技术领域technical field
本发明属于生命安全、生物医学工程以及可穿戴设备领域，具体涉及一种基于呼吸信号的自动报警求助设备及其检测方法。The invention belongs to the fields of life safety, biomedical engineering and wearable devices, and in particular relates to an automatic alarm and help-seeking device based on a breathing signal and a detection method thereof.
技术背景technical background
随着社会的进步人们所处的治安环境状况越来越好，但是一些恶劣事件还是偶有发生，例如儿童被拐、夜跑女性被劫杀等。如果在这些事件中被害人可以及时地、隐蔽地报警，那么很多恶性事件都可以被及时地中止。With the progress of society, the public security environment in which people live is getting better and better, but some bad incidents still happen occasionally, such as children being abducted, women who run at night are robbed and killed, etc. If the victim can timely and covertly call the police in these events, many vicious events can be stopped in time.
虽然说市面上已经有一些能够起到紧急报警作用的设备。例如，名称为“一种紧急报警器”的实用新型专利(授权公告号CN 204440608 U)。当用户遇到危险时可以通过其上面的按键发送紧急呼救信号。又如，名称为“一种SOS按键结构及其老人手机”的实用新型专利(授权公告号CN 202617219 U)。当老年人遇到危险时可以通过其上面的紧急按键进行发送SOS信号。但是这些报警设备都需要用户去按某个或者某些特殊按键。这种方式既不够隐蔽，当遇到歹徒时做出这些特殊动作有可能会激怒歹徒使得事件更加恶化；这种方式又不够安全有效，当被害人处于被绑等严重限制自身活动的状况下，很难有效的触发这些设备的报警功能，并且由于用户在移动状态下，可能会触碰到，从而出现误报的情况。Although there are already some devices on the market that can play an emergency alarm role. For example, a utility model patent titled "an emergency alarm" (authorized announcement number CN 204440608 U). When the user is in danger, the emergency call signal can be sent through the buttons on it. Another example is the utility model patent (authorized announcement number CN 202617219 U) titled "A SOS Button Structure and Its Mobile Phone for the Elderly". When the elderly are in danger, they can send SOS signals through the emergency button on it. But these alarm devices all need the user to press some or some special keys. This method is not concealed enough. When encountering criminals, making these special actions may irritate the criminals and make the incident worse; this method is not safe and effective enough. It is difficult to effectively trigger the alarm function of these devices, and because the user may touch it while moving, false alarms may occur.
因此开发一种隐蔽、不误报和安全有效的紧急求助和报警的设备和方法具有重要的意义。Therefore it is of great significance to develop a concealed, non-false alarm and safe and effective emergency help and alarm equipment and method.
发明内容Contents of the invention
本发明的目的是为了解决上述技术问题，提供了一种基于呼吸信号的自动报警求助设备及其检测方法。图1、图2示出了本发明的设备及其检测方法示意图。The purpose of the present invention is to solve the above-mentioned technical problems, and to provide an automatic alarm and help-seeking device based on a breathing signal and a detection method thereof. Fig. 1 and Fig. 2 show schematic diagrams of the device and its detection method of the present invention.
一种基于呼吸信号的自动报警求助设备，包括电源模块、呼吸信号采集模块、马达微振动模块、GPS模块、无线蓝牙模块、GSM模块和MCU中央处理模块；An automatic alarm and help-seeking device based on respiratory signals, including a power supply module, a respiratory signal acquisition module, a motor micro-vibration module, a GPS module, a wireless Bluetooth module, a GSM module and an MCU central processing module;
所述电源模块包含锂电池、电压转换电路和充放电管理保护电路；为整个设备提供电源保护功能和所需的各种工作电压；The power module includes a lithium battery, a voltage conversion circuit and a charge-discharge management protection circuit; it provides power protection functions and various required working voltages for the entire device;
锂电池为整个设备提供能源；充放电管理保护电路用来管理和保护锂电池充放电过程，防止过压、过流和欠压现象损坏锂电池工作性能；电压转换电路用来把锂电池提供的电压转换成工作电压以供给其他的模块使用；The lithium battery provides energy for the entire device; the charge and discharge management protection circuit is used to manage and protect the charging and discharging process of the lithium battery, preventing overvoltage, overcurrent and undervoltage from damaging the performance of the lithium battery; the voltage conversion circuit is used to convert the lithium battery provided The voltage is converted into a working voltage for use by other modules;
所述呼吸信号采集模块用于采集用户的呼吸信号并把呼吸信号数据输送至MCU中央处理模块；The respiratory signal acquisition module is used to collect the user's respiratory signal and transmit the respiratory signal data to the MCU central processing module;
所述马达微振动模块给予用户触觉反馈的作用；当设备检测到用户做出设定的特定呼吸状态时或设备成功报警时，由MCU中央处理模块控制马达微振动模块进行一连串的振动从而提醒用户设备已经检测到或完成用户的要求；The motor micro-vibration module gives the user the function of tactile feedback; when the device detects the specific breathing state set by the user or the device successfully alarms, the MCU central processing module controls the motor micro-vibration module to perform a series of vibrations to remind the user The device has detected or completed the user's request;
所述GPS模块为设备提供用户的地理位置；当设备检测到用户做出特定呼吸状态时，MCU中央处理模块打开GPS模块，GPS模块实时监测用户的地理位置，并将其传给MCU中央处理模块；The GPS module provides the user's geographic location for the device; when the device detects that the user has made a specific breathing state, the MCU central processing module turns on the GPS module, and the GPS module monitors the user's geographic location in real time and transmits it to the MCU central processing module ;
所述无线蓝牙模块用于用户通过外部的智能手机或者电脑与MCU中央处理模块进行通信；The wireless bluetooth module is used for the user to communicate with the MCU central processing module through an external smart phone or computer;
所述GSM模块插有电话卡用于拨打110、亲友号码和发送MCU中央处理模块提供的实时地理位置信息，并在拨打电话完成后反馈一个信息至MCU中央处理模块；Described GSM module is inserted with telephone card and is used for dialing 110, the number of relatives and friends and sending the real-time geographic location information that MCU central processing module provides, and feeds back a message to MCU central processing module after making a call;
所述MCU中央处理模块控制除开电源模块外的各个模块，通过无线蓝牙模块与手机或电脑通信并使其进行特定呼吸状态和亲友号码的设定，并且对接收到的呼吸信号数据进行处理、分析，并确认呼吸信号是否为设定的特定呼吸状态；是则通过GSM模块拨打求助电话，并同时打开GPS模块，实时地监测用户的地理位置，并通过GSM模块持续发出；求助完成后MCU中央处理模块接收GSM模块的反馈信息，然后控制马达微振动模块给予用户触觉反馈；否则继续监测呼吸信号。The MCU central processing module controls each module except the power supply module, communicates with the mobile phone or computer through the wireless bluetooth module and makes it carry out the setting of the specific breathing state and the number of relatives and friends, and processes and analyzes the received breathing signal data , and confirm whether the breathing signal is in the set specific breathing state; if yes, make a call for help through the GSM module, and turn on the GPS module at the same time, monitor the user's geographical location in real time, and send it continuously through the GSM module; after the help is completed, the MCU central processing The module receives the feedback information from the GSM module, and then controls the motor micro-vibration module to give the user tactile feedback; otherwise, continue to monitor the breathing signal.
