Passing and guiding integrated wireless ad hoc network pedestrian positioning system and implementation method thereof

Abstract

The invention discloses a communication and navigation integrated wireless ad hoc network pedestrian positioning system and an implementation method thereof, and relates to the technical field of wireless ad hoc networks and positioning navigation. The method of the system comprises the following steps: construction of the conduction integrated wireless ad hoc network is divided into ad hoc network anchor node initial positioning and ad hoc network anchor node fine positioning; the wireless ad hoc network assisted pedestrian positioning is divided into mobile label signal acquisition and calibration position fusion positioning. According to the positioning system and the implementation method thereof, a large number of base station facilities do not need to be arranged in a positioning area in advance, a positioning database and a training model do not need to be constructed, positioning anchor points are arranged in real time, real-time area expansion of a navigation positioning network is achieved, follow-up pedestrian track positioning is completed, and initial positioning and fine positioning of anchor nodes are achieved; the function that the navigation positioning network expands along with the area under the condition that no preset base station exists in an underground complex scene is achieved, and safe, stable and high-precision position service on a mobile terminal is provided for a user in various complex environments.

Description

Technical Field

[0001] The present invention relates to the field of wireless self-organizing network and positioning navigation technology, and in particular to a communication and navigation integrated wireless self-organizing network pedestrian positioning system and an implementation method thereof. Background Art

[0002] As location-based services (LBS) penetrate into all aspects of social activities, people's demand for indoor positioning is also increasing. At present, there are two main indoor positioning methods: wireless network-based and inertial principle-based. Wireless network methods are seriously affected by multipath and non-line-of-sight, and inertial principle-based methods have errors that diverge over time due to the defects of the sensor itself. The single application of the above methods has certain limitations, and it is difficult to obtain high-precision usable navigation observation results. Therefore, in terms of positioning methods, it is necessary to achieve higher-precision positioning through multi-source fusion.

[0003] In addition, based on the positioning device hardware, the current common positioning systems are based on the early base station deployment and positioning database learning to achieve wireless network positioning information acquisition. It is necessary to complete the early scene setting to meet the subsequent positioning requirements, resulting in poor applicability of the method. It is impossible to achieve high-precision positioning in any environment and it is impossible to quickly build a positioning benchmark.

[0004] A wireless ad hoc network is a temporary multi-hop autonomous system composed of a group of mobile nodes with wireless transceivers. It does not rely on preset infrastructure and has the characteristics of temporary networking, rapid deployment, no control center, and strong anti-destruction capabilities. It has broad application prospects in military, civil and civilian aspects and is a hot issue in network research.

[0005] Positioning technology is developing towards low cost, easy portability, high precision, multi-fusion, and seamless indoor and outdoor positioning. Ultra-wideband (UWB) can obtain high-precision distance information, but it is seriously affected by multipath and non-line-of-sight (NLOS); pedestrian dead reckoning (PDR) has strong anti-interference ability and can accurately estimate relative motion changes in a short period of time, but the error will diverge over time. The two technologies complement each other, and high-precision and high-stability positioning results can be obtained through information fusion. At the same time, the indoor environment is complex and changeable, and it is crucial to quickly build a positioning benchmark. The positioning solution based on wireless self-organizing networks is flexible and easy to implement, and may solve the positioning problem in harsh or unknown environments.

[0006] The current positioning methods mainly include traditional GNSS positioning, wireless signal positioning (based on WIFI, Bluetooth, ultra-wideband, etc.) and multi-source sensor positioning. Although GNSS has always been the main solution for LBS, the problem of difficulty in use in areas such as cities, canyons and indoors due to severe signal attenuation and multipath phenomena has not been fundamentally solved; the indoor wireless signal positioning technology of opportunity signals can be mainly divided into fingerprint positioning and geometric intersection positioning. Among them, the opportunity signal fingerprint positioning method requires a lot of manpower and material costs to carry out a large amount of preliminary data collection work and regularly update the fingerprint data. Although the traditional geometric intersection indoor positioning method can obtain better navigation and positioning observation results from high signal-to-noise ratio signals in scenes with weak multipath effects. However, in most GNSS-denied areas, these geometric intersection positioning methods also face signals with strong multipath effects and low signal-to-noise ratios. In addition, the expansion of the navigation and positioning network is too complicated, and once it reaches a blind environment without a preset base station, positioning services cannot be provided.

