Wireless network positioning method, server and system

Abstract

The invention discloses a wireless network positioning method, server and system, and relates to the technical field of wireless communication and positioning navigation. The method comprises the following steps: receiving a signal sent by a to-be-positioned target node through a plurality of positioning reference nodes and extracting an original positioning parameter sequence; then, improved Z-score exception elimination based on median and median absolute deviation and adaptive attenuation coefficient smoothing based on average inter-frame hopping are carried out on the sequence, and a stable estimated value and a measurement variance are output; and finally, according to the number of the effective positioning reference nodes, adaptively switching a CHAN algorithm or a weighted least square method to carry out position calculation. The improved Z-score filtering and dynamic weight fusion technology is adopted, the defects in the prior art are overcome, the problem that a fixed algorithm is poor in robustness when the reference node topology change is positioned is solved through an algorithm self-adaptive switching mechanism, the precision and adaptability are remarkably improved in a complex scene, and meanwhile the deployment and maintenance cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of wireless communication and positioning navigation technology, specifically to wireless network positioning methods, servers, and systems. Background Technology

[0002] Wireless positioning technology, as a core support of modern information society, determines the target location by extracting positioning-related parameters (such as time of arrival and signal strength) from the propagation of radio waves and using geometric measurement or statistical derivation methods. This technology is widely used in public safety, the Internet of Things, and intelligent transportation, and its basic process includes three stages: signal feature extraction, positioning parameter estimation, and spatial coordinate calculation. Positioning technology based on communication networks reuses existing network infrastructure (such as base stations and access points), offering coverage advantages in indoor or urban canyon environments where satellite signals are blocked, and has become a key component of the integrated "sensing-communication" architecture in 5G / 6G communication systems. For example, early cellular positioning (Cell-ID) roughly estimated location using base station identification, while modern 4G / 5G high-precision positioning utilizes multi-base station collaborative measurement to improve accuracy. However, with the increasing complexity of application scenarios (such as dynamic environments and the coexistence of multiple network standards), existing technologies have revealed many limitations.

[0003] At the specific solution level, Chinese invention patent application CN202511223911.5 discloses a fusion positioning method based on 5G-A base stations and UWB base stations. By optimizing the deployment location and number of base stations, it leverages the complementary advantages of 5G-A's wide coverage and UWB's high precision to reduce deployment costs. This solution first plans the base station network based on environmental information, then filters ranging signals based on signal strength, and introduces algorithms such as Kalman filtering for trajectory prediction. However, its core problem lies in the high computational complexity of filtering algorithms (such as extended Kalman filtering) and their strict dependence on preset motion models (such as constant velocity models). When the terminal makes a sudden turn or stop, it is prone to filtering divergence. Furthermore, the solution has stringent requirements for the physical layout of base stations, necessitating meticulous planning in the early stages of construction, making it difficult to adapt to already deployed heterogeneous networks. Another related approach, Chinese invention patent application CN201810076225.3, discloses an indoor positioning method based on ranging error correction. This method establishes an error compensation function model by pre-measuring environmental data and linearly corrects the ranging values ​​to suppress deviations caused by non-line-of-sight and multipath effects. However, this approach requires extensive manual sampling to construct a static prior model, which becomes ineffective when the environment changes (such as moving furniture), resulting in high maintenance costs. Furthermore, the fixed function model cannot identify sudden impulse interference, causing the corrected data to still contain outlier errors.

[0004] A deeper problem lies in the fact that existing technologies generally lack environmental adaptability: 1) At the algorithm level, a single solution strategy (such as the trilateration method) cannot dynamically respond to changes in the number of base stations (such as base station failure due to occlusion), and the solution stability is poor when base stations are sparse; 2) At the data level, reliance on static modeling or simple linear processing makes it difficult to cope with non-Gaussian noise under complex channels. These problems together result in insufficient robustness and poor deployment flexibility of positioning systems in dynamic scenarios.

[0005] Therefore, there is an urgent need for a wireless network positioning solution that can sense environmental fluctuations in real time and adaptively adjust processing strategies to break through the dependence of existing technologies on hardware deployment and static models, and achieve high-precision positioning in all scenarios. Summary of the Invention

[0006] This invention provides a wireless network positioning method, server, and system to address the problems of insufficient robustness, poor deployment flexibility, and reliance on static models in existing technologies, thereby achieving high-precision positioning across all scenarios.