所述特定呼吸状态为n≥1个呼吸状态进行排列组合；呼吸状态为：正常呼吸、急促呼吸或屏息，每个呼吸状态中均包括次数和时间，且通过事先设定。The specific breathing state is arranged and combined with n≥1 breathing states; the breathing states are: normal breathing, rapid breathing or breath holding, and each breathing state includes the frequency and time, and is set in advance.
所述电压转换电路为+3.3V&+1.8V生成电路。The voltage converting circuit is a +3.3V&+1.8V generating circuit.
所述呼吸信号采集模块包括呼吸信号传感器、高频电流信号产生电路、电压-电阻转换电路以及模/数转换电路；高频电流信号产生电路通过呼吸信号传感器向人体注入高频电流信号，同时呼吸信号传感器测量人体的电压，通过电压-电阻转换电路将其转换成胸腔阻抗值，模/数转换电路将胸腔阻抗值转换成数字信号输送至MCU中央处理模块。The respiratory signal acquisition module includes a respiratory signal sensor, a high-frequency current signal generating circuit, a voltage-resistance conversion circuit, and an analog/digital conversion circuit; the high-frequency current signal generating circuit injects a high-frequency current signal into the human body through the respiratory signal sensor, and breathes simultaneously. The signal sensor measures the voltage of the human body, and converts it into a chest cavity impedance value through a voltage-resistance conversion circuit, and the analog/digital conversion circuit converts the chest cavity impedance value into a digital signal and sends it to the MCU central processing module.
上述基于呼吸信号的自动报警求助设备的检测方法，包括以下步骤：The detection method of the above-mentioned automatic alarm and help-seeking equipment based on the breathing signal comprises the following steps:
步骤1、基于先验信息或快速傅里叶变换FFT明确需要检测的呼吸信号当中是否含有直流成分；Step 1. Based on prior information or fast Fourier transform FFT, it is determined whether the respiratory signal to be detected contains a DC component;
所述先验信息是指根据呼吸信号测量方法来明确所需检测的呼吸信号是否带有直流成分；The prior information refers to determining whether the respiratory signal to be detected has a DC component according to the respiratory signal measurement method;
所述快速傅里叶变换是指对信号进行时域到频域的转换然后分析0Hz成分的大小从而明确所需检测的呼吸信号是否带有直流成分；The fast Fourier transform refers to converting the signal from the time domain to the frequency domain and then analyzing the size of the 0Hz component so as to clarify whether the respiratory signal to be detected has a DC component;
步骤2、通过阈值算法或特征算法把需要检测的呼吸信号转换成只含有-1,1两种值的方波信号；对于不含有直流成分的呼吸信号首选阈值算法作为转换方法，对于含有直流成分的呼吸信号首选特征算法作为转换方法；Step 2. Convert the breathing signal to be detected into a square wave signal containing only two values of -1 and 1 through a threshold algorithm or a feature algorithm; the threshold algorithm is the preferred conversion method for breathing signals that do not contain DC components, and the conversion method for breathing signals that contain DC components The preferred feature algorithm of the respiratory signal is used as the conversion method;
所述阈值算法是指：将所需检测的呼吸信号中的每点按照特殊映射f1(n)进行阈值比较从而得到一组与输入数据等长的只含有-1,1两种值的方波信号；如果某点数值大于X值，f1(n)设为1；如果某点数值小于-X值，f1(n)设为-1；如果某点数值处于-X和X之间时，f1(n)保持f1(n-1)的值；The threshold algorithm refers to: compare each point in the respiratory signal to be detected according to the special map f1 (n) to obtain a set of squares with the same length as the input data and only contain two values -1 and 1. Wave signal; if the value of a certain point is greater than X value, f1 (n) is set to 1; if the value of a certain point is less than -X value, f1 (n) is set to -1; if the value of a certain point is between -X and X When , f1 (n) keeps the value of f1 (n-1);
其中，n为所需检测的呼吸信号中某点；x(n)为所需检测的呼吸信号中n点的值；f1(n-1)为n-1点的方波信号值；X值大小取决于先验信息、数值可调；f1(n)为本步骤输出的方波信号；Wherein, n is a certain point in the respiratory signal to be detected; x(n) is the value of n points in the respiratory signal to be detected; f1 (n-1) is the square wave signal value of n-1 point; X The value depends on the prior information, and the value is adjustable; f1 (n) is the square wave signal output by this step;
所述特征算法是指：首先，寻找所需检测的呼吸信号中的极大值和极小值；具体如下：The characteristic algorithm refers to: first, find the maximum value and the minimum value in the respiratory signal to be detected; specifically as follows:
①将所需检测的呼吸信号中的每点与其下一点做差，得一组差值数据diff1(n)；① Make a difference between each point in the respiratory signal to be detected and the next point to obtain a set of difference data diff1 (n);
diff1(n)＝x(n+1)-x(n)(n＝0，1，...)diff1 (n)=x(n+1) -x(n) (n=0,1,...)