[0007] In summary, complex scenes such as underground and indoors cannot effectively receive satellite signals for positioning. Limited by the diverse spatial layout of the indoor positioning environment, the complex topological structure, and the harsh channel environment, it is still quite difficult and challenging to achieve high-precision and high-availability indoor positioning technology. Existing research on indoor positioning mainly focuses on cellular signals, UWB, Wi-Fi, BLE, images, geomagnetic fields, motion signals, etc. Although each has its own advantages, it is still very challenging to achieve high-precision, high-availability, and wide-coverage indoor positioning technology, especially in complex spaces without pre-deployed positioning infrastructure. High-precision and high-availability positioning requires urgent breakthroughs. Summary of the invention

[0008] The purpose of the present invention is to provide a wireless ad hoc network pedestrian positioning system with integrated communication and navigation and an implementation method thereof, so as to quickly establish a positioning reference in a target scene and realize pedestrian fusion positioning.

[0009] To achieve the above object, the present invention provides a communication and navigation integrated wireless ad hoc network pedestrian positioning system and an implementation method thereof, comprising the following steps:

[0010] S1. Construct a wireless ad hoc network with integrated communication and navigation, as shown below:

[0011] S11, initial positioning of ad hoc network anchor nodes;

[0012] S12, precise positioning of ad hoc network anchor nodes;

[0013] S2, wireless ad hoc network assisted pedestrian positioning, as shown below:

[0014] S21, mobile tag signal collection;

[0015] S22, calibrated position fusion positioning.

[0016] Preferably, in step S11, in the initial positioning stage of the ad hoc network anchor node, a pedestrian dead reckoning algorithm PDR is used, including detecting gait, estimating step length and estimating heading, as shown below:

[0017] S111, gait detection uses an adaptive peak detection method, which can automatically set the threshold according to the data situation, as shown below:

[0018]

[0019] Where a={a1,a2,…,a N} is the acceleration sequence without gravity effect, N is the length of the sequence; sort(·) is the sorting operation from small to large, and 85% of a sequence is taken as the standard for setting the threshold;

[0020] S112, dead reckoning uses a dynamic model to estimate the step length of each step;

[0021] S113, heading estimation uses a method combining three sensors, gyroscope, accelerometer and magnetometer, to obtain the current three-axis attitude through complementary filtering CF;

[0022] S114, the position of each anchor node is obtained through adaptive gait detection, dynamic step length estimation and complementary filter CF heading estimation to complete the initial positioning.

[0023] Preferably, in the precise positioning stage of the ad hoc network anchor node in step S12, the wire network adjustment method is adopted, and the updated value of the coordinate position is obtained through cyclic iteration, and the next calculation is performed until the range of the updated value variation is less than the set threshold or the number of iterations exceeds the set value. The cycle is ended to obtain the final position estimate and complete the precise positioning of the anchor point coordinates.

[0024] Preferably, in step S21 of collecting mobile tag signals, the tag system and the anchor point system have similar hardware structures and are flexibly configured according to requirements;

[0025] During the tag system experiment, MEMS sensors are used to collect real-time inertial information of targets with positioning requirements. UWB sensors are used to measure the distance to each anchor point within the range in real time. The data collection process ensures time synchronization and completes real-time data recording and updating.

[0026] Preferably, step S22 of calibrating the position and fusion positioning specifically includes the following steps:

[0027] S221, first, the tag collects the accelerometer, gyroscope and magnetometer counter through the inertial sensor for preprocessing and filters out some noise;

[0028] S222. Then, execute the PDR algorithm to detect whether gait occurs, and calculate the length and direction of each step.

[0029] Preferably, in step S222, when a gait occurs and the UWB distance is measured, a particle filtering process is induced:

[0030] First, the particle state is updated by PDR and the state is estimated by UWB distance;

[0031] Then, in each recursive process, the chi-square test is used to determine whether there are abnormal observations, and the observation variance is adjusted according to the test results;

[0032] Finally, weight calculation is performed to suppress the influence of outliers.

