卸载能耗虚拟能量队列由Ei(t+1)＝max(Ei(t)+ò(ei(t)-e0),0)更新，ò和E0分别定义为正比例因子和能量阈值；Total system energy consumption Including local computing energy consumption and unloading energy consumption The virtual energy queue is updated by Ei (t+1)=max(Ei (t)+ò(ei (t)−e0 ),0), where ò and E0 are defined as the positive proportionality factor and energy threshold, respectively;
当时间平均队列长度时，离散时间队列Qi(t)强稳定，其中期望相对于系统随机事件，包括信道衰落和任务到达，根据Little定律，平均延迟与平均队列长度成正比，数据队列转换为每个数据的有限处理延迟。When the average queue length The discrete-time queueQi (t) is strongly stable when , where the expectation is relative to system random events, including channel fading and task arrivals. According to Little's law, the average delay is proportional to the average queue length, and the data queue translates into a finite processing delay for each data.
优选地，卸载至UAV过程中，地面-UAV通信的路径损耗Liu为：Preferably, during the unloading process to the UAV, the path loss Liu of the ground-UAV communication is:
其中，diu表示IoTi与UAVu的距离，fu表示载波频率，vc表示光速，ι和η分别表示视距和非视距链路在自由空间路径损耗顶部产生的平均加性损耗；Where diu represents the distance between IoTi and UAVu,fu represents the carrier frequency, vc represents the speed of light, ι and η represent the average additive loss on top of the free space path loss for line-of-sight and non-line-of-sight links, respectively;
LoS和NLoS链路概率PLOS和PNLOS分别为：The LoS and NLoS link probabilities PLOS and PNLOS are:
PNLOS＝1-PLOSPNLOS = 1-PLOS
其中，hu和hiu分别表示UAVu中的飞行高度和IoT设备i与UAVu之间的水平距离，a、b、ι和η的值由环境状态决定；Where hu and hiu represent the flight altitude in UAVu and the horizontal distance between IoT device i and UAVu, respectively, and the values of a, b, ι and η are determined by the environmental state;
IoT-UAV链路的传输速率riu(t)为：The transmission rate of the IoT-UAV link riu (t) is:
其中，Biu、Piu、σ2分别表示带宽、传输功率和噪声功率；Where,Biu ,Piu ,σ2 represent bandwidth, transmission power and noise power respectively;
UAV处理卸载任务耗时Tiu为：The time taken by UAV to process the offload taskis :
其中，Ci表示所需的总CPU周期，riu(t)表示IoT-UAV传输速率，χi表示计算任务大小；WhereCi represents the total CPU cycles required,riu (t) represents the IoT-UAV transmission rate, andχi represents the computing task size;
卸载至卫星过程中，UAVu与卫星s的自由路径损耗为：During the unloading process to the satellite, the free path loss between UAVu and satellite s is:
Lus＝92.44+20log10dus+20log10fsLus =92.44+20log10dus +20log10fs
其中，dus表示UAV u与卫星s的距离，fs表示载波频率；Where, dus represents the distance between UAV u and satellite s, and fs represents the carrier frequency;
UAV-卫星链路的传输速率为：The transmission rate of the UAV-satellite link is:
其中，Bus、Pus分别为带宽、传输功率；Among them,Bus andPus are bandwidth and transmission power respectively;
链路(s,d)在时隙t的数据速率为：The data rate of link (s, d) in time slot t is:
其中，Ps为卫星发射功率，Gtr和Gre分别为发射天线增益和接收天线增益，k是玻尔兹曼常数，Ts是系统总噪声温度，Eb表示每比特接收能量与一定噪声密度所需的比值，M和SR(t)分别表示链路余量和倾斜范围。WherePs is the satellite transmit power,Gtr andGre are the transmit antenna gain and receive antenna gain respectively, k is the Boltzmann constant,Ts is the total system noise temperature,Eb represents the ratio of the received energy per bit to a certain noise density, M and SR(t) represent the link margin and tilt range respectively.
优选地，设任务i的卸载路径为卫星网络中的总卸载延迟Tss为：Preferably, let the uninstall path of task i be The total offloading delay Tss in the satellite network is:
其中，表示任务i在时隙t期间沿pi从卸载到和表示沿pi一跳的传输容量和传播延迟，K表示卫星之间距离跳数，一般来说K≤2，χi表示计算任务大小。in, It means that task imoves from Uninstall to and represents the transmission capacity and propagation delay along one hop of pi , K represents the number of hops between satellites, generally K ≤ 2, and χi represents the size of the computing task.
优选地，步骤S2中，在未来信道状况和数据到达未知情况下，提出任务卸载关联控制和资源分配的联合优化问题P0，τ为卸载时隙，E(t)为功率控制；利用Lyapunov优化将问题P0转化为P1；将P1分解为三个独立的子问题，SP1用于优化IoT设备与UAV之间的关联控制，SP2用于本地计算资源分配，SP3用于服务器资源分配，当卸载决策固定时，依次解决其他子问题。Preferably, in step S2, when the future channel conditions and data arrival are unknown, task offloading is proposed. Association Control and resource allocation The joint optimization problem P0 is proposed, where τ is the offloading time slot and E(t) is the power control. Lyapunov optimization is used to transform problem P0 into P1. P1 is decomposed into three independent sub-problems, SP1 is used to optimize the association control between IoT devices and UAVs, SP2 is used for local computing resource allocation, and SP3 is used for server resource allocation. When the offloading decision is fixed, other sub-problems are solved in turn.
优选地，优化IoT设备与UAV之间的关联控制具体为：Preferably, optimizing the association control between IoT devices and UAVs is specifically as follows:
从IoT设备的偏好列表Φ(n)中选择最受欢迎的UAV，如果选择的UAV有空缺，匹配对(n,m)将直接添加到Φ；Select the most popular UAV from the preference list Φ(n) of IoT devices. If there is a vacancy in the selected UAV, the matching pair (n,m) will be directly added to Φ;
如果Φ(m)＝＝Mm，则将IoT设备n与当前与UAV m匹配的所有其他设备比较；如果它比最差匹配的IoT设备n′更优，则n和n′将被交换；If Φ(m) == Mm , then compare IoT device n with all other devices currently matched with UAV m; if it is better than the worst matching IoT device n′, then n and n′ will be swapped;
当所有IoT设备和UAV的偏好列表为空或没有阻塞对时，实现稳定匹配。Stable matching is achieved when the preference lists of all IoT devices and UAVs are empty or have no blocking pairs.
优选地，本地计算资源分配SP2具体为：Preferably, the local computing resource allocation SP2 is specifically as follows:
其中，ai＝Qi(t)+Vωi，Qi(t)为任务队列长度，V为固定参数，ωi表示权重，fi表示IoT设备分配的CPU周期频率，φi表示计算密度，Ei(t)表示系统能耗，κ表示与硬件结构相关的有效开关电容参数，f表示载波频率，M0表示本地处理计算任务，Qi(t)表示离散时间队列，fi(t)表示IoT设备i在时隙t的CPU周期频率。Wherein, ai =Qi (t)+Vωi ,Qi (t) is the task queue length, V is a fixed parameter, ωi represents the weight,fi represents the CPU cycle frequency allocated to the IoT device, φi represents the computing density, Ei (t) represents the system energy consumption, κ represents the effective switching capacitor parameter related to the hardware structure, f represents the carrier frequency, M0 represents the local processing computing task,Qi (t) represents the discrete time queue, andfi (t) represents the CPU cycle frequency of IoT device i in time slot t.
优选地，服务器资源分配SP3具体为：Preferably, the server resource allocation SP3 is specifically as follows:
其中，ai＝Qi(t)+Vωi，Qi(t)为任务队列长度，V为固定参数，ωi表示权重，g[ρi(λ*)]表示给定λ*的最优函数，ρi表示卸载路径，，Ei(t)表示系统总能耗，hi表示信道增益，riO表示卸载服务器资源分配。Wherein, ai =Qi (t)+Vωi ,Qi (t) is the task queue length, V is a fixed parameter, ωi represents the weight, g[ρi (λ* )] represents the optimal function given λ* , ρi represents the offloading path, Ei (t) represents the total energy consumption of the system, hi represents the channel gain, andriO represents the offloading server resource allocation.
