A method and device for target detection, identification and tracking based on wireless signals under multi-base station cooperation

Abstract

The application discloses a kind of multi-base station cooperation under the target detection identification tracking method and device based on wireless signal, comprising: S1: third party service provider sends request to core network, policy control module formulates strategy, session management module constructs session and synchronizes;S2: selected base station group synchronously emits signal scanning, data is acquired and sent to user plane and perception module;S3: perception module fuses data, detects and identifies target, tracks target and uploads result to core network, then to service provider;S4: judge whether service provider requests to close business, otherwise calculate signal-to-noise ratio and detection times;If it does not meet the condition, return S2;If meet, calculate average speed, determine perception times threshold;Repeat S2 to S4 until threshold is reached, stop perception;The application breaks through the bottleneck that single base station perception resource is limited by sharing the perception result of multiple base stations, realizes high-precision, high-reliability target detection tracking performance.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of information and communication technology, in particular relates to a target detection and identification tracking method and device based on wireless signals under multi-base station cooperation. BACKGROUND

[0002] With the rise of smart cities, intelligent transportation and unmanned factories, the demand for infrastructure with wireless sensing capabilities is increasingly urgent. In this context, benefiting from the rapid development of 5G-A and 6G technology, widely deployed communication base stations are gradually integrating wireless sensing functions, sensing the surrounding environment and completing sensing tasks by transmitting communication signals. At the same time, using existing communication base stations for environmental sensing also greatly reduces the deployment cost of additional sensing devices. For complex scenarios such as smart cities and intelligent transportation, communication base stations performing wireless sensing tasks need to have high-precision and high-reliability sensing capabilities to ensure the robustness of the sensing task. However, compared to radar sensing, wireless signal-based sensing has less time-frequency resources allocated to sensing tasks; on the other hand, the frequency band of communication signals is low, the spectrum environment is complex, and the signal-to-noise ratio is low. This makes it difficult for the system to accurately and reliably complete sensing tasks such as target detection and target parameter estimation. For the above reasons, using single-base station communication signals for sensing cannot achieve sufficient accuracy and stability, which requires multi-base station cooperative sensing to enhance. SUMMARY

[0003] The purpose of the present application is to provide a target detection and identification tracking method and device based on wireless signals under multi-base station cooperation to solve the problems existing in the prior art.

[0004] To achieve the above purpose, the present application provides a target detection and identification tracking method based on wireless signals under multi-base station cooperation, comprising:

[0005] Step 1: sending a target detection and identification tracking service request to the core network through a third-party service provider, formulating a target type and area, then first formulating a service-related strategy according to the target detection and identification tracking service request using a policy control module, the service-related strategy including a tracking strategy, a base station group selection strategy and a target identification strategy, then constructing a target detection and identification tracking session and performing multi-base station synchronization using a session management module, and finally selecting a base station group to perform a sensing task based on the base station group selection strategy in the policy control module;

[0006] Step 2: synchronously transmitting communication signals for sensing scanning using the base station group, obtaining sensing data, and packaging and sending the sensing data to a user plane module and a sensing module;

[0007] Step 3: data fusion and target detection and identification are performed on the perception data by using the perception module, if no target is detected, return to step 2; if a target point cloud cluster is detected, target tracking is performed on the target corresponding to the detected target point cloud cluster, and the target state data of the target corresponding to the target point cloud cluster is taken as the perception result, the perception result is stored and uploaded, the perception of the current perception round is completed, and the perception result corresponding to the current round of perception is uploaded to the third party service provider through the core network;

[0008] Step 4: it is judged whether the third party service provider sends a service closing request to the core network, if yes, the core network ends the target detection and identification tracking session; if no, the average signal-to-noise ratio of the target point cloud cluster detected in the first set of perception round numbers is calculated, if the average signal-to-noise ratio is less than the preset signal-to-noise ratio threshold or the number of targets detected in the first set of perception round numbers is less than half, return to step 2, if the average signal-to-noise ratio is greater than the preset signal-to-noise ratio threshold, execute the subsequent steps;

[0009] When the perception number reaches the preset decision period, the average speed of the target in the second set of perception round numbers is calculated, and the perception number threshold is determined based on the calculated average speed;

[0010] Steps 2 to 4 are repeated until the current perception round reaches the perception number threshold, and the perception is stopped.