[0033] Preferably, the process of building the communication and navigation integrated wireless ad hoc network in step S1 is divided into two stages: scene layout and algorithm processing.

[0034] Preferably, in the scene layout stage, the specific steps are as follows:

[0035] First, place an anchor node in an area close to the outdoors. The location of the anchor node is easy to measure and is recorded as the known base station KA.

[0036] Then, carry IMU and several base stations from one of the KA points and deploy base stations during the walking process. Since the positions of these base stations are not known in advance, they are recorded as unknown base stations UA;

[0037] Finally, anchor nodes can be placed at every corner, and it is sufficient to ensure line of sight with adjacent base stations. Once the base stations are deployed, the entire positioning network can be built by walking to another known base station.

[0038] Preferably, the algorithm processing stage needs to determine the coordinate information of the unknown base station, which is specifically divided into two processes. The specific steps are as follows:

[0039] First, the gait is detected based on the sensor data collected by the IMU, and the heading and step length are calculated to implement the pedestrian dead reckoning algorithm and perform the initial positioning of each unknown base station;

[0040] Then, according to the ranging information between adjacent base stations and the heading angle at each base station, the traverse network adjustment based on the ranging and heading information constraints is implemented to obtain the precise positioning result of the base station.

[0041] Therefore, the present invention adopts the above-mentioned integrated wireless self-organizing network pedestrian positioning system and its implementation method. There is no need to deploy a large number of base station facilities in the positioning area in advance, nor is there a need to build a positioning database or training model. Instead, a UWB self-organizing network of the channel is constructed with the assistance of PDR, so that the positioning anchor points are arranged in real time and the navigation and positioning network is expanded in real time, and the subsequent pedestrian trajectory positioning can be completed; the initial positioning and precise positioning of the anchor nodes are realized based on the inertial navigation / conductor network adjustment, and a integrated wireless self-organizing network of communication and navigation is constructed, which has the benchmark construction in blind environments or emergency scenarios, and realizes the navigation and positioning network expansion function with the region in complex underground scenarios without preset base stations, providing users with safe, stable and high-precision positioning services on mobile terminals in a variety of complex environments.

[0042] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is the overall technical route of the integrated wireless ad hoc network pedestrian positioning system and its implementation method;

[0044] Figure 2 It is a description of the network construction process of the integrated wireless ad hoc network pedestrian positioning system and its implementation method;

[0045] Figure 3 It is a fusion algorithm flow chart of the communication and navigation integrated wireless ad hoc network pedestrian positioning system and its implementation method;

[0046] Figure 4 The invention is a self-organizing network scheme of a wireless self-organizing network pedestrian positioning system with integrated communication and navigation and its implementation method; wherein (a) is a schematic diagram of scheme 1, and (b) is a schematic diagram of scheme 2;

[0047] Figure 5 It is a plan view of a network scene of a wireless ad hoc network pedestrian positioning system with integrated communication and navigation and its implementation method; wherein (a) is a plan view of a network scene of scene 1, and (b) is a plan view of a network scene of scene 2;

[0048] Figure 6 It is the positioning error of different algorithms of the integrated wireless ad hoc network pedestrian positioning system and its implementation method; where (a) is the positioning error of different algorithms in scenario 1 without the introduction of base station error; (b) is the positioning error of different algorithms in scenario 1 with the introduction of base station error; (c) is the positioning error of different algorithms in scenario 2 without the introduction of base station error; (d) is the positioning error of different algorithms in scenario 2 with the introduction of base station error. DETAILED DESCRIPTION

[0049] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0050] Example 1

[0051] like Figure 1 As shown, the communication and navigation integrated wireless ad hoc network pedestrian positioning system and its implementation method of the present invention are divided into two main parts: communication and navigation integrated wireless ad hoc network construction and wireless ad hoc network assisted pedestrian positioning, which are specifically shown as follows:

[0052] S1. Construction of wireless ad hoc network with integrated communication and navigation.

[0053] The construction process of the integrated wireless ad hoc network is divided into two stages: scene layout and algorithm processing. Figure 2 shown.