其中，n为所需检测的呼吸信号中某点；n+1为所需检测的呼吸信号中n点的下一点；x(n)为所需检测的呼吸信号中n点的值；x(n+1)为所需检测的呼吸信号中n+1点的值；diff1(n)为输出的差值数据；Wherein, n is a certain point in the respiratory signal to be detected; n+1 is the next point of n points in the respiratory signal to be detected; x(n) is the value of n points in the respiratory signal to be detected; x( n+1) is the value of n+1 points in the respiratory signal to be detected; diff1 (n) is the difference data of output;
②寻找diff1(n)数据中符合diff1(n)≥0且diff1(n-1)≤0或diff1(n)≤0且diff1(n-1)≥0条件的点；这些点即为所需找的呼吸信号中的极大值和极小值；② Find the points in the diff1 (n) data that meet the conditions of diff1 (n) ≥ 0 and diff1 (n-1) ≤ 0 or diff1 (n) ≤ 0 and diff1 (n-1) ≥ 0; these The points are the maximum and minimum values in the respiratory signal to be found;
然后，对信号中的所有极值点进行分析；具体过程如下：Then, analyze all extreme points in the signal; the specific process is as follows:
①将每个极点与其下一个极点做差，得一组差值数据diff2(m)；① Make a difference between each pole and the next pole to obtain a set of difference data diff2 (m);
diff2(m)＝x(m+1)-x(m)diff2 (m)＝x(m+1) -x(m)
其中，m为呼吸信号中某极值点坐标；m+1为m点后下一个极值点坐标；x(m)为所需检测的呼吸信号中极值点m对应的值；x(m+1)为所需检测的呼吸信号中极值点m+1对应的值；diff2(m)为输出的差值数据；Among them, m is the coordinates of an extreme point in the respiratory signal; m+1 is the coordinate of the next extreme point after point m; x(m) is the value corresponding to the extreme point m in the respiratory signal to be detected; x(m +1) is the value corresponding to the extremum point m+1 in the respiratory signal to be detected; diff2 (m) is the output difference data;
②对diff2(m)数据进行分析，将所需检测的呼吸信号中的每点按照特殊映射f2(n)进行转换从而得到一组与输入数据等长的只含有-1,1两种值的方波信号；如果diff2(m)值大于D且极值点m和极值点m+1的时间间隔大于T,则把极值点m和极值点m+1之内的所有f2(n)设为1；如果diff2(m)值小于-D且极值点m和极值点m+1的时间间隔大于T,则把极值点m和极值点m+1之内的所有f2(n)设为-1；其余的f2(n)保持上一个f2(n-1)的值；②Analyze the diff2 (m) data, convert each point in the respiratory signal to be detected according to the special mapping f2 (n) to obtain a set of data with the same length as the input data containing only -1, 1 value of the square wave signal; if the value of diff2 (m) is greater than D and the time interval between extreme point m and extreme point m+1 is greater than T, then all the values within extreme point m and extreme point m+1 f2 (n) is set to 1; if the value of diff2 (m) is less than -D and the time interval between extreme point m and extreme point m+1 is greater than T, then extreme point m and extreme point m+1 All f2 (n) within are set to -1; the remaining f2 (n) keep the value of the previous f2 (n-1);
其中，n为所需检测的呼吸信号中某点，m为呼吸信号中某极值点坐标，m+1为m点后下一个极值点坐标，diff2(m)为极值点m对应的差值数据，tm为极值点m对应的时刻，tm+1为极值点m+1对应的时刻；D值大小取决于先验信息、数值可调；0.0＜T≤1.5秒，取决于先验信息；f2(n)为本步骤输出的方波信号；Among them, n is a certain point in the respiratory signal to be detected, m is the coordinate of an extreme point in the respiratory signal, m+1 is the coordinate of the next extreme point after point m, and diff2 (m) is the corresponding extreme point m tm is the time corresponding to the extreme point m, and tm+1 is the time corresponding to the extreme point m+1; the value of D depends on the prior information and the value is adjustable; 0.0<T≤1.5 seconds , depends on prior information; f2 (n) is the square wave signal output by this step;
步骤3、通过跳变点提取算法寻找出步骤2产生的方波信号中的跳变点并把这些点构成跳变点序列；Step 3, find out the jump points in the square wave signal that step 2 produces by the jump point extraction algorithm and form these points into a jump point sequence;
所述的跳变点提取算法是指：Described jump point extracting algorithm refers to:
①将步骤2产生的方波信号中每点与其下一点做差，得一组差值信号diff3(n)① Make a difference between each point in the square wave signal generated in step 2 and the next point to obtain a set of difference signals diff3 (n)
diff3(n)＝y(n+1)-y(n)(n＝0，1，...)diff3 (n) = y(n+1) -y(n) (n = 0, 1, ...)