[0011] Optionally, the step 1 further comprises: after the third party service provider sends a target detection and identification tracking service request to the core network, the core network performs identity verification on the third party service provider through a unified data management module, and after the identity verification is passed, the core network sends the target detection and identification tracking service to a policy control module for policy selection.

[0012] Optionally, the process of formulating the related policy specifically comprises:

[0013] The core network sends the target detection and identification tracking service request to the policy control module for policy selection, obtains the service related policy, and issues the service related policy to the user plane module, the session management module and the perception module.

[0014] Optionally, the step 1 further comprises: the session management module constructs a target detection and identification tracking session, and the connection of the third party service side, the base station side and the user plane module is completed through the session management module.

[0015] Optionally, the step 2 specifically comprises:

[0016] If the current perception round is the first perception, each base station in the base station group synchronously transmits a communication signal to the specified area for perception scanning;

[0017] If the current sensing round is not the first sensing, each base station in the base station group synchronously transmits a communication signal to a target area for sensing scanning.

[0018] Optionally, the sensing data acquisition process specifically comprises:

[0019] Each base station in the base station group synchronously transmits a communication signal for sensing scanning, and acquires echo data; the corresponding sensing information preprocessing module of each base station performs data cleaning, two-dimensional fast Fourier transform, constant false alarm detection and multiple signal classification on the echo data, obtains sensing data, and sends the sensing data to the user plane module and the sensing module after packaging; the sensing data comprises distance data, pitch angle data, azimuth angle data and four-dimensional point cloud graph data of radial velocity.

[0020] Optionally, the communication signal is an orthogonal frequency division multiplexing signal.

[0021] Optionally, the sensing data is subjected to data fusion and target detection and identification, specifically comprising:

[0022] The point cloud graphs of the base stations are fused through the sensing module, and target identification is performed on the fused point cloud graphs based on a clustering algorithm and a vector machine classification.

[0023] Optionally, the target corresponding to the detected target point cloud cluster is subjected to target tracking, specifically comprising: after detecting the target point cloud cluster, the centroid of the target point cloud cluster is subjected to initial state estimation, target state data is obtained, and an extended Kalman algorithm is used to perform state prediction and state update on the target.

[0024] A target detection, identification and tracking device based on wireless signals under multi-base station cooperation, applied to the target detection, identification and tracking method based on wireless signals under multi-base station cooperation, comprising a core network, the core network comprising a session management module and a unified data management module, a user plane module, a policy control module and a sensing module in communication connection with the session management module;

[0025] The policy control module is also in communication connection with the sensing module;

[0026] The session management module is also in communication connection with a third-party service provider;

[0027] The user plane module is also in communication connection with a plurality of base stations, and each base station is provided with a sensing information preprocessing module.

[0028] The technical effect of the present application is:

[0029] The present application effectively reduces measurement error and overcomes the problem of obstacle shielding by fusing point cloud graphs from different base stations, thereby improving target tracking accuracy and stability.

[0030] The application can effectively improve the performance of the detection, identification and tracking task by evaluating the geographical position, coverage capacity and current network load of each base station, and selecting a group of base stations most suitable for performing the detection, identification and tracking task according to the region or target position.