[0054] In the scene layout stage, first, place anchor nodes in areas close to the outdoors (usually places such as doorways or windows where satellite signals can be received). The location of the anchor nodes is easy to measure and is recorded as the known base station KA (KnownAnchor); then, carry an IMU and several base stations starting from one of the KA points and deploy base stations while walking. Since the locations of these base stations are not known in advance, they are recorded as unknown base stations UA (UnknownAnchor). Anchor nodes can be placed at every corner, as long as the adjacent base stations are visible; when the base stations are deployed, walk to another known base station to complete the construction of the entire positioning network.

[0055] The algorithm processing stage needs to determine the coordinate information of the unknown base stations, which is divided into two processes. First, the gait is detected based on the sensor data collected by the IMU, and the heading and step length are calculated to implement the pedestrian dead reckoning algorithm and perform initial positioning of each unknown base station. Then, based on the ranging information between adjacent base stations and the heading angle at each base station, the traverse network adjustment based on the ranging and heading information constraints is implemented to obtain the precise positioning result of the base station.

[0056] S11. Initial positioning of the ad hoc network anchor node.

[0057] The pedestrian dead reckoning algorithm is used in the initial positioning stage of the ad hoc network anchor node. A classic PDR algorithm includes three steps: gait detection, step length estimation, and heading estimation.

[0058] S111. Gait detection uses an adaptive peak detection method, which can automatically set the threshold according to the data situation to reduce human intervention, as shown below:

[0059]

[0060] Where a={a1,a2,…,a N} is the acceleration sequence without the influence of gravity, N is the length of the sequence; sort(·) is the sorting operation from small to large, and in order to avoid the influence of abnormally large values, 85% of a sequence is taken as the standard for setting the threshold.

[0061] S112. The walking distance in dead reckoning is the sum of all step lengths, so it is necessary to estimate the length of each step. The present invention uses a dynamic model for estimation, and the dynamic model assumes that the length of each step of each person is different, and in addition to height, it is also related to weight, step frequency, exercise habits, etc., so that the accurate step length can be better calculated.

[0062] S113, heading estimation part The method adopted by the present invention is to combine three sensors, namely, a gyroscope, an accelerometer and a magnetometer, and obtain the current three-axis attitude through complementary filtering (CF).

[0063] S114: The position of each anchor node can be obtained through adaptive gait detection, dynamic step length estimation and CF heading estimation to complete the initial positioning.

[0064] S12. Precise positioning of ad hoc network anchor nodes.

[0065] The precise positioning of anchor nodes in the ad hoc network uses the wire network adjustment method, which has two types of observation values, namely the heading angle obtained by PDR heading estimation and the side length between adjacent anchor points obtained by UWB ranging.

[0066] The updated value of the coordinate position is continuously obtained through loop iteration, and the next calculation is performed until the update value variation range is less than the set threshold or the number of iterations exceeds the set value. The loop ends and the final position estimate is obtained to achieve precise positioning of the anchor point coordinates.

[0067] S2. Wireless ad hoc network assists pedestrian positioning.

[0068] After the wireless ad hoc network is built, the location information of each anchor point is obtained. Since the positioning technology based on UWB ranging is affected by non-line-of-sight errors and multipath effects, the ranging accuracy is reduced, and in some areas, only one or one base station signal can be received. If the number of base stations is less than 3, positioning cannot be completed. Therefore, the positioning method of UWB wireless ad hoc network and MEMS sensor fusion is considered. In order to avoid the error accumulation caused by multiple integration of sensor measurement data, the PDR algorithm is selected instead of the traditional INS algorithm. In this positioning system, the data involved include UWB distance measurement and PDR estimated step length and heading. It is necessary to use a multi-sensor fusion filtering method to combine the data from UWB and PDR with model-based predictions to make a more meaningful and accurate estimate of the position.

[0069] S21, mobile tag signal collection.

[0070] The tag system and the anchor system have similar hardware structures and can be flexibly set up according to needs. During the tag system experiment, the MEMS sensor is used to complete the real-time inertial information collection of the target with positioning requirements, and the UWB sensor is used to measure the distance between each anchor point in the range in real time. The data collection process ensures time synchronization and completes real-time data recording and real-time updating.

[0071] S22, calibrated position fusion positioning.