其中，n为所需检测的方波信号中某点，y(n)为所需检测的方波信号中n点的值，y(n+1)为所需检测的方波信号中n+1点的值，diff3(n)为输出的差值数据；Among them, n is a certain point in the square wave signal to be detected, y(n) is the value of n points in the square wave signal to be detected, and y(n+1) is n+ in the square wave signal to be detected The value of 1 point, diff3 (n) is the output difference data;
②对diff3(n)数据进行分析；当diff3(n)不为0时，则代表方波信号在n点处发生跳变，即n点和n+1点为所要寻找的跳变点从而加入跳变点序列；当diff3(n)为0时，则代表方波信号在n点处没有发生跳变，即n点和n+1点不为所要寻找的跳变点从而不加入跳变点序列；② Analyze the data of diff3 (n); when diff3 (n) is not 0, it means that the square wave signal jumps at point n, that is, point n and point n+1 are the jump points to be found Thus adding the jump point sequence; when diff3 (n) is 0, it means that the square wave signal does not jump at point n, that is, point n and n+1 are not the jump points to be found, so do not add jump point sequence;
步骤4、计算步骤3中跳变点序列中相邻两点之间的时间间隔，然后去除那些间隔为一个采样周期的间隔，将剩余的时间间隔按序构成时间间隔序列；Step 4, calculate the time interval between two adjacent points in the jump point sequence in step 3, then remove those intervals that are intervals of a sampling period, and form the time interval sequence with the remaining time intervals in order;
步骤5、将步骤4中所得的时间间隔序列进行特殊映射f3(n)得到一组由0,1,2三种值构成的序列；如果步骤4中所得的时间间隔序列中某点对应的时间间隔小于T0,则把f3(n)设为1；如果步骤4中所得的时间间隔序列中某点对应的时间间隔大于T1,则把f3(n)设为2；其余f3(n)设为0：Step 5. Perform special mapping f3 (n) on the time interval sequence obtained in step 4 to obtain a sequence consisting of three values of 0, 1, and 2; if a certain point in the time interval sequence obtained in step 4 corresponds to If the time interval is less than T0 , set f3 (n) to 1; if the time interval corresponding to a point in the time interval sequence obtained in step 4 is greater than T1 , then set f3 (n) to 2; the rest f3 (n) is set to 0:
其中，n为步骤4中产生的时间间隔序列中的某个点，t(n)为n点所对应的时间间隔，0＜T0≤1.5秒，其值大小取决于先验信息，3秒＜T1,其值大小取决于先验信息，f3(n)为输出的由0,1,2三种值构成的序列；Among them, n is a certain point in the time interval sequence generated in step 4, t(n) is the time interval corresponding to n points, 0<T0 ≤ 1.5 seconds, and its value depends on prior information, 3 seconds <T1 , its value depends on prior information, f3 (n) is the output sequence consisting of three values of 0, 1, and 2;
步骤6、判断步骤5中所得的序列中是否含有0、1和2按照特定顺序构成的组合，即符合设定的特定呼吸状态；当发现符合时MCU中央处理模块控制马达微震动模块给予用户第一次触觉反馈，认为用户处于紧急状况需要进行报警求助，并执行报警求助，同时MCU中央处理模块打开GPS模块，通过GPS模块实时监测用户位置信息；当不符合时，即认为用户处于正常状态，保持监测；Step 6. Determine whether the sequence obtained in step 5 contains a combination of 0, 1 and 2 in a specific order, that is, conforms to the set specific breathing state; when found to be consistent, the MCU central processing module controls the motor micro-vibration module to give the user the first One tactile feedback, it is considered that the user is in an emergency situation and needs to call the police for help, and execute the alarm call for help. At the same time, the MCU central processing module turns on the GPS module, and monitors the user's location information in real time through the GPS module; keep monitoring;
步骤7、报警求助执行后，即接收到GSM模块的反馈信息后，MCU中央处理模块控制马达微震动模块给予用户第二次触觉反馈，并且MCU中央处理模块实时地把GPS模块得到的用户位置信息通过GSM模块持续发送到用户亲友手机上。Step 7. After the alarm for help is executed, that is, after receiving the feedback information from the GSM module, the MCU central processing module controls the motor micro-vibration module to give the user the second tactile feedback, and the MCU central processing module real-time user location information obtained by the GPS module Continuously send to the mobile phone of the user's relatives and friends through the GSM module.
所述报警求助是指MCU中央处理模块通过GSM模块拨打110和亲友号码。The calling for help means that the MCU central processing module dials 110 and the numbers of relatives and friends through the GSM module.
本发明通过检测人为主动产生的呼吸变化，即用户事先把正常呼吸(包括正常呼吸的次数和时间)、急促呼吸(包括急促呼吸的次数和时间)和屏息(包括屏息的次数和时间)三种呼吸状态进行n≥1个呼吸状态的排列组合，并设定成一种特定呼吸状态。当用户遇到紧急情况就按照这种事先设定的特定呼吸状态改变自身呼吸状况，进而使设备产生和发送报警求助信号。本发明装置通过检测用户的呼吸过程中是否出现这种设定的特定呼吸状态从而明确用户是否处于紧急状况。实现了误报率低至不误报的可调控，求助方式隐蔽并且安全有效。并且当检测到特定呼吸状态和报警求助完成时，设备给予用户触觉反馈从而减少使用者紧急状况下负面情绪。The present invention detects artificially generated breathing changes, that is, the user has three types of normal breathing (including the number and time of normal breathing), rapid breathing (including the number and time of rapid breathing) and breath holding (including the number and time of breath holding) in advance. The breathing state is arranged and combined with n≥1 breathing states, and set to a specific breathing state. When the user encounters an emergency, he changes his own breathing state according to the specific breathing state set in advance, and then makes the device generate and send an alarm signal for help. The device of the present invention determines whether the user is in an emergency by detecting whether the set specific breathing state occurs during the breathing process of the user. It realizes the controllability of the false alarm rate as low as no false alarm, and the way of seeking help is concealed, safe and effective. And when a specific breathing state is detected and the alarm for help is completed, the device will give the user tactile feedback to reduce the user's negative emotions in an emergency.
正常成年人每次呼吸时间在3-6秒(吸气时间不小于1.5秒、呼气时间不小于1.5秒),换而言之正常人每次正常呼吸都会产生一个时长为3-6秒的带有波动的呼吸信号。另外，当人主动急促呼吸时可以产生一个时长小于3秒(一般为1秒左右)的带有波动的呼吸信号，而当人主动屏息时可以产生数秒甚至数十秒的几乎没有波动的呼吸信号。The breathing time of a normal adult is 3-6 seconds each time (the inhalation time is not less than 1.5 seconds, and the exhalation time is not less than 1.5 seconds). With a fluctuating breathing signal. In addition, when a person actively breathes rapidly, a fluctuating breathing signal with a duration of less than 3 seconds (generally about 1 second) can be generated, while when a person actively holds one's breath, a breathing signal with almost no fluctuation can be generated for several seconds or even tens of seconds .