[0007] To achieve the above objectives, the present invention provides the following technical solution: In one aspect, this invention protects a wireless network positioning method, comprising: The system receives wireless signals from the target node to be located through multiple positioning reference nodes and extracts the original positioning parameter sequence corresponding to the target node to be located. The original positioning parameter sequence is sequentially subjected to improved Z-score anomaly removal based on median and median absolute deviation, and adaptive attenuation coefficient smoothing based on average inter-frame jump variable, outputting stable parameter estimates and measurement variance. Based on the stable parameter estimates and measurement variance, and based on the number of positioning reference nodes on which the stable parameter estimates are generated, different geometric calculation algorithms are adaptively switched to calculate the spatial coordinates of the target node to be located.

[0008] Furthermore, the improved Z-score outlier removal based on the median and the absolute deviation of the median specifically includes: Calculate the median and median absolute deviation of the original positioning parameter sequence; Based on the median and the median absolute deviation, the improved Z-score value corresponding to each positioning parameter is calculated; The improved Z-score value is compared with a preset threshold, and positioning parameters that exceed the preset threshold are removed.

[0009] Furthermore, the improved Z-score is obtained by calculating the absolute value of the difference between the positioning parameter and the median, and comparing it with the median absolute deviation adjusted by the standardization factor.

[0010] Furthermore, the adaptive attenuation coefficient smoothing process based on the average inter-frame jump variable specifically includes: For the cleaned parameter sequence obtained after anomaly removal, calculate its average inter-frame jump variable to quantify the degree of parameter fluctuation; The adaptive attenuation coefficient is dynamically determined based on the relationship between the average inter-frame jump variable and the preset reference fluctuation threshold. The time decay weight of each parameter sample is calculated based on the adaptive decay coefficient, and the cleaned parameter sequence is then weighted and fused.

[0011] Furthermore, the adaptive attenuation coefficient is limited to a preset minimum and maximum value, and is obtained by dynamic mapping or limiting based on the ratio of the average inter-frame jump variable to the reference fluctuation threshold.

[0012] Furthermore, the adaptive switching of different geometric solution algorithms specifically includes: When the quantity is less than a preset threshold, the Chan analytical algorithm is used to calculate the position. When the quantity is greater than or equal to the preset threshold, the position is calculated using the weighted least squares method.

[0013] Furthermore, when using the weighted least squares method, an observation weight matrix is ​​constructed; wherein the observation weight matrix is ​​a diagonal matrix, and each element on its diagonal corresponds one-to-one with the stable parameter estimate involved in the solution, and each element value is the reciprocal of the sum of the corresponding measurement variance and the measurement variance of a certain reference node, which is used to characterize the confidence distribution of signals from different reference nodes in the current environment.

[0014] In another aspect, the present invention provides a server device for wireless network positioning, comprising a processor, a memory, and a communication interface, wherein: The processor and memory are connected via a system bus; The communication interface is connected to the system bus via an input / output interface; The memory includes internal memory and a non-volatile storage medium, the non-volatile storage medium being used to store the operating system, computer programs, and database; The processor is configured to execute the computer program, call the internal memory, and receive the original positioning parameter sequence through the communication interface to implement the wireless network positioning method.

[0015] In another aspect, the present invention provides a wireless network positioning system, comprising: The target node to be located is used to transmit or receive wireless signals; Multiple positioning reference nodes are used to wirelessly communicate with the target node to be positioned and to extract positioning parameters; The aforementioned server device is used to obtain positioning parameters from the positioning reference node and perform positioning calculations to output the position coordinates of the target node to be located.

[0016] In another aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned wireless network positioning method.

[0017] Technical effects of the present invention Compared with the prior art, the present invention has the following advantages: (1) Significantly improves the adaptability and robustness of the positioning system in complex dynamic environments. Addressing the shortcomings of existing technologies that rely on static pre-modeling and struggle to cope with changes in environmental layout and sudden pulse interference, this invention introduces an improved Z-score anomaly removal mechanism. Utilizing the insensitivity of the median and median absolute deviation to extreme values, it can accurately identify and filter out sudden jumps in ranging values ​​caused by multipath effects or non-line-of-sight transmission online and in real-time, ensuring the purity of the input data from the source. Simultaneously, the adaptive smoothing algorithm based on the average inter-frame jump variable can dynamically adjust the filtering weights according to the severity of environmental fluctuations. Deep smoothing is achieved in stable environments to improve accuracy, while rapid response is provided in dynamic environments to reduce lag. Thus, high-precision positioning is achieved in various unknown and variable scenarios without any prior modeling, significantly enhancing the system's generalization ability and robustness.