优选地，步骤S3中，低复杂度映射策略包括：Preferably, in step S3, the low complexity mapping strategy includes:
参与者模块：接受输入X(t)并输出一组候选卸载动作xi(t)；Actor module: accepts input X(t) and outputs a set of candidate offloading actions xi (t);
评论员模块：计算xi(t)并选择最佳卸载动作x(t)；Critic module: calculatesxi (t) and selects the best unloading action x(t);
策略更新模块：不断改进参与者模块的策略；Strategy update module: continuously improve the strategies of participant modules;
队列模块，在执行卸载操作后更新系统队列Queue module, updates the system queue after performing an uninstall operation
参与者模块、评论员模块、策略更新模块和队列模块通过与随机环境的重复交互，依次迭代和运行。The actor module, critic module, policy update module, and queue module iterate and run sequentially through repeated interactions with the random environment.
第二方面，本发明实施例提供了一种基于SAGIN的任务卸载与资源分配联合优化系统，包括：In a second aspect, an embodiment of the present invention provides a SAGIN-based task offloading and resource allocation joint optimization system, including:
收集模块，通过SAG-IoT系统收集任务及调度信息；The collection module collects task and scheduling information through the SAG-IoT system;
处理模块，利用任务数据及调度信息，基于匹配博弈算法实现UAV与IoT设备的最优关联，并对任务卸载做出最优决策，最大化SAG-IoT系统计算性能；The processing module uses task data and scheduling information to achieve the optimal association between UAV and IoT devices based on the matching game algorithm, and makes the best decision for task offloading to maximize the computing performance of the SAG-IoT system;
优化模块，基于最优关联、任务卸载最优决策以及最大化SAG-IoT系统计算性能，基于深度强化学习的在线卸载框架，构建从输入X(t)到最优动作x*(t)的低复杂度映射策略，基于不同模块与随机环境的重复交互，依次迭代和运行，实现任务卸载与资源分配联合优化。The optimization module, based on the optimal association, optimal decision of task offloading and maximizing the computing performance of the SAG-IoT system, constructs a low-complexity mapping strategy from input X(t) to optimal action x* (t) based on the online offloading framework based on deep reinforcement learning. Based on the repeated interaction between different modules and the random environment, it iterates and runs in sequence to achieve joint optimization of task offloading and resource allocation.
第三方面，一种计算机设备，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现上述基于SAGIN的任务卸载与资源分配联合优化方法的步骤。In a third aspect, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned SAGIN-based task offloading and resource allocation joint optimization method when executing the computer program.
第四方面，本发明实施例提供了一种计算机可读存储介质，包括计算机程序，所述计算机程序被处理器执行时实现上述基于SAGIN的任务卸载与资源分配联合优化方法的步骤。In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, including a computer program, which, when executed by a processor, implements the steps of the above-mentioned SAGIN-based task offloading and resource allocation joint optimization method.
与现有技术相比，本发明至少具有以下有益效果：Compared with the prior art, the present invention has at least the following beneficial effects:
一种基于SAGIN的任务卸载与资源分配联合优化方法，充分利用卫星上的可用资源，利用ISL的快速性和可靠性来实现星间合作。不仅增加了系统容量和覆盖范围，而且降低了单颗卫星的资源消耗，提高了卫星网络的整体资源容量。利用高容量LEO卫星和基于无人机(UAV)的SAGIN海量边缘接入进行任务卸载，设计了最优系统模型；此外，在本发明中，利用低轨道卫星集群网络来弥补每颗卫星容量的局限性。设在不知道未来通道条件和数据到达的情况下做出每个时间框架的决策；在保证网络稳定性和平均功率约束的条件下，提出一种联合在线优化算法，以最大限度地提高网络数据处理能力，实现动态任务卸载、资源分配和关联控制。A joint optimization method for task offloading and resource allocation based on SAGIN fully utilizes the available resources on the satellite and uses the rapidity and reliability of ISL to achieve inter-satellite cooperation. It not only increases the system capacity and coverage, but also reduces the resource consumption of a single satellite, and improves the overall resource capacity of the satellite network. The optimal system model is designed by using high-capacity LEO satellites and SAGIN massive edge access based on unmanned aerial vehicles (UAVs) for task offloading; in addition, in the present invention, a low-orbit satellite cluster network is used to compensate for the limitations of the capacity of each satellite. It is assumed that the decision for each time frame is made without knowing the future channel conditions and data arrival; under the condition of ensuring network stability and average power constraints, a joint online optimization algorithm is proposed to maximize the network data processing capability and realize dynamic task offloading, resource allocation and association control.
进一步的，提出一个结合Lyapunov优化和DRL优点的框架。具体的，利用Lyapunov优化将多时隙随机MINLP问题解耦为单时隙确定性问题。然后在每个时隙中，将基于模型的优化和无模型的DRL相结合，以低计算复杂度来解决确定性问题。所提出的框架不仅保证了长期队列稳定性和平均功率约束，而且获得了最优的在线计算率性能。Furthermore, a framework combining the advantages of Lyapunov optimization and DRL is proposed. Specifically, Lyapunov optimization is used to decouple the multi-slot stochastic MINLP problem into a single-slot deterministic problem. Then, in each slot, model-based optimization and model-free DRL are combined to solve the deterministic problem with low computational complexity. The proposed framework not only ensures long-term queue stability and average power constraints, but also obtains the optimal online computation rate performance.
进一步的，提出的框架采用actor-critic结构。actor模块是一个DNN框架，它根据输入环境参数(包括所有IoT设备的通道增益和队列积压)学习最佳二进制卸载动作。critic模块通过解决资源最优分配问题和关联控制来评估二进制卸载行为。与传统的行为者-批评结构相比，该方法利用模型信息获得对动作的准确评价，从而使训练过程具有更强的鲁棒性和更快的收敛性。Furthermore, the proposed framework adopts an actor-critic structure. The actor module is a DNN framework that learns the best binary offloading actions based on the input environment parameters, including channel gains and queue backlogs of all IoT devices. The critic module evaluates the binary offloading behavior by solving the resource optimal allocation problem and associated control. Compared with the traditional actor-critic structure, this method uses model information to obtain accurate evaluation of actions, making the training process more robust and convergent faster.
可以理解的是，上述第二方面的有益效果可以参见上述第一方面中的相关描述，在此不再赘述。It can be understood that the beneficial effects of the second aspect mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.
综上所述，本发明考虑时变无线信道和任务动态到达，建立空天地一体化通信模型，充分利用卫星丰富的计算资源，提供快速、灵活、广泛的通信和覆盖，在此基础上提出基于Lyapunov深度强化学习的在线任务卸载和资源分配方案；将优化问题表述为多时隙随机混合整数非线性规划问题，并利用Lyapunov优化解耦为若干确定性单时隙子问题；将基于模型的优化与无模型的DRL相结合，分别以较低的计算复杂度求解各子问题；在稳定系统队列的同时获得了最优的计算性能。In summary, the present invention considers time-varying wireless channels and dynamic arrival of tasks, establishes an integrated air-space-ground communication model, makes full use of the rich computing resources of satellites, provides fast, flexible and extensive communication and coverage, and on this basis proposes an online task offloading and resource allocation scheme based on Lyapunov deep reinforcement learning; formulates the optimization problem as a multi-slot random mixed integer nonlinear programming problem, and uses Lyapunov optimization to decouple it into several deterministic single-slot sub-problems; combines model-based optimization with model-free DRL, and solves each sub-problem with lower computational complexity; and achieves optimal computing performance while stabilizing the system queue.