综上所述，本发明的设备实现了求助隐蔽、误报率低、安全有效的紧急求助、减少使用者紧急状况下负面情绪和求助条件的可控性设定。To sum up, the device of the present invention realizes concealment of help-seeking, low false alarm rate, safe and effective emergency help-seeking, reduction of user's negative emotions in emergency situations and controllable setting of help-seeking conditions.
附图说明Description of drawings
图1为本发明的检测方法示意图；Fig. 1 is a schematic diagram of the detection method of the present invention;
图2为本发明的设备结构示意框图；Fig. 2 is a schematic block diagram of the device structure of the present invention;
图3为实施例设定的特定呼吸状态示意图；Fig. 3 is the schematic diagram of the specific breathing state set by the embodiment;
图4为带有特定呼吸状态顺序的使用阈值算法的结果图；Fig. 4 is the result figure of using the threshold value algorithm with specific respiratory state sequence;
图5为不带有特定呼吸状态顺序的使用阈值算法的结果图；Fig. 5 is a result diagram of using a threshold algorithm without a specific respiratory state sequence;
图6为带有特定呼吸状态顺序的使用特征算法的结果图；Fig. 6 is the result figure of using characteristic algorithm with specific breathing state sequence;
图7为不带有特定呼吸状态顺序的使用特征算法的结果图。Fig. 7 is a graph of the result of using the feature algorithm without a specific order of breathing states.
具体实施方案specific implementation plan
下面结合附图和具体的实施对本发明做进一步的阐述。The present invention will be further elaborated below in conjunction with the accompanying drawings and specific implementation.
选用电压转换电路为+3.3V&+1.8V生成电路。图3为本实施例设定的特定呼吸状态示意图，事先设定急促呼吸3次然后紧接着屏息4秒以上作为特定呼吸状态。当发现用户出现急促呼吸3次然后屏息4秒以上的现象即认为用户处于紧急状况从而帮助用户报警求助。The voltage conversion circuit is selected as +3.3V&+1.8V generating circuit. FIG. 3 is a schematic diagram of the specific breathing state set in this embodiment. The specific breathing state is set in advance as the specific breathing state after 3 short breaths followed by breath holding for more than 4 seconds. When it is found that the user has 3 short breaths and then holds his breath for more than 4 seconds, it is considered that the user is in an emergency situation and helps the user to call the police for help.
对于图3所示的特定呼吸状态，上述基于呼吸信号的自动报警求助设备的检测方法，包括以下步骤：For the specific breathing state shown in Figure 3, the detection method of the above-mentioned automatic alarm and help-seeking equipment based on the breathing signal includes the following steps:
步骤1、基于先验信息或快速傅里叶变换FFT明确需要检测的呼吸信号当中是否含有直流成分；Step 1. Based on prior information or fast Fourier transform FFT, it is determined whether the respiratory signal to be detected contains a DC component;
所述先验信息是指根据呼吸信号测量方法来明确所需检测的呼吸信号是否带有直流成分；The prior information refers to determining whether the respiratory signal to be detected has a DC component according to the respiratory signal measurement method;
所述快速傅里叶变换是指对信号进行时域到频域的转换然后分析0Hz成分的大小从而明确所需检测的呼吸信号是否带有直流成分；The fast Fourier transform refers to converting the signal from the time domain to the frequency domain and then analyzing the size of the 0Hz component so as to clarify whether the respiratory signal to be detected has a DC component;
步骤2、通过阈值算法或特征算法把需要检测的呼吸信号转换成只含有-1,1两种值的方波信号；对于不含有直流成分的呼吸信号首选阈值算法作为转换方法，对于含有直流成分的呼吸信号首选特征算法作为转换方法；Step 2. Convert the breathing signal to be detected into a square wave signal containing only two values of -1 and 1 through a threshold algorithm or a feature algorithm; the threshold algorithm is the preferred conversion method for breathing signals that do not contain DC components, and the conversion method for breathing signals that contain DC components The preferred feature algorithm of the respiratory signal is used as the conversion method;
所述的阈值算法是指：将所需检测的呼吸信号中的每点按照特殊映射f1(n)进行阈值比较从而得到一组与输入数据等长的只含有-1,1两种值的方波信号；如果某点数值大于X值，f1(n)设为1；如果某点数值小于-X值，f1(n)设为-1；如果某点数值处于-X和X之间时，f1(n)保持f1(n-1)的值；The threshold algorithm refers to: compare each point in the respiratory signal to be detected according to the special map f1 (n) to obtain a set of values equal to the input data and only contain two values -1 and 1. Square wave signal; if the value of a certain point is greater than X value, f1 (n) is set to 1; if the value of a certain point is less than -X value, f1 (n) is set to -1; if the value of a certain point is between -X and X time, f1 (n) maintains the value of f1 (n-1);
其中，n为所需检测的呼吸信号中某点；x(n)为所需检测的呼吸信号中n点的值；f1(n-1)为n-1点的方波信号值；X值取0.00005；f1(n)为本步骤输出的方波信号；Wherein, n is a certain point in the respiratory signal to be detected; x(n) is the value of n points in the respiratory signal to be detected; f1 (n-1) is the square wave signal value of n-1 point; X The value is 0.00005; f1 (n) is the square wave signal output by this step;
所述的特征算法是指：首先，寻找所需检测的呼吸信号中的极大值和极小值；具体过程如下：The characteristic algorithm refers to: first, find the maximum value and the minimum value in the respiratory signal to be detected; the specific process is as follows:
①将所需检测的呼吸信号中的每点与其下一点做差，得一组差值数据diff1(n)；① Make a difference between each point in the respiratory signal to be detected and the next point to obtain a set of difference data diff1 (n);
diff1(n)＝x(n+1)-x(n)(n＝0，1，...)diff1 (n)=x(n+1) -x(n) (n=0,1,...)