[0018] (2) It achieves algorithm-level adaptability, breaking through the bottleneck of rigid algorithm architecture in existing technologies. Compared with existing solutions, this invention constructs a dual adaptive mechanism based on real-time perception. At the solution level, the system seamlessly switches between the CHAN parsing algorithm (for sparse positioning reference nodes, ensuring convergence) and the weighted least squares method (for redundant positioning reference nodes, improving accuracy) based on the number of currently available effective positioning reference nodes, ensuring optimal positioning across all scenarios from the network edge to the central area. At the data processing level, a smoothing strategy that adapts to parameter fluctuations replaces the traditional fixed parameter filtering, enabling the system to autonomously adapt to different motion states of the target node to be located. This algorithm-level flexibility effectively overcomes the excessive dependence of existing technologies on specific hardware deployments, fixed motion models, and ideal positioning reference node topologies.

[0019] (3) Reduced system deployment and maintenance costs, and enhanced practical value. Since this invention does not rely on offline environmental data acquisition and modeling, nor does it require complex hybrid deployment planning of specific positioning reference nodes (such as 5G-A and UWB) to improve accuracy, it achieves "ready to use" rapid deployment capability. This not only reduces reliance on manual intervention and lowers initial construction costs, but also enables the solution to be directly applied to existing heterogeneous communication networks with unevenly distributed positioning reference nodes, greatly improving the applicability and economy of the technology in real-world scenarios.

[0020] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention.

[0021] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0022] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the overall architecture of a wireless network positioning system in an embodiment of the present invention, wherein (a) is a schematic diagram of a wireless positioning system based on Wi-Fi signals, (b) is a schematic diagram of a wireless positioning system based on 5G cellular networks, and (c) is a schematic diagram of a wireless positioning system based on Bluetooth signals. Figure 2 This is a flowchart of a wireless network positioning method according to an embodiment of the present invention; Figure 3 This is a framework diagram of the parameter extraction module in an embodiment of the present invention; Figure 4 This is a framework diagram of the multi-frame fusion filtering module in an embodiment of the present invention; Figure 5 This is a framework diagram of the adaptive position calculation module in an embodiment of the present invention; Figure 6 This is a framework diagram of a server device according to an embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0025] Example 1 Figure 2 The diagram shows a flowchart of a wireless network positioning method according to the present invention. The specific method includes: Step 1: Receive wireless signals from the target node to be located through multiple positioning reference nodes, and extract the original positioning parameter sequence corresponding to the target node to be located; Step 2: The original positioning parameter sequence is sequentially subjected to improved Z-score anomaly removal based on median and median absolute deviation and adaptive attenuation coefficient smoothing based on average inter-frame jump variable, and stable parameter estimates and measurement variance are output. Step 3: Based on the stable parameter estimates and measurement variance, and based on the number of positioning reference nodes on which the stable parameter estimates are generated, adaptively switch different geometric solution algorithms to calculate the spatial coordinates of the target node to be located.

[0026] In some embodiments of this example, before performing step 1, the following may also be included: System environment setup: This solution is applied to a wireless communication system, which contains multiple positioning reference nodes, several target nodes to be located, and a positioning server.

[0027] It should be noted that the "communication network" mentioned in this article is not limited to cellular mobile communication networks, but encompasses various wireless standards, including but not limited to 5G / 6G cellular networks, Wi-Fi, and Bluetooth communication systems. Therefore, the physical entities described in this article may have different specific forms in different application scenarios. The target node to be located (hereinafter referred to as "terminal") refers to the mobile device to be located. For example, in a cellular network, it is represented by a smartphone, a vehicle module (C-V2X), etc.; in a wireless local area network, it can be represented by a station (STA); in the Industrial Internet of Things, it can be represented by a UWB tag or an active RFID tag, etc. The target node to be located includes the following functional modules: 1) Signal receiving module: receives downlink signals from each positioning reference node; 2) Signal transmitting module: transmits specific reference signals to each positioning reference node.