下面通过附图和实施例，对本发明的技术方案做进一步的详细描述。The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
图1为天空地一体化IoT系统的总体架构图；Figure 1 is the overall architecture diagram of the sky-ground integrated IoT system;
图2是基于Lyapunov深度强化学习的在线任务卸载和资源分方案图；Figure 2 is a diagram of online task offloading and resource allocation scheme based on Lyapunov deep reinforcement learning;
图3是本发明的仿真参数图；Fig. 3 is a simulation parameter diagram of the present invention;
图4是本发明的一个仿真结果图。FIG. 4 is a simulation result diagram of the present invention.
图5是本发明的另一个仿真结果图。FIG. 5 is another simulation result diagram of the present invention.
图6为本发明一实施例提供的计算机设备的示意图；FIG6 is a schematic diagram of a computer device provided by an embodiment of the present invention;
图7为本发明根据一实施例提供的一种芯片的框图。FIG7 is a block diagram of a chip provided according to an embodiment of the present invention.
具体实施方式DETAILED DESCRIPTION
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
在本发明的描述中，需要理解的是，术语“包括”和“包含”指示所描述特征、整体、步骤、操作、元素和/或组件的存在，但并不排除一个或多个其它特征、整体、步骤、操作、元素、组件和/或其集合的存在或添加。In the description of the present invention, it should be understood that the terms “include” and “comprises” indicate the presence of described features, wholes, steps, operations, elements and/or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and/or collections thereof.
还应当理解，在本发明说明书中所使用的术语仅仅是出于描述特定实施例的目的而并不意在限制本发明。如在本发明说明书和所附权利要求书中所使用的那样，除非上下文清楚地指明其它情况，否则单数形式的“一”、“一个”及“该”意在包括复数形式。It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.
还应当进一步理解，在本发明说明书和所附权利要求书中使用的术语“和/或”是指相关联列出的项中的一个或多个的任何组合以及所有可能组合，并且包括这些组合，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，本发明中字符“/”，一般表示前后关联对象是一种“或”的关系。It should be further understood that the term "and/or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations. For example, A and/or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character "/" in the present invention generally indicates that the associated objects are in an "or" relationship.
应当理解，尽管在本发明实施例中可能采用术语第一、第二、第三等来描述预设范围等，但这些预设范围不应限于这些术语。这些术语仅用来将预设范围彼此区分开。例如，在不脱离本发明实施例范围的情况下，第一预设范围也可以被称为第二预设范围，类似地，第二预设范围也可以被称为第一预设范围。It should be understood that, although the terms first, second, third, etc. may be used to describe preset ranges, etc. in the embodiments of the present invention, these preset ranges should not be limited to these terms. These terms are only used to distinguish preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.
取决于语境，如在此所使用的词语“如果”可以被解释成为“在……时”或“当……时”或“响应于确定”或“响应于检测”。类似地，取决于语境，短语“如果确定”或“如果检测(陈述的条件或事件)”可以被解释成为“当确定时”或“响应于确定”或“当检测(陈述的条件或事件)时”或“响应于检测(陈述的条件或事件)”。The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.
在附图中示出了根据本发明公开实施例的各种结构示意图。这些图并非是按比例绘制的，其中为了清楚表达的目的，放大了某些细节，并且可能省略了某些细节。图中所示出的各种区域、层的形状及它们之间的相对大小、位置关系仅是示例性的，实际中可能由于制造公差或技术限制而有所偏差，并且本领域技术人员根据实际所需可以另外设计具有不同形状、大小、相对位置的区域/层。Various structural schematic diagrams of the embodiments disclosed in the present invention are shown in the accompanying drawings. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions/layers with different shapes, sizes, and relative positions according to actual needs.
本发明提供了一种基于SAGIN的任务卸载与资源分配联合优化方法，采用考虑时变无线信道和任务动态到达的基于Lyapunov深度强化学习的在线任务卸载和资源分配(LyDRO-TORA)方案；具体而言，首先利用Lyapunov优化将多时隙随机混合整数非线性规划(MINLP)问题解耦为若干确定性单时隙子问题。然后，将基于模型的优化与无模型的(DeepReinforcement Learning DRL)相结合，分别以较低的计算复杂度进行求解；与其他基准算法相比，本发明在稳定系统队列的同时获得了最优的计算性能。The present invention provides a joint optimization method for task offloading and resource allocation based on SAGIN, which adopts the online task offloading and resource allocation (LyDRO-TORA) scheme based on Lyapunov deep reinforcement learning considering time-varying wireless channels and dynamic arrival of tasks; specifically, the multi-slot random mixed integer nonlinear programming (MINLP) problem is first decoupled into several deterministic single-slot sub-problems using Lyapunov optimization. Then, the model-based optimization is combined with the model-free (Deep Reinforcement Learning DRL) to solve them respectively with lower computational complexity; compared with other benchmark algorithms, the present invention achieves the best computational performance while stabilizing the system queue.
其次，本发明旨在保证长期队列稳定性和平均功率约束，以及最优计算性能；提出空天地一体化架构，充分利用卫星的广覆盖性能及丰富的计算资源，实现灵活统一的资源分配；本发明提出的优化框架除了利用二进制计算卸载外，还可以扩展到由多个独立子任务组成的在线部分计算卸载策略。Secondly, the present invention aims to ensure long-term queue stability and average power constraints, as well as optimal computing performance; proposes an integrated air-space-ground architecture to fully utilize the wide coverage performance and abundant computing resources of satellites to achieve flexible and unified resource allocation; in addition to utilizing binary computing offloading, the optimization framework proposed in the present invention can also be extended to an online partial computing offloading strategy composed of multiple independent subtasks.
本发明一种基于SAGIN的任务卸载与资源分配联合优化方法，包括以下步骤：The present invention provides a joint optimization method for task offloading and resource allocation based on SAGIN, comprising the following steps:
S1、SAG-IoT系统收集任务及调度信息；S1,SAG-IoT system collects task and scheduling information;
SAG-IoT系统在时隙t做出任务卸载决策有卸载向量表示分别在IoT设备、UAV边缘服务器和LEO卫星中处理的任务。The SAG-IoT system makes a task offloading decision at time slot t have Unload Vector Represents the tasks processed in IoT devices, UAV edge servers, and LEO satellites respectively.
请参阅图1，利用离散时隙模型，其中时间周期表示为T＝1,2,...t，持续时间为τ，设该场景为准静态，即网络状态在一个时隙内保持不变，但在不同时隙内动态变化。Please refer to Figure 1. Using a discrete time slot model, where the time period is represented by T = 1, 2, ... t and the duration is τ, the scenario is assumed to be quasi-static, that is, the network state remains unchanged within a time slot, but changes dynamically in different time slots.
S101、任务由不同处理器处理时，计算模型和通信模型分别如下所示：S101. When tasks are processed by different processors, the calculation model and communication model are as follows:
1)本地计算1) Local computing
任务由IoT设备处理时，完成的任务量为：When the task is processed by the IoT device, the amount of completed task is:
其中，fi(t)为IoT设备i在时隙t的CPU周期频率，上界为fimax,φi为计算密度，即每比特所需的计算周期数。Wherefi (t) is the CPU cycle frequency of IoT device i in time slot t, its upper bound isfimax , andφi is the computationaldensity , i.e., the number of computational cycles required per bit.
本地计算的延迟为：The latency of local computation is:
其中，Ci＝xiφi表示所需的总CPU周期。Wherein,Ci =xiφi representsthe total CPU cycles required.
2)卸载到UAV2) Offloading to UAV
若卸载至UAV，利用Gale-Shapley(GS)算法利用IoT设备和UAV的偏好列表实现多对一匹配。具体来说，主动方和被动方都申请匹配，并根据优先级进行比较，只保留最合适的匹配应用，最终实现稳定匹配；完成的任务量为：If offloaded to UAV, the Gale-Shapley (GS) algorithm is used to achieve many-to-one matching using the preference lists of IoT devices and UAVs. Specifically, both the active and passive parties apply for matching, and the comparison is made based on the priority, and only the most suitable matching application is retained, and finally a stable matching is achieved; the amount of tasks completed is:
设UAV以固定速度v和飞行高度Hu飞行。上行链路采用正交传输，允许多个但有限的IoT设备同时卸载任务。Assume that the UAV flies at a fixed speed v and altitude Hu . The uplink adopts orthogonal transmission, allowing multiple but limited IoT devices to offload tasks simultaneously.