[0072] Considering that there are two different types of errors in UWB positioning and pedestrian dead reckoning: one is the measurement error caused by non-line-of-sight observation or multipath effect, and the other is the cumulative error caused by sensor drift. Many studies have shown that these errors can be effectively reduced by fusion and filtering. However, the model of the UWB / PDR fusion positioning system is nonlinear, and the linearization of the equation introduces linear errors and increases the amount of calculation. In addition, different test environments will change the propagation path of electromagnetic waves, and the non-line-of-sight errors caused by obstacles with different physical properties are not fixed, resulting in different effects of errors on UWB. The method of fixedly setting one or several variances will weaken the robustness of the positioning system and cannot meet the positioning accuracy requirements of various environments. Therefore, the present invention proposes an adaptive particle filter (APF) fusion positioning algorithm based on outlier detection.

[0073] This process is based on the above-mentioned deployment of the self-organizing network base station, and the location coordinates of the base station can be obtained, such as Figure 3 As shown. First, the tag collects accelerometer, gyroscope and magnetometer data through inertial sensors, performs preprocessing, and filters out some noise; then, the PDR algorithm is executed to detect whether a gait occurs and calculate the step length and direction of each step. If the UWB distance is measured when a gait occurs, the particle filtering process will be induced: the particle state is updated by PDR, and the state is estimated by the UWB distance. In each recursive process, the chi-square test is used to determine whether there are abnormal observations, and the observation variance is adjusted according to the test results, and then the weight calculation is performed to suppress the influence of outliers. Improve the effectiveness and stability of the positioning method. Finally, the high-precision positioning result of the tag is obtained.

[0074] Example 2

[0075] Considering the complex actual environment, in order to realize the wireless ad hoc network construction method with the minimum number of known anchor nodes, two schemes can be used to design the network construction route, such as Figure 4 As shown in the figure. The first one is that the location of an anchor node is known, and starting from this point, the route is a closed loop, such as Figure 4As shown in (a) in the figure; the second is to know the positions of two anchor nodes, starting from one point and walking to the other point, as shown in Figure 4 As shown in (b) of FIG. KA and UA represent known base stations and unknown base stations respectively.

[0076] (1) Scene and equipment construction.

[0077] During the test, WITMOTION's WT931 attitude sensor was selected as the inertial navigation system unit, and DecaWave's DWM1001 module was selected as the UWB ranging unit. The two units were assembled and integrated into a synchronous system. The system only needs to modify the software program to synchronously meet the base station requirements and tag requirements. When the tester walks in the positioning area with the device, the acceleration, angular velocity and magnetism during the movement, as well as the distance between base stations and between base stations and tags can be recorded, and different data can be synchronized to the same time system for later processing and fusion.

[0078] (2) Gait detection.

[0079] Field tests were carried out based on the WT931 attitude sensor to verify the adaptive peak detection method mentioned in the invention, and compared with the traditional peak detection method to analyze the effectiveness and stability of the proposed algorithm.

[0080] Table 1 Gait detection accuracy evaluation

[0081]

[0082] The step count results of user 1 and user 2 under different motion states are shown in Table 1. It can be seen that in the traditional peak detection method, since the acceleration threshold is set according to the acceleration collected by user 1 under normal walking conditions, the threshold is not adjusted during the whole process. The threshold is too large for the acceleration collected by user 2 at a slow speed, resulting in missing data gait; and for user 1 itself, when walking quickly, the speed is not stable, resulting in drastic changes in the acceleration value. When some gaits occur, the acceleration value is less than the threshold and cannot be detected. Only 85 steps out of 99 steps can be detected. Comprehensively considering the six cases in the table, the average accuracy of adaptive peak detection is 99.7%, which is higher than the traditional peak detection, and can correctly detect gaits under different users and different walking speeds, verifying the stability and effectiveness of the proposed method.

[0083] (3)UWB ranging calibration.

[0084] The UWB ranging accuracy is not only affected by the environment, but also by the sensor itself. In order to eliminate the ranging error caused by the sensor itself as much as possible, the UWB anchor node needs to be calibrated to achieve better ranging accuracy.

[0085] (4) Construction of self-organizing network.