[0028] Location reference node (hereinafter referred to as "base station"): refers to infrastructure whose absolute or relative position is known. For example, in 5G / 6G networks, it is a base station; in WLAN networks, it is a wireless access point (AP); in Bluetooth systems, it is a Bluetooth tag, etc. A location reference node includes the following functional modules: 1) Signal transmitting module: transmits downlink signals to the terminal; 2) Signal receiving module: receives wireless transmission signals from different target nodes; 3) Parameter extraction module: processes the received signals and extracts the location parameters corresponding to different terminals. Location server (hereinafter referred to as "server"): refers to the computing unit that executes the core solution logic. It can be the Location Management Function (LMF) in the core network, a Multi-access Edge Computing (MEC) server deployed at the network edge, or even an embedded computing unit integrated into the main base station. The location server is the computing hub of the system and mainly includes the following functional modules: 1) Parameter aggregation module: aggregates positioning parameters from multiple base stations; 2) Parameter preprocessing module: performs anomaly removal and weighted fusion on the positioning parameters; 3) Location solution module: adaptively selects the solution algorithm according to the number of base stations to obtain the final positioning result.

[0029] The system involves communication between the terminal and the base station via uplink and downlink; each base station receives transmitted signals from the terminal, extracts positioning parameters, and aggregates these parameters to the positioning server for location tracking. The system architecture is as follows: Figure 1 As shown.

[0030] Figure 1 for Example diagram of system architecture for three independent positioning technologies: 5G cellular and Bluetooth. Figure 1 (a) is based on The wireless positioning system for signals involves the target node communicating with the access point (AP) via the 802.11 protocol. The positioning data collected by the AP is aggregated by the Wireless LAN Controller (WLC) and uploaded to the positioning server. Finally, the location service is output to the user by the location-based service application (LBS APP). Figure 1 (b) is a wireless positioning system based on a 5G cellular network, in which the target node to be located is located via... The interface (New Radio Uu Interface) interacts with the base station, which then communicates with the base station via... The NextGeneration-C Interface transmits location-related data to the Access and Mobility Management Function (AMF) of the core network. The AMF then forwards the data to the LMF via the NLs Interface (NLMF Service Based Interface) to complete the location calculation. The calculation result is pushed to the Location Service (LCS) client via the Hypertext Transfer Protocol (HTTP) for use by upper-layer applications. Figure 1 (c) is a Bluetooth-based wireless positioning system. In this system, the target node communicates with the Bluetooth tag using Bluetooth Low Energy (BLE) protocol. The Bluetooth gateway collects the tag signal data and uploads it to the positioning server. Finally, the LBS APP presents the positioning result. Three technical architectures are presented independently, each corresponding to a different positioning implementation method in a different scenario. In this embodiment, the number of base stations... The maximum value is 3 or higher, with no upper limit requirement, and the algorithm can automatically adapt to the optimal algorithm when the number of base stations changes dynamically. This embodiment is also applicable to line-of-sight (LOS) transmission or mixed line-of-sight and non-line-of-sight (NLOS) transmission scenarios.

[0031] Step 1: Receive wireless signals from the target node to be located via multiple positioning reference nodes, and extract the original positioning parameter sequence corresponding to the target node. This process is implemented through a parameter extraction module, and the overall framework is as follows: Figure 3 The specific process is as follows: 1.1) in the Time slot, The transmitted signals received at each base station from different terminals can be represented as follows: : in Indicates the first The received signal corresponding to each base station .

[0032] 1.2) Assuming the environment has For the terminal to be located, for the first... One base station, which receives signals Can be regarded as The signals transmitted by each terminal are superimposed after transmission through the channel. Each base station, based on the signal characteristics of different terminals (such as specific pilot sequences, frequencies, or spreading codes), will... Separate into independent signal components corresponding to different terminals. For the first... Each base station has a corresponding received signal that can be represented as: in, Indicates the first The base station received from the first The signal components of each terminal.

[0033] 1.3) Extract positioning parameters for different terminals from each base station. Here, depending on the positioning algorithm selected by the system, different positioning parameters are extracted for each terminal, such as signal characteristic values ​​like Time of Arrival (TOA), Time Difference of Arrival (TDOA), or Received Signal Strength Indicator (RSSI).