地面-UAV通信的路径损耗为：The path loss for ground-UAV communication is:
其中，diu表示IoTi与UAVu的距离，fu表示载波频率，vc表示光速，ι和η分别表示视距(LoS)和非视距(NLoS)链路在自由空间路径损耗顶部产生的平均加性损耗。where diu represents the distance between IoTi and UAVu,fu represents the carrier frequency,vc represents the speed of light, ι and η represent the average additive losses incurred by line-of-sight (LoS) and non-line-of-sight (NLoS) links on top of the free-space path loss, respectively.
LoS和NLoS链路概率分别为：The LoS and NLoS link probabilities are:
PNLOS＝1-PLOSPNLOS = 1-PLOS
其中，hu和hiu分别表示UAVu中的飞行高度和IoT设备i与UAVu之间的水平距离，a、b、ι和η的值由环境状态决定。Where hu and hiu represent the flight altitude in UAVu and the horizontal distance between IoT device i and UAVu, respectively, and the values of a, b, ι and η are determined by the environmental status.
IoT-UAV链路的传输速率为：The transmission rate of the IoT-UAV link is:
其中，Biu、Piu、σ2分别表示带宽、传输功率和噪声功率。Among them,Biu ,Piu , andσ2 represent bandwidth, transmission power, and noise power, respectively.
UAV处理卸载任务耗时：UAV processing offload task time:
3)卸载到LEO卫星3) Offload to LEO satellite
若卸载至卫星，接入卫星接收来自UAV的任务请求，并根据其当前的工作负载状态进行决策，如果单颗卫星计算资源不足，则通过卫星链路将任务传输给最近的CH，由CH决定计算或转交给CMs协同计算。If offloading to the satellite, the access satellite receives the task request from the UAV and makes a decision based on its current workload status. If the computing resources of a single satellite are insufficient, the task is transmitted to the nearest CH via the satellite link, and the CH decides whether to calculate or transfer it to the CMs for collaborative computing.
完成的任务量为：The amount of tasks completed is:
对于UAV-卫星链路，UAVu与卫星s的自由路径损耗为：For the UAV-satellite link, the free path loss between UAVu and satellite s is:
Lus＝92.44+20log10dus+20log10fsLus =92.44+20log10dus +20log10fs
其中，dus表示UAV u与卫星s的距离，fs表示载波频率。Where dus represents the distance between UAV u and satellite s, and fs represents the carrier frequency.
UAV-卫星链路的传输速率为：The transmission rate of the UAV-satellite link is:
其中，Bus、Pus分别为带宽、传输功率。Among them,Bus andPus are bandwidth and transmission power respectively.
链路(s,d)在时隙t的数据速率为：The data rate of link (s, d) in time slot t is:
其中，自由空间损失为Ps为卫星发射功率；Gtr和Gre分别为发射天线增益和接收天线增益；k是玻尔兹曼常数，Ts是系统总噪声温度，Eb＝(N0+N1)表示每比特接收能量与一定噪声密度所需的比值，M和SR(t)分别表示链路余量和倾斜范围。Among them, free space loss forPs is the satellite transmit power;Gtr andGre are the transmit antenna gain and receive antenna gain respectively; k is the Boltzmann constant,Ts is the total system noise temperature,Eb = (N0 +N1 ) represents the ratio of each bit of received energy to a certain noise density, M and SR(t) represent the link margin and tilt range respectively.
设任务i的卸载路径为卫星网络中的总卸载延迟包括传输延迟和传播延迟，表示为：Assume that the uninstall path of task i is The total offloading delay in the satellite network includes transmission delay and propagation delay, which is expressed as:
其中，表示任务i在时隙t期间沿pi从卸载到对应的和表示沿pi一跳的传输容量和传播延迟。in, It means that task imoves from Uninstall to Corresponding and represents the transmission capacity and propagation delay along one hop of pi .
S102、任务由不同处理器处理时，能耗和队列更新模型。S102, when tasks are processed by different processors, energy consumption and queue update models.
系统总能耗包括本地计算能耗和卸载能耗表示为：Total system energy consumption Including local computing energy consumption and unloading energy consumption It is expressed as:
其中，κ>0为与硬件结构相关的有效开关电容参数。Among them, κ>0 is the effective switch capacitance parameter related to the hardware structure.
设T＝1，则平均计算率和功耗分别为：Assume T = 1, then the average computing rate and power consumption are:
虚拟能量队列由Ei(t+1)＝max(Ei(t)+ò(ei(t)-e0),0)更新，ò和E0分别定义为正比例因子和能量阈值。The virtual energy queue is updated by Ei (t+1)=max(Ei (t)+ò(ei (t)−e0 ),0), where ò and E0 are defined as a positive proportionality factor and an energy threshold, respectively.
同样地，本地和卸载任务的队列长度更新分别为：Similarly, the queue lengths for local and offloaded tasks are updated as:
其中，[x]+＝max(x,0)，有Qi(1)＝0和Di(t)<Qi(1)，意味着Qi(1)≥0对时隙t成立。Among them, [x]+ = max(x,0),Qi (1) = 0 and Di (t)<Qi (1), which means thatQi (1)≥0 holds for time slot t.
当时间平均队列长度时，离散时间队列Qi(t)强稳定，其中期望相对于系统随机事件，包括信道衰落和任务到达。根据Little定律，平均延迟与平均队列长度成正比。因此，数据队列转换为每个数据的有限处理延迟。When the average queue length The discrete-time queueQi (t) is strongly stable when , where the expectation is relative to system random events, including channel fading and task arrival. According to Little's law, the average delay is proportional to the average queue length. Therefore, the data queue translates into a finite processing delay for each data.
S2、在满足长期队列稳定性和平均功率约束下，最大化系统计算性能；S2, maximize system computing performance while satisfying long-term queue stability and average power constraints;
在未来信道状况和数据到达未知情况下，提出任务卸载关联控制和资源分配的联合优化问题。其中，表示为卸载时隙，为功率控制。MINLP问题(P0)为：In the case of unknown future channel conditions and data arrival, task offloading is proposed Association Control and resource allocation The joint optimization problem of Represented as unloading time slot, is power control. The MINLP problem (P0) is:
其中，表示平均加权和速率，C1,C2分别表示设备端和服务器端的计算能力约束，C3表示卸载决策约束，C4表示卸载时间约束，C5表示关联控制，C6表示时延约束，C7表示数据队列长度约束，C8对应功率阈值约束，e0为功率阈值，C9,C10为数据稳定性和功率分配约束。in, represents the average weighted sum rate,C1 andC2 represent the computing power constraints on the device side and the server side respectively,C3 represents the offloading decision constraint,C4 represents the offloading time constraint,C5 represents the association control,C6 represents the delay constraint,C7 represents the data queue length constraint,C8 corresponds to the power threshold constraint,e0 is the power threshold,C9 andC10 are the data stability and power allocation constraints.
面对P0，利用Lyapunov drift-plus-penalty框架来保证数据队列的稳定性和平均功耗，解决长期排队延迟约束与短期决策耦合问题，将长期随机优化问题解耦为一系列短期确定性优化问题，并逐时隙求解。应用机会期望最小化技术，观察队列的积压情况Ω(t)，并最小化上式中的上界来决定联合任务卸载和资源分配。利用Lyapunov优化将问题P0转化为P1：In the face of P0, the Lyapunov drift-plus-penalty framework is used to ensure the stability and average power consumption of the data queue, solve the coupling problem between long-term queuing delay constraints and short-term decision-making, decouple the long-term stochastic optimization problem into a series of short-term deterministic optimization problems, and solve them slot by slot. Apply the opportunity expectation minimization technique, observe the queue backlog Ω(t), and minimize the upper bound in the above formula to determine the joint task offloading and resource allocation. Use Lyapunov optimization to transform problem P0 into P1:
将P1分解为三个独立的子问题，其中SP1用于优化IoT设备与UAV之间的关联控制，SP2用于本地计算资源分配，SP3用于服务器资源分配。当卸载决策固定时，依次解决其他子问题。P1 is decomposed into three independent sub-problems, where SP1 is used to optimize the association control between IoT devices and UAVs, SP2 is used for local computing resource allocation, and SP3 is used for server resource allocation. When the offloading decision is fixed, the other sub-problems are solved in turn.