其中，n为所需检测的呼吸信号中某点；n+1为所需检测的呼吸信号中n点的下一点；x(n)为所需检测的呼吸信号中n点的值；x(n+1)为所需检测的呼吸信号中n+1点的值；diff1(n)为输出的差值数据；Wherein, n is a certain point in the respiratory signal to be detected; n+1 is the next point of n points in the respiratory signal to be detected; x(n) is the value of n points in the respiratory signal to be detected; x( n+1) is the value of n+1 points in the respiratory signal to be detected; diff1 (n) is the difference data of output;
②寻找diff1(n)数据中符合diff1(n)≥0且diff1(n-1)≤0或diff1(n)≤0且diff1(n-1)≥0条件的点；这些点即为所需找的呼吸信号中的极大值和极小值；② Find the points in the diff1 (n) data that meet the conditions of diff1 (n) ≥ 0 and diff1 (n-1) ≤ 0 or diff1 (n) ≤ 0 and diff1 (n-1) ≥ 0; these The points are the maximum and minimum values in the respiratory signal to be found;
然后，对信号中的所有极值点进行分析；具体过程如下：Then, analyze all extreme points in the signal; the specific process is as follows:
①将每个极点与其下一个极点做差，得一组差值数据diff2(m)；① Make a difference between each pole and the next pole to obtain a set of difference data diff2 (m);
diff2(m)＝x(m+1)-x(m)diff2 (m)＝x(m+1) -x(m)
其中，m为呼吸信号中某极值点坐标；m+1为m点后下一个极值点坐标；x(m)为所需检测的呼吸信号中极值点m对应的值；x(m+1)为所需检测的呼吸信号中极值点m+1对应的值；diff2(m)为输出的差值数据；Among them, m is the coordinates of an extreme point in the respiratory signal; m+1 is the coordinate of the next extreme point after point m; x(m) is the value corresponding to the extreme point m in the respiratory signal to be detected; x(m +1) is the value corresponding to the extremum point m+1 in the respiratory signal to be detected; diff2 (m) is the output difference data;
②对diff2(m)数据进行分析，将所需检测的呼吸信号中的每点按照特殊映射f2(n)进行转换从而得到一组与输入数据等长的只含有-1,1两种值的方波信号；如果diff2(m)值大于D且极值点m和极值点m+1的时间间隔大于T,则把极值点m和极值点m+1之内的所有f2(n)设为1；如果diff2(m)值小于-D且极值点m和极值点m+1的时间间隔大于T,则把极值点m和极值点m+1之内的所有f2(n)设为-1；其余的f2(n)保持上一个f2(n-1)的值；②Analyze the diff2 (m) data, convert each point in the respiratory signal to be detected according to the special mapping f2 (n) to obtain a set of data with the same length as the input data containing only -1, 1 value of the square wave signal; if the value of diff2 (m) is greater than D and the time interval between extreme point m and extreme point m+1 is greater than T, then all the values within extreme point m and extreme point m+1 f2 (n) is set to 1; if the value of diff2 (m) is less than -D and the time interval between extreme point m and extreme point m+1 is greater than T, then extreme point m and extreme point m+1 All f2 (n) within are set to -1; the remaining f2 (n) keep the value of the previous f2 (n-1);
其中，n为所需检测的呼吸信号中某点，m为呼吸信号中某极值点坐标，m+1为m点后下一个极值点坐标，diff2(m)为极值点m对应的差值数据，tm为极值点m对应的时刻，tm+1为极值点m+1对应的时刻；D值取0.00005；T值取0.3；f2(n)为本步骤输出的方波信号；Among them, n is a certain point in the respiratory signal to be detected, m is the coordinate of an extreme point in the respiratory signal, m+1 is the coordinate of the next extreme point after point m, and diff2 (m) is the corresponding extreme point m tm is the time corresponding to extreme point m, tm+1 is the time corresponding to extreme point m+1; D value is 0.00005; T value is 0.3; f2 (n) is the output of this step square wave signal;
步骤3、通过跳变点提取算法寻找出步骤2产生的方波信号中的跳变点并把这些点构成跳变点序列；Step 3, find out the jump points in the square wave signal that step 2 produces by the jump point extraction algorithm and form these points into a jump point sequence;
所述的跳变点提取算法是指：Described jump point extracting algorithm refers to:
①将步骤2产生的方波信号中每点与其下一点做差，得一组差值信号diff3(n)① Make a difference between each point in the square wave signal generated in step 2 and the next point to obtain a set of difference signals diff3 (n)
diff3(n)＝y(n+1)-y(n)(n＝0，1，...)diff3 (n) = y(n+1) -y(n) (n = 0, 1, ...)