[0034] Specifically, this embodiment does not limit the specific positioning parameter extraction algorithm. The base station extracts the first When obtaining the positioning parameters of a terminal, the following method can be used: When extracting TOA or TDOA positioning parameters, the phase received signal can be analyzed. Cross-correlation calculations are performed to extract the corresponding positioning parameters; when extracting RSSI positioning parameters, this can be achieved by... Power measurement is performed to extract parameters.

[0035] To ensure standardized input, each base station will output positioning parameters in a unified manner. (The last sentence appears to be incomplete and possibly refers to a different context.) Taking the first base station as an example, in the... One data frame, corresponding to The location parameters output for each terminal are as follows: ,in Representative at the Within the data frame, the first The data was extracted from the base station, corresponding to the [number]th base station. Positioning parameters of each terminal.

[0036] Furthermore, this system considers using Within each time frame, positioning parameters are extracted. For the first time frame... The corresponding base station obtained All positioning parameters of a terminal can be written as: .

[0037] Step 2 involves sequentially performing improved Z-score anomaly removal based on the median and median absolute deviation, and adaptive attenuation coefficient smoothing based on the average inter-frame jump variable, on the original positioning parameter sequence, outputting stable parameter estimates and measurement variance. This process is implemented through a multi-frame fusion filtering module, and its overall framework is as follows: Figure 4 As shown. Specifically includes: 2.1) The improved Z-score anomaly removal based on median and median absolute deviation specifically includes: calculating the median and median absolute deviation of the original positioning parameter sequence; calculating the improved Z-score value corresponding to each positioning parameter based on the median and the median absolute deviation; comparing the improved Z-score value with a preset threshold, and removing positioning parameters that exceed the preset threshold. The improved Z-score value is obtained by calculating the absolute value of the difference between the positioning parameter and the median, and comparing it with the median absolute deviation adjusted by a standardization factor.

[0038] In one specific embodiment, each base station employs an improved Z-score method based on the median and median absolute deviation to perform anomaly removal. By calculating the discrete statistical characteristics of the parameter sequence, it accurately identifies and removes outlier parameter values ​​that deviate from the normal distribution. The specific calculation steps are as follows: Median calculation: For the first The corresponding base station obtained The positioning parameters of each terminal across all time frames are: The median was calculated. .

[0039] Median absolute deviation calculation: Calculate the values ​​of each parameter relative to the median. Median absolute deviation between Then calculate the median of these absolute deviations. .

[0040] Improved Z-score calculation: Calculate the Z-score corresponding to each positioning parameter: Where C is the standardization factor, which is essentially the quantile of the cumulative distribution function of the standard normal distribution. Its value is... (This is the cumulative distribution function of the standard normal distribution), its function is to... Standardization provides a reliable estimate of the standard deviation under a normal distribution scenario, thus providing a unified reference standard for the Z-score determination threshold.

[0041] Anomaly detection: In the anomaly detection stage, a threshold is set. (In engineering, a value of 3.2 is typically used. This threshold corresponds to a 99.9% confidence interval under a normal distribution, balancing the accuracy of anomaly detection with the false negative rate.) If the Z-score corresponding to the localization parameter sample satisfies... If the sample location parameter fails to pass verification, it is considered an outlier and is removed, retaining only valid location parameters that pass the verification. Finally, each base station outputs... A valid positioning parameter that has passed verification is represented as follows: .

[0042] 2.2) The adaptive attenuation coefficient smoothing process based on the average inter-frame hop variable specifically includes: calculating the average inter-frame hop variable of the cleaned parameter sequence obtained after anomaly removal to quantify the degree of parameter fluctuation; dynamically determining an adaptive attenuation coefficient based on the relationship between the average inter-frame hop variable and a preset reference fluctuation threshold; calculating the time attenuation weight of each parameter sample based on the adaptive attenuation coefficient, and performing weighted fusion on the cleaned parameter sequence. The adaptive attenuation coefficient is limited to a preset minimum and maximum value, and is obtained by dynamic mapping or limiting processing based on the ratio of the average inter-frame hop variable to the reference fluctuation threshold.

[0043] To further enhance the filtering module's ability to utilize time-series data, this invention proposes a dual-dimensional weighted fusion strategy. Combining the dimensions of timeliness and environmental adaptability, it achieves a weighting strategy of "high weight for new data, low weight for old data." First, the fluctuation of parameters across multiple frames is quantified, then the adaptive attenuation coefficient is dynamically adjusted, and finally, the weights are calculated based on this coefficient. The specific steps are as follows: Quantifying parameter fluctuations: Calculating the average jump variable The following formula is used to measure the degree of fluctuation in positioning parameters: in, Indicates the first The first base station is for the first The average inter-frame hop variable of the effective positioning parameters of each terminal is used to measure the degree of fluctuation of the effective positioning parameters.