S201、优化IoT设备和UAV关联控制，匹配问题(SP1)如下：S201. Optimize the IoT device and UAV association control. The matching problem (SP1) is as follows:
IoT设备和UAV关联控制匹配博弈φ被定义为两组之间的多对一映射：The IoT device and UAV association control matching game φ is defined as a many-to-one mapping between two groups:
1)IoT设备最大连接数：和1) Maximum number of IoT device connections: and
2)UAV最大连接数：和2) Maximum number of UAV connections: and
3)IoT设备与UAV相互匹配：Φ(n)＝m，当且仅当n∈Φ(m)。3) The IoT device and the UAV match each other: Φ(n) = m, if and only if n∈Φ(m).
每个IoT设备首先选择可用UAV集合中信噪比最高的信道；其中，IoT设备n与UAV mPn(t)＝Piuhiu(t)/σ2之间的信噪比定义为IoT设备n的偏好函数。为了使UAV的处理能力最大化，UAV首先选择SP1中目标函数值最优的IoT设备。Each IoT device first selects the channel with the highest signal-to-noise ratio in the available UAV set; where the signal-to-noise ratio between IoT device n and UAVmPn (t) = Piu hiu (t) / σ2 is defined as the preference function of IoT device n. In order to maximize the processing power of the UAV, the UAV first selects the IoT device with the best objective function value in SP1.
由于IoT设备在UAV覆盖范围内密集部署，UAV的偏好列表会受到其他UAV匹配结果的影响。因此，在匹配过程中，UAV和IoT设备的优先级会发生变化。为了保证匹配的稳定性，在GS算法的基础上引入了阻塞对，即：如果Φ中没有阻塞对，则认为Φ是稳定的。Since IoT devices are densely deployed within the coverage of UAVs, the preference list of UAVs will be affected by the matching results of other UAVs. Therefore, during the matching process, the priorities of UAVs and IoT devices will change. In order to ensure the stability of matching, blocking pairs are introduced on the basis of the GS algorithm, that is, if there is no blocking pair in Φ, Φ is considered stable.
在IoT设备与UAV之间稳定匹配Φ中，为Φ的阻塞对时，有以下条件：In the stable matching Φ between IoT devices and UAVs, When is a blocking pair of Φ, the following conditions apply:
1)优先列表为的IoT设备n未被服务，或优先选择UAV m而不是当前匹配的UAVΦ(n)；1) The priority list is IoT device n is not served, or prefers UAV m over the currently matched UAV Φ(n);
2)未充分利用的UAV m更偏向IoT设备n，而不是当前匹配的IoT设备集Φ(m)中的一个。2) The underutilized UAV m prefers IoT device n rather than one of the currently matched IoT device set Φ(m).
本发明提出一种基于匹配博弈的IoT设备与UAV关联控制优化算法，以寻找稳定的匹配结果，具体如下：This paper proposes an IoT device and UAV association control optimization algorithm based on matching game to find a stable matching result, which is as follows:
首先，从IoT设备的偏好列表Φ(n)中选择最受欢迎的UAV。如果选择的UAV有空缺，匹配对(n,m)将直接添加到Φ；First, the most popular UAV is selected from the preference list Φ(n) of IoT devices. If there is a vacancy in the selected UAV, the matching pair (n,m) will be directly added to Φ;
然而，如果Φ(m)＝＝Mm，则将IoT设备n与当前与UAV m匹配的所有其他设备比较；如果它比最差匹配的IoT设备n′更优，则n和n′将被交换；However, if Φ(m) == Mm , then IoT device n is compared with all other devices currently matched with UAV m; if it is better than the worst matching IoT device n′, then n and n′ are swapped;
最后，当所有IoT设备和UAV的偏好列表为空或没有阻塞对时，实现稳定匹配。Finally, stable matching is achieved when the preference lists of all IoT devices and UAVs are empty or have no blocking pairs.
S202、通过最大化SP2来确定本地计算资源分配：S202, determining local computing resource allocation by maximizing SP2:
其中，ai＝Qi(t)+Vωi，闭式最优解为：Among them, ai =Qi (t)+Vωi , the closed-form optimal solution is:
S203、通过最大化SP3来确定服务器端资源分配；S203, determining server-side resource allocation by maximizing SP3;
其中，SP3划分为SP4和SP5；Among them, SP3 is divided into SP4 and SP5;
SP4为：SP4 is:
SP4为凸优化问题，通过CVX求解。SP4 is a convex optimization problem and is solved using CVX.
SP5为：SP5 is:
SP5是一个不等式约束优化问题，利用拉格朗日乘子方法求解。SP5 is an inequality constrained optimization problem, which is solved using the Lagrange multiplier method.
构造拉格朗日函数：Construct the Lagrangian function:
其中，λ表示对偶变量，为了最小化d(λ)，对偶函数可以分解为并行子问题：Among them, λ represents the dual variable. In order to minimize d(λ), the dual function can be decomposed into parallel sub-problems:
当C11保持最优，将上式约束C12等价为0≤riO/τi≤Blog(1+Pimaxh(t)/N0)＝rimax，获得变换为：when C11 remains optimal, and the above constraint C12 is equivalent to0≤riO/τi≤Blog (1+Pimaxh (t)/N0 )=rimax ,andwe get Transformed into:
进一步等价为：Further equivalent is:
利用Lambert-W函数解决问题：Solve the problem using Lambert-W function:
其中，和是给定λ时的固定参数。W(x)表示Lambert-W函数，它是J(z)＝zexp(z)＝x,i.e.,z＝W(x)的反函数。in, and is a fixed parameter when λ is given. W(x) represents the Lambert-W function, which is the inverse function of J(z)=zexp(z)=x,ie,z=W(x).
最优解等效表示为ρi(λ)是给定的固定参数λ，表示IoT设备i的最优通信数据速率，即最优传输时间在传输速率ρi(λ)固定的情况下随riO线性增加。问题被重写为：Optimal solution The equivalent expression is ρi (λ) is a given fixed parameter λ, which represents the optimal communication data rate of IoT device i, that is, the optimal transmission time When the transmission rate ρi (λ) is fixed, it increases linearly with riO. The problem can be rewritten as:
其最优解为：The optimal solution is:
最后，将原优化问题进一步改写为：Finally, the original optimization problem is further rewritten as:
其最优解为The optimal solution is
在给定卸载决策x(t)和参数X(t)的情况下，通过优化y(t)，将G(x(t),X(t))表示为P1的最优值；求解P1等价于找到最优卸载决策x*(t)，其中Given an unloading decision x(t) and parameter X(t), G(x(t), X(t)) is expressed as the optimal value of P1 by optimizing y(t); solving P1 is equivalent to finding the optimal unloading decision x* (t), where
S3、基于DRL的在线卸载框架，构建从输入X(t)到最优动作x*(t)的低复杂度映射策略，实现任务卸载与资源分配联合优化。S3. Based on the DRL-based online offloading framework, a low-complexity mapping strategy from input X(t) to optimal action x* (t) is constructed to achieve joint optimization of task offloading and resource allocation.
低复杂度映射策略包括：Low-complexity mapping strategies include:
参与者模块：接受输入X(t)并输出一组候选卸载动作xi(t)；Actor module: accepts input X(t) and outputs a set of candidate offloading actions xi (t);
评论员模块：计算xi(t)并选择最佳卸载动作x(t)；Critic module: calculatesxi (t) and selects the best unloading action x(t);
策略更新模块：不断改进参与者模块的策略；Strategy update module: continuously improve the strategies of participant modules;
队列模块，在执行卸载操作后更新系统队列Queue module, updates the system queue after performing an uninstall operation
这四个模块通过与随机环境的重复交互，依次迭代和运行。These four modules iterate and run sequentially through repeated interactions with the random environment.