[0031] The application can flexibly adjust the time of the group of base stations reselection according to the target speed information and target perceived signal-to-noise ratio, and can fully realize the trade-off between the utilization of computing resources and the detection and tracking performance. BRIEF DESCRIPTION OF DRAWINGS

[0032] The accompanying drawings, which form a part of the present application, are intended to provide further understanding of the present application and are incorporated herein for a description of the present application. The use of the illustrations in the specification together with the detailed description is to explain embodiments of the present application and is not intended to limit the present application. In the drawings:

[0033] Figure 1 The scene schematic diagram in the embodiment of the application;

[0034] Figure 2 The flow chart of the target detection, identification and tracking method in the embodiment of the application;

[0035] Figure 3 The structure diagram of the target detection, identification and tracking device in the embodiment of the application. DETAILED DESCRIPTION

[0036] The detailed description of the various exemplary embodiments of the present application is not to be considered as limiting the present application, but is to be understood as a more detailed description of some aspects, characteristics and embodiments of the present application.

[0037] It should be understood that the terms described in the present application are only for describing the specific embodiments, and are not used to limit the present application. In addition, for the numerical range in the present application, it should be understood that each intermediate value between the upper limit and the lower limit of the range is also specifically disclosed. Each smaller range between any stated value or intermediate value in the stated range and any other stated value or intermediate value in the stated range is also included in the present application. The upper limit and the lower limit of these smaller ranges can be independently included or excluded from the range.

[0038] As for "include", "comprise", "have", "contain", and the like used herein, they are all open terms, that is, they mean to include but not limited to.

[0039] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0040] Embodiment one

[0041] As Figure 1 -Figure 3 As shown, the embodiment provides a target detection and identification tracking method based on wireless signals under multi-base station cooperation, comprising:

[0042] Step 1: sending a target detection and identification tracking service request to the core network through a third-party service provider, formulating a target type and a region, then first formulating a service-related strategy according to the target detection and identification tracking service request by using a policy control module, the service-related strategy including a tracking strategy, a base station group selection strategy and a target identification strategy, then constructing a target detection and identification tracking session and performing multi-base station synchronization by using a session management module, and finally selecting a base station group to perform a perception task based on the base station group selection strategy in the policy control module;

[0043] Step 2: synchronously transmitting a communication signal by using the base station group to perform a perception scan, obtaining perception data, and sending the perception data to a user plane module and a perception module after packaging;

[0044] Step 3: using the perception module to perform data fusion and target detection and identification on the perception data, if no target is detected, returning to step 2; if a target point cloud cluster is detected, performing target tracking on the target corresponding to the detected target point cloud cluster, and storing and uploading target state data of the target corresponding to the target point cloud cluster as a perception result, completing the perception of the current perception round, and uploading the perception result corresponding to the current round of perception to the third-party service provider through the core network;

[0045] Step 4: determining whether the third-party service provider sends a service closing request to the core network, if yes, the core network ends the target detection and identification tracking session; if not, calculating an average signal-to-noise ratio of the target point cloud cluster detected in a first set number of perception rounds, if the average signal-to-noise ratio is less than a preset signal-to-noise ratio threshold or the number of detected targets in the first set number of perception rounds is less than half, returning to step 2, if the average signal-to-noise ratio is greater than the preset signal-to-noise ratio threshold, performing a subsequent step;

[0046] When the perception number reaches a preset decision period, calculating an average speed of the target in a second set number of perception rounds, and determining a perception number threshold based on the calculated average speed;

[0047] Repeating steps 2 to 4 until the current perception round reaches the perception number threshold, and stopping perception.

[0048] A target detection and identification tracking device based on wireless signals under multi-base station cooperation, applied to the target detection and identification tracking method based on wireless signals under multi-base station cooperation, comprising a core network, the core network comprising a session management module and a unified data management module, a user plane module, a policy control module and a perception module in communication connection with the session management module;

[0049] The policy control module is also in communication connection with the perception module;

[0050] The session management module is also in communication connection with a third-party service provider;

[0051] The user plane module is also in communication connection with a plurality of base stations, and each of the base stations is provided with a perception information preprocessing module.