[0044] Adaptive attenuation coefficient Dynamic adjustment: using gain factor and reference fluctuation threshold The system establishes the following adaptive mapping relationship: in, For the first For the first base station The adaptive attenuation coefficient of each terminal's effective parameter sample. The minimum attenuation coefficient, The maximum attenuation coefficient, This is the gain factor, used to adjust... Sensitivity to environmental changes. The gain factor. The calculation formula is as follows: The above reference fluctuation threshold This threshold is used to characterize the maximum allowable statistical fluctuation deviation of positioning parameters per unit time for a target under normal motion conditions. This threshold is a configurable system parameter, and its specific value needs to be preset based on empirical data from the communication standard used and the target application scenario, rather than a fixed value. The setting principle is: in scenarios where the positioning parameter is distance information, for high-precision positioning systems such as Ultra-Wideband (UWB), due to their strong multipath suppression capability and small channel measurement error, It can be set to a lower range (e.g., 0.1 meters to 0.3 meters); for cellular network (such as 4G / 5G) positioning systems with relatively narrow bandwidth and greater susceptibility to environmental interference, This requires setting within a higher range (e.g., 1 meter to 3 meters) to accommodate its greater inherent measurement fluctuations. The configuration can be adjusted within or outside this typical range based on the actual system's performance indicators and scenario requirements.

[0045] Calculation of time decay weighting factor: For the first base station The terminal's first The formula for calculating the effective parameter sample time decay weighting factor is: in, , For the first For the first base station The terminal's first Time decay weighting factor for each effective parameter sample, The timestamp of the current moment. For the first The timestamp of the frame.

[0046] Weight normalization: Due to differences in the number of effective samples and timestamp distribution among base stations, directly using the time decay weight factor will lead to an imbalance in weight allocation. Therefore, weight normalization is required. The formula is as follows: in, For the first For the first base station The terminal's first The final normalized weights of each valid parameter sample, the normalized weights. satisfy .

[0047] 2.3) After obtaining the normalized weights of each sample, the weighted fusion of parameters from multiple frames can be carried out, and the measurement variance of the effective parameter samples can be calculated to provide a data accuracy reference for subsequent positioning calculations.

[0048] No. For the first base station The estimated stable measurement parameters for each terminal are: To quantify the accuracy of the estimated measurement parameters, the variance of the parameter measurements is calculated simultaneously: in, This is to measure the lower limit of variance. Finally, the estimated stable measurement parameters output by each base station are... and the corresponding parameter measurement variance This will serve as the core input data for the adaptive positioning solution module, providing solid data support for the accurate calculation of the terminal's location.

[0049] Step 3: Based on the stable parameter estimates and measurement variances, and considering the number of positioning reference nodes upon which the stable parameter estimates are generated, adaptively switch between different geometric calculation algorithms to calculate the spatial coordinates of the target node to be located. Specifically, the adaptive switching between different geometric calculation algorithms involves: when the number of reference nodes is less than a preset threshold, using the Chan analytical algorithm for position calculation; and when the number of reference nodes is greater than or equal to the preset threshold, using weighted least squares for position calculation. When using weighted least squares, a weight matrix is ​​constructed where each element is the reciprocal of the sum of the measurement variance of the corresponding reference node and the measurement variance of a specific reference node, representing the confidence distribution of signals from different reference nodes in the current environment.

[0050] This step employs an adaptive position calculation module. Figure 5 Based on the stable parameter information output by the multi-frame fusion filtering module, the target is located. This process is not limited to a specific positioning scheme and can employ time-of-arrival (TOA), time-difference-of-arrival (TDOA), angle-based positioning (AOA), etc. For the TDOA-based positioning scheme, this embodiment designs an improved algorithm. In the TDOA positioning system, for the same terminal, the raw measurement parameters reported by each base station are processed by the multi-frame fusion filtering module to obtain a stable time parameter estimate. and its measurement variance ;in The module assigns base station numbers. Based on the data quality feedback from the multi-frame fusion filtering module, this module dynamically selects high-quality base stations to participate in the calculation, and uses the measurement variance from the filtering feedback to adjust the solution weights in real time, ensuring high-precision positioning results are output even in various complex environments. , This is the output of the module.