本发明再一个实施例中，提供一种基于SAGIN的任务卸载与资源分配联合优化系统，该系统能够用于实现上述基于SAGIN的任务卸载与资源分配联合优化方法，具体的，该基于SAGIN的任务卸载与资源分配联合优化系统包括收集模块、处理模块以及优化模块。In another embodiment of the present invention, a SAGIN-based task offloading and resource allocation joint optimization system is provided. The system can be used to implement the above-mentioned SAGIN-based task offloading and resource allocation joint optimization method. Specifically, the SAGIN-based task offloading and resource allocation joint optimization system includes a collection module, a processing module and an optimization module.
其中，收集模块，通过SAG-IoT系统收集任务及调度信息；Among them, the collection module collects task and scheduling information through the SAG-IoT system;
处理模块，利用任务数据及调度信息，基于匹配博弈算法实现UAV与IoT设备的最优关联，并对任务卸载做出最优决策，最大化SAG-IoT系统计算性能；The processing module uses task data and scheduling information to achieve the optimal association between UAV and IoT devices based on the matching game algorithm, and makes the best decision for task offloading to maximize the computing performance of the SAG-IoT system;
优化模块，基于最优关联、任务卸载最优决策以及最大化SAG-IoT系统计算性能，基于深度强化学习的在线卸载框架，构建从输入X(t)到最优动作x*(t)的低复杂度映射策略，基于不同模块与随机环境的重复交互，依次迭代和运行，实现任务卸载与资源分配联合优化。The optimization module, based on the optimal association, optimal decision of task offloading and maximizing the computing performance of the SAG-IoT system, constructs a low-complexity mapping strategy from input X(t) to optimal action x* (t) based on the online offloading framework based on deep reinforcement learning. Based on the repeated interaction between different modules and the random environment, it iterates and runs in sequence to achieve joint optimization of task offloading and resource allocation.
本发明再一个实施例中，提供了一种终端设备，该终端设备包括处理器以及存储器，所述存储器用于存储计算机程序，所述计算机程序包括程序指令，所述处理器用于执行所述计算机存储介质存储的程序指令。处理器可能是中央处理单元(Central ProcessingUnit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor、DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable GateArray，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等，其是终端的计算核心以及控制核心，其适于实现一条或一条以上指令，具体适于加载并执行一条或一条以上指令从而实现相应方法流程或相应功能；本发明实施例所述的处理器可以用于基于SAGIN的任务卸载与资源分配联合优化方法的操作，包括：In another embodiment of the present invention, a terminal device is provided, the terminal device includes a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding functions; the processor described in the embodiment of the present invention can be used for the operation of the joint optimization method of task offloading and resource allocation based on SAGIN, including:
通过SAG-IoT系统收集任务及调度信息；利用任务数据及调度信息，基于匹配博弈算法实现UAV与IoT设备的最优关联，并对任务卸载做出最优决策，最大化SAG-IoT系统计算性能；基于最优关联、任务卸载最优决策以及最大化SAG-IoT系统计算性能，基于深度强化学习的在线卸载框架，构建从输入X(t)到最优动作x*(t)的低复杂度映射策略，基于不同模块与随机环境的重复交互，依次迭代和运行，实现任务卸载与资源分配联合优化。The SAG-IoT system is used to collect task and scheduling information. The task data and scheduling information are used to achieve the optimal association between UAV and IoT devices based on the matching game algorithm, and the optimal decision is made for task offloading to maximize the computing performance of the SAG-IoT system. Based on the optimal association, the optimal decision for task offloading and the maximization of the computing performance of the SAG-IoT system, an online offloading framework based on deep reinforcement learning is used to construct a low-complexity mapping strategy from input X(t) to optimal action x* (t). Based on the repeated interaction between different modules and the random environment, they are iterated and run in sequence to achieve joint optimization of task offloading and resource allocation.
本发明再一个实施例中，本发明还提供了一种存储介质，具体为计算机可读存储介质(Memory)，所述计算机可读存储介质是终端设备中的记忆设备，用于存放程序和数据。可以理解的是，此处的计算机可读存储介质既可以包括终端设备中的内置存储介质，当然也可以包括终端设备所支持的扩展存储介质。计算机可读存储介质提供存储空间，该存储空间存储了终端的操作系统。并且，在该存储空间中还存放了适于被处理器加载并执行的一条或一条以上的指令，这些指令可以是一个或一个以上的计算机程序(包括程序代码)。需要说明的是，此处的计算机可读存储介质可以是高速RAM存储器，也可以是非不稳定的存储器(Non-Volatile Memory)，例如至少一个磁盘存储器。In another embodiment of the present invention, the present invention further provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory, or a non-volatile memory (Non-Volatile Memory), such as at least one disk memory.
可由处理器加载并执行计算机可读存储介质中存放的一条或一条以上指令，以实现上述实施例中有关基于SAGIN的任务卸载与资源分配联合优化方法的相应步骤；计算机可读存储介质中的一条或一条以上指令由处理器加载并执行如下步骤：The processor may load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the SAGIN-based task offloading and resource allocation joint optimization method in the above embodiment; the processor may load and execute the following steps of one or more instructions in the computer-readable storage medium:
通过SAG-IoT系统收集任务及调度信息；利用任务数据及调度信息，基于匹配博弈算法实现UAV与IoT设备的最优关联，并对任务卸载做出最优决策，最大化SAG-IoT系统计算性能；基于最优关联、任务卸载最优决策以及最大化SAG-IoT系统计算性能，基于深度强化学习的在线卸载框架，构建从输入X(t)到最优动作x*(t)的低复杂度映射策略，基于不同模块与随机环境的重复交互，依次迭代和运行，实现任务卸载与资源分配联合优化。The SAG-IoT system is used to collect task and scheduling information. The task data and scheduling information are used to achieve the optimal association between UAV and IoT devices based on the matching game algorithm, and the optimal decision is made for task offloading to maximize the computing performance of the SAG-IoT system. Based on the optimal association, the optimal decision for task offloading and the maximization of the computing performance of the SAG-IoT system, an online offloading framework based on deep reinforcement learning is used to construct a low-complexity mapping strategy from input X(t) to optimal action x* (t). Based on the repeated interaction between different modules and the random environment, they are iterated and run in sequence to achieve joint optimization of task offloading and resource allocation.
请参阅图6，终端设备为计算机设备，该实施例的计算机设备60包括：处理器61、存储器62以及存储在存储器62中并可在处理器61上运行的计算机程序63，该计算机程序63被处理器61执行时实现实施例中的储层改造井筒中流体组成计算方法，为避免重复，此处不一一赘述。或者，该计算机程序63被处理器61执行时实现实施例储层改造井筒中流体组成计算系统中各模型/单元的功能，为避免重复，此处不一一赘述。Please refer to FIG6 , the terminal device is a computer device, and the computer device 60 of this embodiment includes: a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When the computer program 63 is executed by the processor 61, the method for calculating the composition of fluid in the reservoir transformation wellbore in the embodiment is implemented. To avoid repetition, it is not described one by one here. Alternatively, when the computer program 63 is executed by the processor 61, the functions of each model/unit in the calculation system for the composition of fluid in the reservoir transformation wellbore in the embodiment are implemented. To avoid repetition, it is not described one by one here.
计算机设备60可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。计算机设备60可包括，但不仅限于，处理器61、存储器62。本领域技术人员可以理解，图6仅仅是计算机设备60的示例，并不构成对计算机设备60的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如计算机设备还可以包括输入输出设备、网络接入设备、总线等。The computer device 60 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art may understand that FIG. 6 is only an example of the computer device 60 and does not constitute a limitation on the computer device 60. The computer device 60 may include more or fewer components than shown in the figure, or may combine certain components, or different components. For example, the computer device may also include input and output devices, network access devices, buses, etc.