其中，n为所需检测的方波信号中某点，y(n)为所需检测的方波信号中n点的值，y(n+1)为所需检测的方波信号中n+1点的值，diff3(n)为输出的差值数据；Among them, n is a certain point in the square wave signal to be detected, y(n) is the value of n points in the square wave signal to be detected, and y(n+1) is n+ in the square wave signal to be detected The value of 1 point, diff3 (n) is the output difference data;
②对diff3(n)数据进行分析；当diff3(n)不为0时，则代表方波信号在n点处发生跳变，即n点和n+1点为所要寻找的跳变点从而加入跳变点序列；当diff3(n)为0时，则代表方波信号在n点处没有发生跳变，即n点和n+1点不为所要寻找的跳变点从而不加入跳变点序列；② Analyze the data of diff3 (n); when diff3 (n) is not 0, it means that the square wave signal jumps at point n, that is, point n and point n+1 are the jump points to be found Thus adding the jump point sequence; when diff3 (n) is 0, it means that the square wave signal does not jump at point n, that is, point n and n+1 are not the jump points to be found, so do not add jump point sequence;
步骤4、计算步骤3中跳变点序列中相邻两点之间的时间间隔，然后去除那些间隔为一个采样周期的间隔，将剩余的时间间隔按序构成时间间隔序列；Step 4, calculate the time interval between two adjacent points in the jump point sequence in step 3, then remove those intervals that are intervals of a sampling period, and form the time interval sequence with the remaining time intervals in order;
步骤5、将步骤4中所得的时间间隔序列进行特殊映射f3(n)得到一组由0,1,2三种值构成的序列；如果步骤4中所得的时间间隔序列中某点对应的时间间隔小于T0,则把f3(n)设为1；如果步骤4中所得的时间间隔序列中某点对应的时间间隔大于T1,则把f3(n)设为2；其余f3(n)设为0：Step 5. Perform special mapping f3 (n) on the time interval sequence obtained in step 4 to obtain a sequence consisting of three values of 0, 1, and 2; if a certain point in the time interval sequence obtained in step 4 corresponds to If the time interval is less than T0 , set f3 (n) to 1; if the time interval corresponding to a point in the time interval sequence obtained in step 4 is greater than T1 , then set f3 (n) to 2; the rest f3 (n) is set to 0:
其中，n为步骤4中产生的时间间隔序列中的某个点，t(n)为n点所对应的时间间隔，T0值取1.25，T1值取4，f3(n)为输出的由0,1,2三种值构成的序列；Among them, n is a certain point in the time interval sequence generated in step 4, t(n) is the time interval corresponding to point n, the value of T0 is 1.25, the value of T1 is 4, and f3 (n) is the output A sequence consisting of three values of 0, 1, and 2;
步骤6、判断步骤5中所得的序列中是否含有0、1和2按照特定顺序构成的组合，即符合设定的特定呼吸状态，如图3所示的特定呼吸状态下需要检测是否存在“1111112”这种组合；当发现符合时MCU中央处理模块控制马达微震动模块给予用户第一次触觉反馈，认为用户处于紧急状况需要进行报警求助，并执行报警求助，同时MCU中央处理模块打开GPS模块，通过GPS模块实时监测用户位置信息；当不符合时，即认为用户处于正常状态，保持监测；Step 6. Determine whether the sequence obtained in step 5 contains a combination of 0, 1 and 2 in a specific order, that is, conforms to the set specific breathing state. In the specific breathing state shown in Figure 3, it is necessary to detect whether there is "1111112 "This combination; when the MCU central processing module controls the motor micro-vibration module to give the user the first tactile feedback, it thinks that the user is in an emergency and needs to call for help, and executes the call for help. At the same time, the MCU central processing module turns on the GPS module, Monitor the user's location information in real time through the GPS module; when it does not match, the user is considered to be in a normal state and keeps monitoring;
步骤7、报警求助执行后，即接收到GSM模块的反馈信息后，MCU中央处理模块控制马达微震动模块给予用户第二次触觉反馈，并且MCU中央处理模块实时地把GPS模块得到的用户位置信息通过GSM模块持续发送到用户亲友手机上。Step 7. After the alarm for help is executed, that is, after receiving the feedback information from the GSM module, the MCU central processing module controls the motor micro-vibration module to give the user the second tactile feedback, and the MCU central processing module real-time user location information obtained by the GPS module Continuously send to the mobile phone of the user's relatives and friends through the GSM module.
本方法检测人为主动产生的呼吸变化具有十分高的准确性，具体效果如下：This method has very high accuracy in the detection of artificially generated breathing changes, and the specific effects are as follows:
如图4带有特定呼吸状态顺序的使用阈值算法的结果图所示，图中第一个波形为所需检测的呼吸信号波形；第二个波形为经过阈值算法转换的方波波形，其中黑点为经过跳变点提取算法找到的跳变点；第三个波形为后续步骤处理的结果，其中虚线为检测到不含有特定呼吸状态顺序的部分，实线为检测到的含有特定呼吸状态顺序的部分。本方法准确地发现本次检测中带有某种事先设定的特定呼吸状态，即急促呼吸3次、屏息4秒。As shown in Figure 4, the results of using threshold algorithm with a specific respiratory state sequence, the first waveform in the figure is the respiratory signal waveform to be detected; the second waveform is the square wave waveform converted by the threshold algorithm, in which black The dots are the jump points found by the jump point extraction algorithm; the third waveform is the result of subsequent steps, where the dotted line is the part that does not contain a specific respiratory state sequence, and the solid line is the detected sequence that contains a specific respiratory state part. This method accurately finds that there is a certain pre-set specific breathing state in this test, that is, rapid breathing for 3 times and breath holding for 4 seconds.
如图5不带有特定呼吸状态顺序的使用阈值算法的结果图所示，图中第一个波形为所需检测的呼吸信号波形；第二个波形为经过阈值算法转换的方波波形，其中黑点为经过跳变点提取算法找到的跳变点；第三个波形为后续步骤处理的结果，虚线为检测到不含有特定呼吸状态顺序的部分。本方法准确地发现本次检测中不带有某种事先设定的特定呼吸状态。As shown in Figure 5, the result of using the threshold algorithm without a specific respiratory state sequence, the first waveform in the figure is the respiratory signal waveform to be detected; the second waveform is a square wave waveform converted by the threshold algorithm, where The black dots are the jump points found by the jump point extraction algorithm; the third waveform is the result of subsequent steps, and the dotted line is the part that does not contain a specific respiratory state sequence. The method accurately finds that there is no specific breathing state set in advance in this detection.
如图6带有特定呼吸状态顺序的使用特征算法的结果图所示，图中第一个波形为所需检测的呼吸信号波形，其中黑点为经过特征算法找寻到各个极值点；第二个波形为经过特征算法转换的方波波形，其中黑点为经过跳变点提取算法找到的跳变点；第三个波形为后续步骤处理的结果，其中虚线为检测到不含有特定呼吸状态顺序的部分，实线为检测到的含有特定呼吸状态顺序的部分。本方法准确地发现本次检测中带有某种事先设定的特定呼吸状态，即急促呼吸3次、屏息4秒。As shown in Figure 6, the results of using the characteristic algorithm with a specific respiratory state sequence, the first waveform in the figure is the respiratory signal waveform to be detected, and the black dots are the extreme points found through the characteristic algorithm; the second The first waveform is the square wave waveform transformed by the feature algorithm, and the black dots are the jump points found by the jump point extraction algorithm; the third waveform is the result of the subsequent steps, and the dotted line is the detection sequence that does not contain a specific respiratory state The part of , the solid line is the detected part containing a specific sequence of respiratory states. This method accurately finds that there is a certain pre-set specific breathing state in this test, that is, rapid breathing for 3 times and breath holding for 4 seconds.