[0052] In a single-base-station target detection and tracking scene based on wireless signals, due to the small amount of time-frequency resources allocated to the perception task and the complex spectrum environment in which the frequency band is located, the signal-to-noise ratio of the perception signal will be low, which will directly lead to a decrease in the accuracy of target detection and tracking. To solve the above problem, the embodiment provides a target detection and tracking method and device based on wireless signals in a multi-base-station cooperation mode, which breaks through the bottleneck of limited perception resources of a single base station by sharing the perception results of multiple base stations, and realizes high-precision and high-reliability target detection and tracking performance.

[0053] The specific implementation process of the embodiment is as follows:

[0054] Step 1: The core network triggers a target detection and tracking session, the policy control module decides the tracking strategy and the base station group selection strategy, and sends them to the corresponding modules. Each module executes the corresponding strategy.

[0055] Step 1.1: When the third-party service provider needs to perform a detection and recognition tracking task for a specified type of target in a specified area, it sends a target detection and recognition tracking service request to the core network. The core network triggers a target detection and recognition tracking session according to the service request. After the tracking service request is sent, the core network will start identity verification for the third-party service provider through the unified data management module.

[0056] Step 1.2: The detection and recognition tracking service request of the third-party service provider is transmitted to the policy control module by the core network. The policy control module decides the tracking strategy, the base station group selection strategy, and the target recognition strategy. Then the corresponding strategies are sent to the corresponding modules, including the user plane module, the session management module, and the perception module.

[0057] Step 1.3: The session management module establishes the session flow of target detection and recognition tracking according to the session module strategy sent by the policy control module in step 1.2, completes the connection of the third-party service side, the base station side, and the user plane module, completes the data path, and performs multi-base-station synchronization.

[0058] Step 1.4 The strategy selection module selects the strategy according to the base station group. If it is the first time to perform base station group selection, the strategy selection module selects the base station group most suitable for performing the detection and identification tracking task based on the specified area location and the geographic location, coverage capability and current network load information of each base station. If it is not the first time, the strategy selection module selects the base station group most suitable for performing the detection and identification tracking task based on the target location and the geographic location, coverage capability and current network load information of each base station.

[0059] Step 2 At each time of sensing, each base station transmits a communication signal to the environment, and the sensing information preprocessing module at the base station locally processes the echo signal to obtain a point cloud map.

[0060] Step 2.1 At each time of sensing, the selected multiple base stations synchronously scan the specified area and transmit orthogonal frequency division multiplexing signals. At the same time, the sensing receiver is used to receive the echo signal.

[0061] Step 2.2 At each time of sensing, after the sensing receiver of each base station obtains the echo signal, the sensing information preprocessing module at the base station pre-processes the sensing signal to obtain a four-dimensional point cloud map of distance, pitch angle, azimuth angle and radial velocity.

[0062] Step 2.3 The sensing information preprocessing module sends the encapsulated and packaged data of the echo information sensed by the base station side after preprocessing to the user plane module, and then uploads it to the sensing module.

[0063] Step 3 The sensing module fuses the point cloud maps of the base stations and performs target identification and target tracking on the fused point cloud map.

[0064] Step 3.1 The sensing module performs sensing data fusion and target identification after obtaining the point cloud maps of the base stations. If a point cloud cluster is identified as a specified type of target, the subsequent steps are continued. Otherwise, jump to step 2 for the next sensing.

[0065] Step 3.2 After obtaining the fused point cloud map, the sensing data processing module starts target tracking. Then, the sensing module stores all the data of this sensing and uploads the target state as the sensing result to the core network.

[0066] Step 3.3 After each sensing is completed, the core network returns the sensing result to the third-party service provider.

[0067] Step 4 The core network performs base station group update judgment and session closing judgment.

[0068] Step 4.1 It is judged whether the third-party service provider sends a service closing request to the core network. If it has been sent, the core network ends the target detection and identification tracking session and completes the target detection and identification tracking based on the wireless signal. If it has not been sent, the subsequent steps are continued.