所称处理器61可以是中央处理单元(Central Processing Unit，CPU)，还可以是其它通用处理器、中央处理器、图形处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现场可编程门阵列(Field-Programmable Gate Array，FPGA)或者其它可编程逻辑器件、分立门或者晶体管逻辑器件、基于量子计算的数据处理逻辑器、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。The processor 61 may be a central processing unit (CPU), or other general-purpose processors, central processing units, graphics processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, data processing logic based on quantum computing, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.
存储器62可以是计算机设备60的内部存储单元，例如计算机设备60的硬盘或内存。存储器62也可以是计算机设备60的外部存储设备，例如计算机设备60上配备的插接式硬盘，智能存储卡(Smart Media Card，SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)等。The memory 62 may be an internal storage unit of the computer device 60, such as a hard disk or memory of the computer device 60. The memory 62 may also be an external storage device of the computer device 60, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the computer device 60.
进一步地，存储器62还可以既包括计算机设备60的内部存储单元也包括外部存储设备。存储器62用于存储计算机程序以及计算机设备所需的其它程序和数据。存储器62还可以用于暂时地存储已经输出或者将要输出的数据。Furthermore, the memory 62 may include both an internal storage unit of the computer device 60 and an external storage device. The memory 62 is used to store computer programs and other programs and data required by the computer device. The memory 62 may also be used to temporarily store data that has been output or is to be output.
本申请所提供的各实施例中所使用的对存储器、数据库或其它介质的任何引用，均可包括非易失性和易失性存储器中的至少一种。非易失性存储器可包括只读存储器(Read-Only Memory，ROM)、磁带、软盘、闪存、光存储器、高密度嵌入式非易失性存储器、阻变存储器(ReRAM)、磁变存储器(Magnetoresistive Random Access Memory，MRAM)、铁电存储器(Ferroelectric Random Access Memory，FRAM)、相变存储器(Phase Change Memory，PCM)、石墨烯存储器等。易失性存储器可包括随机存取存储器(Random Access Memory，RAM)或外部高速缓冲存储器等。作为说明而非局限，RAM可以是多种形式，比如静态随机存取存储器(Static Random Access Memory，SRAM)或动态随机存取存储器(Dynamic RandomAccess Memory，DRAM)等。Any reference to memory, database or other media used in the embodiments provided in the present application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.
本申请所提供的各实施例中所涉及的数据库可包括关系型数据库和非关系型数据库中至少一种。非关系型数据库可包括基于区块链的分布式数据库等，不限于此。本申请所提供的各实施例中所涉及的处理器可为通用处理器、中央处理器、图形处理器、数字信号处理器、可编程逻辑器、基于量子计算的数据处理逻辑器等，不限于此。The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.
请参阅图7，终端设备为芯片，该实施例的芯片600包括处理器622，其数量可以为一个或多个，以及存储器632，用于存储可由处理器622执行的计算机程序。存储器632中存储的计算机程序可以包括一个或一个以上的每一个对应于一组指令的模块。此外，处理器622可以被配置为执行该计算机程序，以执行上述的可泛化通用单目绝对深度图估计方法。Referring to FIG. 7 , the terminal device is a chip, and the chip 600 of this embodiment includes a processor 622, which may be one or more, and a memory 632 for storing a computer program executable by the processor 622. The computer program stored in the memory 632 may include one or more modules, each corresponding to a set of instructions. In addition, the processor 622 may be configured to execute the computer program to execute the above-mentioned generalizable monocular absolute depth map estimation method.
另外，芯片600还可以包括电源组件626和通信组件650，该电源组件626可以被配置为执行芯片600的电源管理，该通信组件650可以被配置为实现芯片600的通信，例如，有线或无线通信。此外，该芯片600还可以包括输入/输出接口658。芯片600可以操作基于存储在存储器632的操作系统。In addition, the chip 600 may further include a power supply component 626 and a communication component 650, wherein the power supply component 626 may be configured to perform power management of the chip 600, and the communication component 650 may be configured to implement communication, for example, wired or wireless communication, of the chip 600. In addition, the chip 600 may further include an input/output interface 658. The chip 600 may operate based on an operating system stored in the memory 632.
为使本发明实施例的目的、技术方案和优点更加清楚，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。通常在此处附图中的描述和所示的本发明实施例的组件可以通过各种不同的配置来布置和设计。因此，以下对在附图中提供的本发明的实施例的详细描述并非旨在限制要求保护的本发明的范围，而是仅仅表示本发明的选定实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention described and shown in the drawings here can usually be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
仿真验证Simulation Verification
为了模拟和评估提出的LyDRO-TORA算法，本发明采用如下的仿真设置：In order to simulate and evaluate the proposed LyDRO-TORA algorithm, the present invention adopts the following simulation settings:
设所有的计算都在TensorFlow 2.0平台进行评估，该平台拥有Intel Core i5-4570 3.2GHz CPU和12gb内存；Assume that all computations are evaluated on the TensorFlow 2.0 platform, which has an Intel Core i5-4570 3.2GHz CPU and 12gb of memory;
所有IoT设备的任务数据到达遵循平均速率相等的指数分布E[Ai(t)＝λi]。The arrival of task data from all IoT devices follows an exponential distribution E[Ai (t)=λi ] with the same average rate.
请参阅图3，仿真参数设置如下：Please refer to Figure 3, the simulation parameters are set as follows:
16颗高度780千米的LEO卫星组成4×4walker星座。16 LEO satellites at an altitude of 780 kilometers form a 4×4 walker constellation.
4个高度为10km的UAVs均匀分布在100km×100km的区域内。Four UAVs with an altitude of 10 km are evenly distributed in an area of 100 km × 100 km.
数据量为[10Mbit，200Mbit]的地面用户随机分布在UAVs覆盖的区域内。Ground users with data volumes of [10Mbit, 200Mbit] are randomly distributed in the area covered by UAVs.
在这种情况下，低轨道卫星的周期为100分钟。UAVs与LEO卫星的连接以及LEO卫星的网络拓扑由卫星工具包(STK)导出。In this case, the period of the LEO satellite is 100 minutes. The connectivity of UAVs to LEO satellites and the network topology of LEO satellites are derived from the Satellite Tool Kit (STK).
本发明提出的LyDRO-TORA框架在Actor模块中采用全连接多层感知器，由1个输入层、2个隐藏层和1个输出层组成，其中第一层和第二层隐藏层分别有120和80个隐藏神经元。The LyDRO-TORA framework proposed in the present invention adopts a fully connected multilayer perceptron in the Actor module, which consists of 1 input layer, 2 hidden layers and 1 output layer, where the first and second hidden layers have 120 and 80 hidden neurons, respectively.
请参阅图4和图5，考虑10,000个时间框架，数据到达率Mbps，绘制随时间的平均功耗性能。分析LyDRO-TORA算法的网络稳定性，揭示了数据处理能力与网络稳定性之间的权衡关系。Refer to Figures 4 and 5, considering 10,000 time frames, data arrival rate Mbps, and plotting the average power consumption performance over time. Analyzing the network stability of the LyDRO-TORA algorithm reveals the trade-off between data processing capacity and network stability.
仿真表明本发明方法具有更高的计算能力，并且所提出的算法在所有时间框架内保持较低的数据队列长度；随着数据到达率的增加，三种方案的平均能耗稳步上升至约束值。与比例分配方法相比，本发明方法实现了更大的稳定容量区域，因此在高负荷和严格的功率约束下具有更强的鲁棒性。The simulation shows that the proposed method has higher computing power and the proposed algorithm maintains a lower data queue length in all time frames; as the data arrival rate increases, the average energy consumption of the three schemes steadily increases to the constraint value. Compared with the proportional allocation method, the proposed method achieves a larger stable capacity area and is therefore more robust under high load and strict power constraints.