[0086] This embodiment uses two networking scenarios for illustration, such as Figure 5 As shown, in scenario 1, Figure 5 As shown in (a) of the figure, a known base station is deployed, and the remaining three base stations are deployed through a closed route; in scenario 2, as shown in Figure 5 As shown in (b), a known base station is set up near the French windows at both ends, and the remaining three base stations are set up at several corners during the walking process. The actual position of the base station is measured by the rangefinder.

[0087] In scenario 1, the maximum error point of the initial positioning is base station 2, with a horizontal coordinate error of 0.94m, and the final maximum error is 0.85m. In scenario 2, base stations 1 and 5 are known, and the maximum error point of the initial positioning by PDR is base station 4, with a horizontal coordinate error of more than 2.5m, while the final positioning error is less than 0.8m.

[0088] Table 2 Network construction accuracy

[0089]

[0090] (5) Pedestrian fusion positioning.

[0091] Scene 1 is relatively small, with four base stations in the space. Although there are pillars blocking it, most areas can see through at least three base stations, mainly affected by the multipath effect. Scene 2 is relatively large, with five base stations installed, but due to the large number of corners, in many areas, only one base station can see through. In some areas, all base stations are not in sight due to the blockage of the tester's body, which is seriously affected by non-line-of-sight. Hold the positioning device horizontally in front of the chest and move it to finally get the positioning result.

[0092] The positioning accuracy of this embodiment is achieved based on existing equipment, and higher-precision positioning can be achieved later by improving the measurement accuracy of sensors.

[0093] The indoor pedestrian tracking error statistics results based on the embodiment of the channel-integrated wireless ad hoc network pedestrian positioning system are as follows: Figure 6 As shown. In two scenarios, the embodiments of the invention evaluate the positioning performance of four algorithms, LS, PDR, PF and adaptive PF, and whether base station errors are introduced. Among them, since the PDR algorithm is only related to the movement of the device itself and is not affected by base station errors; LS, PF and APF all use the distance information between the mobile device and the base station during positioning. When the base station position is inaccurate, the positioning result will also change.

[0094] In scenario 1, the ranging is relatively accurate. After the base station error is introduced, the accuracy of LS, PF, and APF decreases, and the average error increases by 0.21m at most. Among them, the APF positioning error increases by only 0.16m. Figure 6 As shown in (a) and (b) in .

[0095] In scenario 2, the positioning accuracy of LS and PF does not decrease significantly, which may be due to the large ranging error of the original ones. The base station error is introduced to compensate for part of the ranging error. The average error of APF increases by 0.09m, but the accuracy is still higher than that of the other three algorithms. Figure 6 As shown in (c) and (d) in Figure 3, the influence of the base station error on positioning is relatively small.

[0096] Therefore, the present invention adopts the above-mentioned integrated wireless self-organizing network pedestrian positioning system and its implementation method. There is no need to deploy a large number of base station facilities in the positioning area in advance, nor is there a need to build a positioning database or training model. Instead, a UWB self-organizing network of the channel is constructed with the assistance of PDR, so that the positioning anchor points are arranged in real time and the navigation and positioning network is expanded in real time, and the subsequent pedestrian trajectory positioning can be completed; the initial positioning and precise positioning of the anchor nodes are realized based on the inertial navigation / conductor network adjustment, and a integrated wireless self-organizing network of communication and navigation is constructed, which has the benchmark construction in blind environments or emergency scenarios, and realizes the navigation and positioning network expansion function with the region in complex underground scenarios without preset base stations, providing users with safe, stable and high-precision positioning services on mobile terminals in a variety of complex environments.

[0097] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A wireless ad hoc network pedestrian positioning system with integrated communication and navigation and its implementation method, characterized in that: The following steps are involved: S1. Construct a wireless ad hoc network with integrated communication and navigation, as shown below: S11, initial positioning of ad hoc network anchor nodes; S12, precise positioning of ad hoc network anchor nodes; S2, wireless ad hoc network assisted pedestrian positioning, as shown below: S21, mobile tag signal collection; S22, calibrated position fusion positioning.