如图7不带有特定呼吸状态顺序的使用阈值算法的结果图所示，图中第一个波形为所需检测的呼吸信号波形，其中黑点为经过特征算法找寻到各个极值点；第二个波形为经过特征算法转换的方波波形，其中黑点为经过跳变点提取算法找到的跳变点；第三个波形为后续步骤处理的结果，虚线为检测到不含有特定呼吸状态顺序的部分。本方法准确地发现本次检测中不带有某种事先设定的特定呼吸状态。As shown in Figure 7, the results of using the threshold algorithm without a specific respiratory state sequence, the first waveform in the figure is the respiratory signal waveform to be detected, and the black dots are the extreme points found through the characteristic algorithm; the second The second waveform is the square wave waveform converted by the feature algorithm, and the black dots are the jump points found by the jump point extraction algorithm; the third waveform is the result of the subsequent steps, and the dotted line is the detection sequence that does not contain a specific respiratory state part. The method accurately finds that there is no specific breathing state set in advance in this detection.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610407597.0ACN105877749B (en) | 2016-06-08 | 2016-06-08 | A kind of automatic alarm help-seeking equipment and its detection method based on breath signal |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610407597.0ACN105877749B (en) | 2016-06-08 | 2016-06-08 | A kind of automatic alarm help-seeking equipment and its detection method based on breath signal |
| Publication Number | Publication Date |
|---|---|
| CN105877749Atrue CN105877749A (en) | 2016-08-24 |
| CN105877749B CN105877749B (en) | 2019-04-05 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610407597.0AExpired - Fee RelatedCN105877749B (en) | 2016-06-08 | 2016-06-08 | A kind of automatic alarm help-seeking equipment and its detection method based on breath signal |
| Country | Link |
|---|---|
| CN (1) | CN105877749B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH1188546A (en)* | 1997-09-12 | 1999-03-30 | Nec Corp | Crisis notice system |
| CN101620649A (en)* | 2009-08-07 | 2010-01-06 | 四川长虹电器股份有限公司 | Real-time monitoring method for human health based on network communication |
| CN202904817U (en)* | 2012-10-18 | 2013-04-24 | 华北电力大学（保定） | Soundless sleep apnea prompting system |
| CN203016930U (en)* | 2013-01-30 | 2013-06-26 | 冯军营 | Electronic information terminal for medical treatment |
| CN103705214A (en)* | 2013-12-09 | 2014-04-09 | 合肥博谐电子科技有限公司 | Health secretary |
| CN205019041U (en)* | 2015-08-24 | 2016-02-10 | 浙江大学 | Data acquisition system based on intelligence bracelet |
| CN105496428A (en)* | 2015-12-14 | 2016-04-20 | 北京奇虎科技有限公司 | Wearable device control implementation method and device |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH1188546A (en)* | 1997-09-12 | 1999-03-30 | Nec Corp | Crisis notice system |
| CN101620649A (en)* | 2009-08-07 | 2010-01-06 | 四川长虹电器股份有限公司 | Real-time monitoring method for human health based on network communication |
| CN202904817U (en)* | 2012-10-18 | 2013-04-24 | 华北电力大学（保定） | Soundless sleep apnea prompting system |
| CN203016930U (en)* | 2013-01-30 | 2013-06-26 | 冯军营 | Electronic information terminal for medical treatment |
| CN103705214A (en)* | 2013-12-09 | 2014-04-09 | 合肥博谐电子科技有限公司 | Health secretary |
| CN205019041U (en)* | 2015-08-24 | 2016-02-10 | 浙江大学 | Data acquisition system based on intelligence bracelet |
| CN105496428A (en)* | 2015-12-14 | 2016-04-20 | 北京奇虎科技有限公司 | Wearable device control implementation method and device |
| Publication number | Publication date |
|---|---|
| CN105877749B (en) | 2019-04-05 |
| Publication | Publication Date | Title |
|---|---|---|
| CN204731935U (en) | A kind of device of intellectual monitoring Falls in Old People | |
| CN106128475A (en) | Wearable intelligent safety equipment based on abnormal emotion speech recognition and control method | |
| EP1710766A1 (en) | Human body monitoring system | |
| CN104822174B (en) | A kind of communication pattern method of adjustment | |
| CN113205661A (en) | Anti-cheating implementation method and system, intelligent wearable device and storage medium | |
| WO2015096567A1 (en) | User behavior safety monitoring method and device | |
| CN108464821A (en) | A kind of intelligent wristwatch for human body physical sign monitoring | |
| CN105847561A (en) | Environmental monitoring method, device and handset | |
| Jijesh et al. | A method for the personal safety in real scenario | |
| CN204719923U (en) | Based on the grading automatical warning device of mobile phone in danger of wearable device | |
| CN108670219A (en) | A kind of multifunctional intellectual watch | |
| CN111820880A (en) | Campus overlord early warning system and method | |
| CN108597182A (en) | A kind of old man's Emergency Assistance calling set | |
| Monalisha et al. | IoT based safety system for women | |
| CN108416978A (en) | emergency call system | |
| CN105877749A (en) | Automatic alarm help-seeking equipment based on respiration signal and detection method of automatic alarm help-seeking equipment | |
| CN206757843U (en) | A kind of wearable emergency alarm help facility | |
| CN201351415Y (en) | Remote restraining device | |
| CN107564244A (en) | A kind of wearable personal security's guard system | |
| CN205486703U (en) | GSM audio alert ware | |
| Kumbhar et al. | Women security system using GSM and GPS | |
| CN204580009U (en) | A kind of intelligent waistband | |
| CN107402321A (en) | A kind of shockproof alarm circuit and wrist type Antistatic alarm | |
| CN103051360B (en) | Based on the system and method for RFID wireless connections | |
| CN207232240U (en) | A kind of shockproof alarm circuit and wrist type Antistatic alarm |
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee | Granted publication date:20190405 |