综上所述，本发明一种基于SAGIN的任务卸载与资源分配联合优化方法及系统，为偏远地区的IoT设备提供无处不在的通信和计算服务。在SAG-IoT中，针对任务卸载与资源分配的联合优化面临着大规模状态空间、时变网络场景、排队延迟的长期约束与短期决策之间的耦合等挑战。提出一种考虑时变无线信道和任务动态到达的基于Lyapunov深度强化学习的在线任务卸载和资源分配方案。具体而言，该方案首先利用Lyapunov优化将多时隙随机MINLP问题解耦为若干确定性单时隙子问题。然后，将基于模型的优化与无模型的DRL相结合，分别以较低的计算复杂度进行求解。优化框架由四个模块组成：参与者模块：接受输入X(t)并输出一组候选卸载动作xi(t)；评论员模块：计算xi(t)并选择最佳卸载动作x(t)；策略更新模块：不断改进参与者模块的策略，队列模块在执行卸载操作后更新系统队列这四个模块通过与随机环境的重复交互，依次迭代和运行。仿真结果表明，本发明方案在稳定系统队列的同时获得了最优的计算性能。本发明提出的优化框架除了利用二进制计算卸载外，还可以扩展到由多个独立子任务组成的在线部分计算卸载策略。In summary, the present invention provides a joint optimization method and system for task offloading and resource allocation based on SAGIN, which provides ubiquitous communication and computing services for IoT devices in remote areas. In SAG-IoT, the joint optimization of task offloading and resource allocation faces challenges such as large-scale state space, time-varying network scenarios, coupling between long-term constraints of queuing delays and short-term decisions. An online task offloading and resource allocation scheme based on Lyapunov deep reinforcement learning considering time-varying wireless channels and dynamic arrival of tasks is proposed. Specifically, the scheme first uses Lyapunov optimization to decouple the multi-slot random MINLP problem into several deterministic single-slot sub-problems. Then, the model-based optimization is combined with the model-free DRL to solve them with lower computational complexity. The optimization framework consists of four modules: the participant module: accepts the input X(t) and outputs a set of candidate offloading actions xi (t); the commentator module: calculates xi (t) and selects the best offloading action x(t); the policy update module: continuously improves the strategy of the participant module, and the queue module updates the system queue after performing the offloading operation. These four modules iterate and run in sequence through repeated interactions with the random environment. The simulation results show that the scheme of the present invention achieves the best computing performance while stabilizing the system queue. In addition to utilizing binary computation offloading, the optimization framework proposed in the present invention can also be extended to an online partial computation offloading strategy consisting of multiple independent subtasks.
所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，仅以上述各功能单元、模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能单元、模块完成，即将所述装置的内部结构划分成不同的功能单元或模块，以完成以上描述的全部或者部分功能。实施例中的各功能单元、模块可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中，上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。另外，各功能单元、模块的具体名称也只是为了便于相互区分，并不用于限制本申请的保护范围。上述系统中单元、模块的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述或记载的部分，可以参见其它实施例的相关描述。In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
本领域普通技术人员可以意识到，结合本发明中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
在本发明所提供的实施例中，应该理解到，所揭露的装置/终端和方法，可以通过其它的方式实现。例如，以上所描述的装置/终端实施例仅仅是示意性的，例如，所述模块或单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通讯连接可以是通过一些接口，装置或单元的间接耦合或通讯连接，可以是电性，机械或其它的形式。In the embodiments provided by the present invention, it should be understood that the disclosed devices/terminals and methods can be implemented in other ways. For example, the device/terminal embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
所述集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明实现上述实施例方法中的全部或部分流程，也可以通过计算机程序来指令相关的硬件来完成，所述的计算机程序可存储于一计算机可读存储介质中，该计算机程序在被处理器执行时，可实现上述各个方法实施例的步骤。其中，所述计算机程序包括计算机程序代码，所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读介质可以包括：能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(Read-Only Memory，ROM)、随机存取存储器(RandomAccess Memory，RAM)、电载波信号、电信信号以及软件分发介质等，需要说明的是，所述计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减，例如在某些司法管辖区，根据立法和专利实践，计算机可读介质不包括是电载波信号和电信信号。If the integrated module/unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.
本申请是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。The present application is described with reference to the flowchart and/or block diagram of the method, device (system) and computer program product according to the embodiment of the present application. It should be understood that each process and/or box in the flowchart and/or block diagram, and the combination of the process and/or box in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for realizing the function specified in one process or multiple processes in the flowchart and/or one box or multiple boxes in the block diagram.
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.
以上内容仅为说明本发明的技术思想，不能以此限定本发明的保护范围，凡是按照本发明提出的技术思想，在技术方案基础上所做的任何改动，均落入本发明权利要求书的保护范围之内。The above contents are only for explaining the technical idea of the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410894405.8ACN118900462A (en) | 2024-07-04 | 2024-07-04 | A joint optimization method and system for task offloading and resource allocation based on SAGIN |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410894405.8ACN118900462A (en) | 2024-07-04 | 2024-07-04 | A joint optimization method and system for task offloading and resource allocation based on SAGIN |
| Publication Number | Publication Date |
|---|---|
| CN118900462Atrue CN118900462A (en) | 2024-11-05 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410894405.8APendingCN118900462A (en) | 2024-07-04 | 2024-07-04 | A joint optimization method and system for task offloading and resource allocation based on SAGIN |
| Country | Link |
|---|---|
| CN (1) | CN118900462A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119255301A (en)* | 2024-12-05 | 2025-01-03 | 南京信息工程大学 | A computational task offloading optimization method considering multiple performance indicators |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119255301A (en)* | 2024-12-05 | 2025-01-03 | 南京信息工程大学 | A computational task offloading optimization method considering multiple performance indicators |
| Publication | Publication Date | Title |
|---|---|---|
| CN113543176B (en) | Unloading decision method of mobile edge computing system based on intelligent reflecting surface assistance | |
| CN118250750B (en) | Satellite edge computing task offloading and resource allocation method based on deep reinforcement learning | |
| US11831708B2 (en) | Distributed computation offloading method based on computation-network collaboration in stochastic network | |
| CN112416554B (en) | Task migration method and device, electronic equipment and storage medium | |
| CN112118287B (en) | Optimal Scheduling Decision Method for Network Resources Based on Alternating Direction Multiplier Algorithm and Mobile Edge Computing | |
| CN114741191B (en) | A multi-resource allocation method for computationally intensive task correlation | |
| CN116489708B (en) | Meta universe oriented cloud edge end collaborative mobile edge computing task unloading method | |
| CN114340016A (en) | A method and system for offloading and distributing power grid edge computing | |
| CN114007231B (en) | Heterogeneous unmanned aerial vehicle data unloading method and device, electronic equipment and storage medium | |
| CN114785692B (en) | A virtual power plant aggregation control communication network traffic balancing method and device | |
| CN117200870A (en) | Online data unloading method and system for space-air-ground integrated network | |
| CN118900462A (en) | A joint optimization method and system for task offloading and resource allocation based on SAGIN | |
| Hu et al. | Dynamic task offloading in MEC-enabled IoT networks: A hybrid DDPG-D3QN approach | |
| CN114205353A (en) | Calculation unloading method based on hybrid action space reinforcement learning algorithm | |
| CN116541106A (en) | Calculation task offloading method, computing device and storage medium | |
| Tyagi et al. | GM-WOA: a hybrid energy efficient cluster routing technique for SDN-enabled WSNs | |
| CN117669741A (en) | Dynamic collaborative reasoning method for UAV cluster size model based on genetic algorithm | |
| Wang | Collaborative task offloading strategy of UAV cluster using improved genetic algorithm in mobile edge computing | |
| Chen et al. | Optimal sequential relay-remote selection and computation offloading in mobile edge computing | |
| CN118265084A (en) | A particle swarm-based end-to-end satellite edge computing task offloading method | |
| CN113573363A (en) | MEC calculation unloading and resource allocation method based on deep reinforcement learning | |
| CN115696587A (en) | Resource allocation method, device, equipment and medium | |
| CN115955479A (en) | Task rapid scheduling and resource management method in cloud edge cooperation system | |
| CN114980160A (en) | Unmanned aerial vehicle-assisted terahertz communication network joint optimization method and device | |
| CN118972958A (en) | D2D communication resource allocation method and device for power emergency communication |
| 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 |