[0051] 3.1) Constructing the Positioning Equation: To make the positioning-related time measurement data easier to understand intuitively, and to provide a unified and easy-to-operate quantitative calculation basis for subsequent solutions, after obtaining the time measurement values, a time-to-distance conversion operation is first performed. That is... Given the first The coordinates of each base station are Assuming the target terminal's location coordinates are Then the target terminal and the first The actual distance between base stations Satisfies the three-dimensional Euclidean distance formula: This technical solution is based on As a reference base station, the distance estimate output by the filtering module is used. Convert to distance difference observations : The difference between the observed value and the true distance Measurement noise exists : 3.2) Location Calculation: After constructing the positioning equation, this technical solution will calculate the location based on the number of effective base stations. Automatically select the location solution method.

[0052] 3.2.1) When the number of base stations is limited ( In this scenario, the CHAN algorithm based on high-quality base station screening is used for location calculation. This algorithm transforms the nonlinear distance difference equation into a linear form by introducing auxiliary variables, and quickly obtains the initial location estimate of the target terminal using a closed-form analytical solution. The base stations are ranked by quality, and those with drastic fluctuations are removed, leaving only the base stations with the highest weights to ensure the reliability of the non-iterative analytical solution.

[0053] 3.2.2) When the number of base stations is sufficient ( When calculating the position, linear least squares based on composite weights is used. The core of this approach lies in fully utilizing redundant observation information by using the adaptive weighting matrix generated based on the filter variance. By introducing regular equations, high-precision observation data takes the lead in the solution process, thereby obtaining more accurate position estimates than ordinary least squares method in noisy environments.

[0054] The innovation of this scheme under redundant observation conditions lies in the construction of a composite weight matrix. Under the TDOA system, the weights not only include the measurement variance of a single base station, but also the correlation error introduced by the "reference base station." This design ensures that if the environment of either the reference base station or the current base station deteriorates, the weights will be adaptively adjusted downwards, which is more reasonable than a single weight.

[0055] Using the observation error covariance matrix Based on this, the inverse matrix is ​​used as the weights. The smaller the error of the observed data, the larger the weight; the larger the error, the smaller the weight. Specifically, it is defined as follows: in, .

[0056] By constructing an adaptive weighting matrix It allows high-precision observation data to dominate the solution process, while adaptively adjusting the weights according to the dynamic changes in base station errors, thus adapting to complex engineering environments.

[0057] 3.3) Final Position This is the final output of the adaptive position calculation module.

[0058] The key point of the method described in this embodiment is: (1) Improved Z-score filtering technology for complex communication environments To address the issue of sudden jumps in ranging values ​​caused by multipath and pulse interference in complex communication environments, it is necessary to accurately identify and remove abnormal measurements to ensure the purity of positioning data. This invention proposes a method for anomaly removal using an improved Z-score, which can effectively clean the original positioning data.

[0059] The key to this data cleaning solution lies in: using the median and MAD, which are insensitive to extreme values, to build a discrimination benchmark; and achieving accurate identification and automated removal of impulse noise while retaining valid data fluctuations.

[0060] (2) Adaptive smoothing and dynamic weight allocation mechanism based on average inter-frame jump variable To address the challenge of balancing smoothness depth and real-time response speed during positioning, filtering parameters need to be adjusted in real-time based on the severity of environmental fluctuations to balance the smoothness of the positioning trajectory with dynamic responsiveness. This invention proposes a method for dynamically adjusting the attenuation coefficient based on inter-frame jump variables, which can be applied to dynamic positioning scenarios with drastic environmental fluctuations.

[0061] The key to this adaptive smoothing scheme lies in: using the average inter-frame jump variable to quantify the degree of environmental fluctuation in real time; and realizing the autonomous adjustment of the filtering intensity by dynamically constructing the attenuation coefficient and time weight.

[0062] (3) Adaptive handover scheme based on base station number awareness calculation algorithm To address the issue of poor positioning robustness caused by the dynamic change in the number of effective base stations in wireless networks, the system needs to be able to automatically select the optimal computational model for different amounts of observation data to ensure positioning accuracy across all scenarios. This invention proposes a method for automatically matching the optimal algorithm based on the number of online base stations, capable of covering all scenarios from base station limitations to base station redundancy.