[0069] Step 4.2 calculates the average signal-to-noise ratio of the tracking target point cloud cluster in the last 10 times of sensing. If the average signal-to-noise ratio is less than a set threshold, or more than 5 times of sensing does not detect the target, jump to step 1.4 to reselect the base station group. Otherwise, continue to execute the subsequent steps.

[0070] Step 4.3 calculates the average speed of the target in the last 10 times of sensing after every K times of sensing. According to this, the required sensing times N for the subsequent base station group selection is determined.

[0071] Step 4.4 judges whether the sensing times n reaches N. If it reaches, jump to step 1.4 to reselect the base station group. If it does not reach, jump to step 2 to perform the next sensing.

[0072] Based on the above method, the embodiment proposes a target detection and recognition tracking method and device based on wireless signal under multi-base station cooperation, which includes a session management module, a unified data management module, a user plane module, a policy control module, a sensing information preprocessing module and a sensing module. The session management module is used to provide session management of third-party service providers and selection control of the user plane module, etc.; the unified data management module is used to realize user subscription management, access authorization, authentication information, etc.; the user plane module is responsible for providing message forwarding, processing, session anchor point, etc.; the policy control module is responsible for base station group selection strategy, target recognition strategy and target tracking strategy, and the corresponding strategies of each module are issued to each module; the sensing information preprocessing module is responsible for preprocessing the echo signal at the base station to obtain a point cloud map; and the sensing module is used to fuse the point cloud maps of multiple base stations and execute target recognition and target tracking.

[0073] The embodiment effectively reduces measurement error and overcomes the problem of obstacle shielding by fusing point cloud maps from different base stations, thereby improving target tracking accuracy and stability.

[0074] The embodiment can effectively improve the performance of the detection and recognition tracking task by evaluating the geographical position, coverage capacity and current network load of each base station, and selecting the most suitable base station group to perform the detection and recognition tracking task in combination with the region or target position.

[0075] The embodiment flexibly adjusts the time of base station group reselection according to target speed information and target sensing signal-to-noise ratio, and can fully realize the trade-off between computing resource utilization and detection and tracking performance.

[0076] The technical scheme of the specific application example 1 of the embodiment is as follows: when the specified target type is a road vehicle, since its speed is fast, the decision period K needs to be set to 20, and at this time, the value is set to be small to ensure that the appropriate base station group is selected in time to perform the detection and recognition tracking task on the target. The specific steps are as follows.

[0077] Step 1. The core network triggers a target detection tracking session, the policy control module decides the tracking policy and the base station group selection policy, and issues them to the corresponding modules. Each module executes the corresponding policy.

[0078] Step 1.1 When the third-party service provider needs to perform a detection and identification tracking task for road vehicles in a specified area, it sends a target detection and identification tracking service request to the core network. The core network triggers a target detection and identification tracking session according to the service request. After sending the tracking service request, the core network will start the identity verification of the third-party service provider through the unified data management module.

[0079] Step 1.2 The detection and identification tracking service request of the third-party service provider is transmitted by the core network to the policy control module. The policy control module decides the tracking policy, the base station group selection policy and the target identification policy. Then the corresponding policies are issued to the corresponding modules, including the user plane module, the session management module and the perception module.

[0080] Step 1.3 The session management module establishes the session flow of target detection and identification tracking according to the session module policy issued by the policy control module in step 1.2, completes the connection of the third-party service side, the base station side and the user plane module, completes the data path, and performs multi-base station synchronization.

[0081] Step 1.4 The policy selection module selects the base station group according to the base station group selection policy. If it is the first time to perform base station group selection, the policy selection module will select the base station group most suitable for performing the detection and identification tracking task based on the specified area location and the geographical location, coverage capacity and current network load information of each base station. If it is not the first time, the policy selection module will select the base station group most suitable for performing the detection and identification tracking task based on the target location and the geographical location, coverage capacity and current network load information of each base station.