2. The communication and navigation integrated wireless ad hoc network pedestrian positioning system and its implementation method according to claim 1 is characterized in that: In step S11, the pedestrian dead reckoning algorithm PDR is used in the initial positioning stage of the ad hoc network anchor node, including detecting gait, estimating step length and estimating heading, as shown below: S111, gait detection uses an adaptive peak detection method, which can automatically set the threshold according to the data situation, as shown below: Where a={a1,a2,…,a N } is the acceleration sequence without gravity effect, N is the length of the sequence; sort(·) is the sorting operation from small to large, and 85% of a sequence is taken as the standard for setting the threshold; S112, dead reckoning uses a dynamic model to estimate the step length of each step; S113, heading estimation uses a method combining three sensors, gyroscope, accelerometer and magnetometer, to obtain the current three-axis attitude through complementary filtering CF; S114, the position of each anchor node is obtained through adaptive gait detection, dynamic step length estimation and complementary filter CF heading estimation to complete the initial positioning.

3. The communication and navigation integrated wireless ad hoc network pedestrian positioning system and its implementation method according to claim 1 is characterized by: In step S12, the wire network adjustment method is used in the precise positioning stage of the self-organizing network anchor node. The updated value of the coordinate position is obtained through cyclic iteration, and the next calculation is performed until the update value variation range is less than the set threshold or the number of iterations exceeds the set value. The cycle is ended to obtain the final position estimate and complete the precise positioning of the anchor point coordinates.

4. The communication and navigation integrated wireless ad hoc network pedestrian positioning system and its implementation method according to claim 1, characterized in that: In step S21 of mobile tag signal collection, the tag system and the anchor point system have similar hardware structures and are flexibly configured according to requirements; During the tag system experiment, MEMS sensors are used to collect real-time inertial information of targets with positioning requirements. UWB sensors are used to measure the distance to each anchor point within the range in real time. The data collection process ensures time synchronization and completes real-time data recording and updating.

5. The communication and navigation integrated wireless ad hoc network pedestrian positioning system and its implementation method according to claim 1, characterized in that: Step S22 is to calibrate the position and fuse the positioning, which specifically includes the following steps: S221, first, the tag collects the accelerometer, gyroscope and magnetometer counter through the inertial sensor for preprocessing and filters out some noise; S222. Then, execute the PDR algorithm to detect whether gait occurs, and calculate the length and direction of each step.

6. The communication and navigation integrated wireless ad hoc network pedestrian positioning system and its implementation method according to claim 5, characterized in that: In step S222, when a gait occurs and the UWB distance is measured, the particle filtering process is induced: First, the particle state is updated by PDR and the state is estimated by UWB distance; Then, in each recursive process, the chi-square test is used to determine whether there are abnormal observations, and the observation variance is adjusted according to the test results; Finally, weight calculation is performed to suppress the influence of outliers.

7. The communication and navigation integrated wireless ad hoc network pedestrian positioning system and its implementation method according to claim 1, characterized in that: The process of building the integrated wireless ad hoc network in step S1 is divided into two stages: scene layout and algorithm processing.

8. The communication and navigation integrated wireless ad hoc network pedestrian positioning system and its implementation method according to claim 7 is characterized in that: In the scene layout stage, the specific steps are as follows: First, place an anchor node in an area close to the outdoors. The location of the anchor node is easy to measure and is recorded as the known base station KA. Then, carry IMU and several base stations from one of the KA points and deploy base stations during the walking process. Since the positions of these base stations are not known in advance, they are recorded as unknown base stations UA; Finally, anchor nodes can be placed at every corner, and it is sufficient to ensure line of sight with adjacent base stations. Once the base stations are deployed, the entire positioning network can be built by walking to another known base station.

9. The communication and navigation integrated wireless ad hoc network pedestrian positioning system and its implementation method according to claim 7, characterized in that: The algorithm processing stage needs to determine the coordinate information of the unknown base station, which is divided into two processes. The specific steps are as follows: First, the gait is detected based on the sensor data collected by the IMU, and the heading and step length are calculated to implement the pedestrian dead reckoning algorithm and perform the initial positioning of each unknown base station; Then, according to the ranging information between adjacent base stations and the heading angle at each base station, the traverse network adjustment based on the ranging and heading information constraints is implemented to obtain the precise positioning result of the base station.

Images