[0082] Step 2 At each perception, each base station transmits a communication signal to the environment, and the perception information preprocessing module at the base station locally processes the echo signal to obtain a point cloud map.

[0083] Step 2.1 At each perception, the selected multiple base stations synchronously scan the specified area and transmit orthogonal frequency division multiplexing signals. At the same time, the perception receiver is used to receive the echo signal.

[0084] Step 2.2 At each perception, after the perception receiver of each base station obtains the echo signal, the perception information preprocessing module at the base station will preprocess the perception signal to obtain a four-dimensional point cloud map of distance, pitch angle, azimuth angle and radial velocity.

[0085] Step 2.3 The perception information preprocessing module sends the encapsulated and packaged data of the echo information preprocessed by the base station side to the user plane module, and then uploads it to the perception module.

[0086] Step 3. The perception module fuses the point cloud maps of the base stations and performs target recognition and target tracking on the fused point cloud map.

[0087] Step 3.1. After obtaining the point cloud maps of the base stations, the perception module performs perception data fusion and target recognition. If a point cloud cluster is identified as a target of a specified type, the subsequent steps are continued. Otherwise, the next perception is performed by jumping to Step 2.

[0088] Step 3.2. After obtaining the fused point cloud map, the perception data processing module begins target tracking. Then, the perception module stores all the data of this perception and uploads the target state as the perception result to the core network.

[0089] Step 3.3. After each perception, the core network returns the perception result to the third-party service provider.

[0090] Step 4. The core network performs base station group update judgment and session closing judgment.

[0091] Step 4.1. It is judged whether the third-party service provider sends a service closing request to the core network. If it has been sent, the core network ends the target detection and tracking session, and completes the target detection and tracking based on wireless signals. If not, the subsequent steps are continued.

[0092] Step 4.2. The average signal-to-noise ratio of the tracked target point cloud cluster in the last 10 perceptions is calculated. If the average signal-to-noise ratio is less than a set threshold, or the target is not detected for more than 5 times, the base station group selection is re-performed by jumping to Step 1.4. Otherwise, the subsequent steps are continued.

[0093] Step 4.3. The average speed of the target in the last 10 perceptions is calculated after every 20 perceptions. According to this, the required number of perceptions N for the selection of the next base station group is determined.

[0094] Step 4.4. It is judged whether the number of perceptions n reaches N. If it does, the base station group selection is re-performed by jumping to Step 1.4. If it does not, the next perception is performed by jumping to Step 2.

[0095] The technical scheme of the specific application example 2 of the embodiment is as follows: when the specified target type is a pedestrian, the decision period K needs to be set to 50 because the speed of the pedestrian is slow. At this time, the value is set to be large to ensure that the network computing resources are reduced as much as possible on the premise of timely selecting a suitable base station group. The specific steps are as follows.

[0096] Step 1. The core network triggers a target detection and tracking session. The policy control module determines the tracking strategy and the base station group selection strategy, and issues them to the corresponding modules. Each module executes the corresponding strategy.

[0097] Step 1.1 When the third-party service provider needs to perform pedestrian detection and tracking in a specified area, the third-party service provider sends a target detection and tracking service request to the core network. The core network triggers a target detection and tracking session according to the service request. After sending the tracking service request, the core network initiates an identity verification for the third-party service provider through the unified data management module.

[0098] Step 1.2 The detection and tracking service request of the third-party service provider is transmitted to the policy control module by the core network. The policy control module determines the tracking policy, the base station group selection policy, and the target identification policy. Then the corresponding policies are issued to the corresponding modules, including the user plane module, the session management module, and the perception module.

[0099] Step 1.3 The session management module establishes the session flow of target detection and tracking according to the session module policy issued by the policy control module in step 1.2, completes the connection of the third-party service side, the base station side, and the user plane module, completes the data path, and performs multi-base station synchronization.

[0100] Step 1.4 The policy selection module selects the base station group according to the base station group selection policy. If it is the first time to perform base station group selection, the policy selection module selects the base station group that is most suitable for performing the detection and tracking task based on the location of the specified area and the geographic location, coverage capability, and current network load information of each base station. If it is not the first time, the policy selection module selects the base station group that is most suitable for performing the detection and tracking task based on the target location and the geographic location, coverage capability, and current network load information of each base station.

[0101] Step 2 In each perception, each base station transmits a communication signal to the environment, and the perception information preprocessing module at the base station locally processes the echo signal to obtain a point cloud map.

[0102] Step 2.1 In each perception, the selected multiple base stations synchronously scan the specified area and transmit orthogonal frequency division multiplexing signals. At the same time, the perception receiver is used to receive the echo signal.

[0103] Step 2.2 In each perception, after the perception receiver of each base station obtains the echo signal, the perception information preprocessing module at the base station pre-processes the perception signal to obtain a four-dimensional point cloud map of distance, pitch angle, azimuth angle, and radial velocity.

[0104] Step 2.3 The perception information preprocessing module sends the encapsulated and packaged data of the pre-processed echo information perceived by the base station side to the user plane module, and then uploads it to the perception module.

[0105] Step 3 The perception module fuses the point cloud maps of each base station and performs target identification and target tracking on the fused point cloud map.

[0106] Step 3.1 After the point cloud map of the base stations is obtained, the perception module performs perception data fusion and target recognition. If a point cloud cluster is identified as a target of a specified type, the subsequent steps are performed. Otherwise, the next perception is performed by jumping to Step 2.

[0107] Step 3.2 After the fused point cloud map is obtained, the perception data processing module starts target tracking. Then, the perception module stores all the data of this perception and uploads the target state as the perception result to the core network.

[0108] Step 3.3 After each perception, the core network returns the perception result to the third-party service provider.

[0109] Step 4 The core network performs base station group update judgment and session closing judgment.

[0110] Step 4.1 It is judged whether the third-party service provider sends a service closing request to the core network. If yes, the core network ends the target detection and recognition tracking session, and completes the target detection and recognition tracking based on wireless signals. If not, the subsequent steps are performed.

[0111] Step 4.2 The average signal-to-noise ratio of the tracked target point cloud cluster in the last 10 perceptions is calculated. If the average signal-to-noise ratio is less than a set threshold, or the target is not detected for more than 5 times, the base station group selection is performed again by jumping to Step 1.4. Otherwise, the subsequent steps are performed.

[0112] Step 4.3 The average speed of the target in the last 10 perceptions is calculated after every 50 perceptions. According to the average speed, the required number of perceptions N for the selection of the next base station group is determined.

[0113] Step 4.4 It is judged whether the number of perceptions n reaches N. If yes, the base station group selection is performed again by jumping to Step 1.4. If not, the next perception is performed by jumping to Step 2.

[0114] The above is only the preferred specific implementation of the present application, but the protection scope of the present application is not limited to this. Any changes or replacements within the technical range disclosed in the present application can be easily thought by those skilled in the art, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for target detection, recognition and tracking based on wireless signals in a multi-base-station cooperative manner, characterized in that, The method comprises the following steps: Step 1: sending a target detection and identification tracking service request to a core network through a third-party service provider, formulating a target type and a region, then first formulating a service-related strategy according to the target detection and identification tracking service request by using a policy control module, the service-related strategy comprising a tracking strategy, a base station group selection strategy and a target identification strategy, then constructing a target detection and identification tracking session and performing multi-base station synchronization by using a session management module, and finally selecting a base station group to perform a sensing task based on the base station group selection strategy in the policy control module; Step 2: synchronously transmitting a communication signal by using the base station group to perform sensing scanning, obtaining sensing data, and sending the sensing data to a user plane module and a sensing module after packaging; Step 3: performing data fusion and target detection and identification on the sensing data by using the sensing module, returning to step 2 if no target is detected, performing target tracking on a target corresponding to a detected target point cloud cluster if the target point cloud cluster is detected, and storing and uploading target state data of the target corresponding to the target point cloud cluster as a sensing result, completing sensing of a current sensing round, and uploading the sensing result corresponding to the current round of sensing to the third-party service provider through the core network; Step 4: determining whether a service closing request is sent by the third-party service provider to the core network, and ending the target detection and identification tracking session if the service closing request is sent; if not, calculating an average signal-to-noise ratio of the target point cloud cluster detected in a first set number of sensing rounds, returning to step 2 if the average signal-to-noise ratio is less than a preset signal-to-noise ratio threshold or the number of targets detected in the first set number of sensing rounds is less than half, and executing a subsequent step if the average signal-to-noise ratio is greater than the preset signal-to-noise ratio threshold; when the sensing number reaches a preset decision period, calculating an average speed of the target in a second set number of sensing rounds, and determining a sensing number threshold based on the calculated average speed; repeating steps 2 to 4 until the current sensing round reaches the sensing number threshold, and stopping sensing.

2. The method of claim 1, wherein, The step 1 further comprises: after the third-party service provider sends the target detection and identification tracking service request to the core network, the core network performs identity authentication on the third-party service provider through a unified data management module, and sends the target detection and identification tracking service to the policy control module for policy selection after the identity authentication is passed.

3. The method of claim 1, wherein, The process of formulating the related strategy specifically comprises: The core network sends the target detection and identification tracking service request to the policy control module for policy selection, obtains a service-related strategy, and issues the service-related strategy to a user plane module, a session management module and a sensing module.

4. The method of claim 1, wherein, The step 1 further comprises: the session management module constructs a target detection and identification tracking session, and completes connection of a third-party service side, a base station side and a user plane module through the session management module.

5. The method of claim 1, wherein, The step 2 specifically comprises: If the current sensing round is the first sensing, each base station in the base station group synchronously transmits a communication signal to a specified region to perform sensing scanning. If the current sensing round is not the first sensing, each base station in the base station group synchronously transmits a communication signal to the target area for sensing scanning.

6. The method of claim 1, wherein, The acquisition process of the sensing data specifically includes: Each base station in the base station group synchronously transmits a communication signal for sensing scanning, and acquires echo data. The corresponding sensing information preprocessing module of each base station performs data cleaning, two-dimensional fast Fourier transform, constant false alarm detection and multiple signal classification on the echo data, obtains sensing data, and sends the sensing data to the user plane module and the sensing module after packaging; the sensing data includes distance data, pitch angle data, azimuth angle data and four-dimensional point cloud graph data of radial velocity.

7. The method of claim 1, wherein, The communication signal is an orthogonal frequency division multiplexing signal.

8. The method of claim 1, wherein, The data fusion and target detection and identification of the sensing data specifically include: The point cloud graphs of the base stations are fused through the sensing module, and target identification is performed on the fused point cloud graphs based on a clustering algorithm and a vector machine classification.

9. The method of claim 1, wherein, The target tracking of the target corresponding to the detected target point cloud cluster specifically includes: after detecting the target point cloud cluster, performing initial state estimation on the centroid of the target point cloud cluster to obtain target state data, and using an extended Kalman algorithm to perform state prediction and state update on the target.

10. A device for target detection, identification and tracking based on wireless signals under multi-base-station cooperation, applied to the method for target detection, identification and tracking based on wireless signals under multi-base-station cooperation according to any one of claims 1-9, characterized in that, The core network includes a session management module; The core network includes a session management module, a unified data management module, a user plane module, a policy control module and a sensing module which are in communication connection; the policy control module is also in communication connection with the sensing module; the session management module is also in communication connection with a third-party service provider; The user plane module is also in communication connection with a plurality of base stations, and each base station is provided with a sensing information preprocessing module.

Images