[0063] The key to this adaptive location calculation scheme lies in: establishing an algorithm traffic splitting logic based on a threshold number of base stations; and achieving seamless switching between the CHAN algorithm (for fewer base stations) and the weighted least squares algorithm (for more base stations).

[0064] In another aspect, the present invention provides a server device for wireless network positioning, such as... Figure 6 As shown, it includes a processor, memory, and a communication interface, wherein: The processor and memory are connected via a system bus; The communication interface is connected to the system bus via an input / output interface; The memory includes internal memory and a non-volatile storage medium, the non-volatile storage medium being used to store the operating system, computer programs, and database; The processor is configured to execute the computer program, call the internal memory, and receive the original positioning parameter sequence through the communication interface to implement the wireless network positioning method.

[0065] Another aspect of the present invention protects a wireless network positioning system, comprising: The target node to be located is used to transmit or receive wireless signals; Multiple positioning reference nodes are used to wirelessly communicate with the target node to be positioned and to extract positioning parameters; The server device described above is used to obtain positioning parameters from the positioning reference node and perform positioning calculations to output the position coordinates of the target node to be located.

[0066] In another aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned wireless network positioning method.

[0067] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0068] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A wireless network positioning method, characterized in that, include: The system receives wireless signals from the target node to be located through multiple positioning reference nodes and extracts the original positioning parameter sequence corresponding to the target node to be located. The original positioning parameter sequence is sequentially subjected to improved Z-score anomaly removal based on median and median absolute deviation, and adaptive attenuation coefficient smoothing based on average inter-frame jump variable, outputting stable parameter estimates and measurement variance. Based on the stable parameter estimates and measurement variance, and based on the number of positioning reference nodes on which the stable parameter estimates are generated, different geometric calculation algorithms are adaptively switched to calculate the spatial coordinates of the target node to be located.

2. The wireless network positioning method according to claim 1, characterized in that, The improved Z-score outlier removal based on the median and median absolute deviation specifically includes: Calculate the median and median absolute deviation of the original positioning parameter sequence; Based on the median and the median absolute deviation, the improved Z-score value corresponding to each positioning parameter is calculated; The improved Z-score value is compared with a preset threshold, and positioning parameters that exceed the preset threshold are removed.

3. The wireless network positioning method according to claim 2, characterized in that, The improved Z-score is obtained by calculating the absolute value of the difference between the positioning parameter and the median, and comparing it with the median absolute deviation adjusted by the standardization factor.

4. The wireless network positioning method according to claim 2, characterized in that, The adaptive attenuation coefficient smoothing process based on the average inter-frame jump variable specifically includes: For the cleaned parameter sequence obtained after anomaly removal, calculate its average inter-frame jump variable to quantify the degree of parameter fluctuation; The adaptive attenuation coefficient is dynamically determined based on the relationship between the average inter-frame jump variable and the preset reference fluctuation threshold. The time decay weight of each parameter sample is calculated based on the adaptive decay coefficient, and the cleaned parameter sequence is then weighted and fused.

5. A wireless network positioning method according to claim 4, characterized in that, The adaptive attenuation coefficient is limited to a preset minimum and maximum value, and is obtained by dynamic mapping or amplitude limiting based on the ratio of the average inter-frame jump variable to the reference fluctuation threshold.

6. The wireless network positioning method according to claim 1, characterized in that, The adaptive switching of different geometric solution algorithms specifically includes: When the quantity is less than a preset threshold, the Chan analytical algorithm is used to calculate the position. When the quantity is greater than or equal to the preset threshold, the position is calculated using the weighted least squares method.

7. A wireless network positioning method according to claim 6, characterized in that, When using the weighted least squares method, an observation weight matrix is ​​constructed. The observation weight matrix is ​​a diagonal matrix, and each element on its diagonal corresponds one-to-one with the stable parameter estimates involved in the solution. Each element value is the reciprocal of the sum of the corresponding measurement variance and the measurement variance of a certain reference node, which is used to characterize the confidence distribution of signals from different reference nodes in the current environment.

8. A server device for wireless network positioning, characterized in that, This includes a processor, memory, and communication interfaces, among which: The processor and memory are connected via a system bus; The communication interface is connected to the system bus via an input / output interface; The memory includes internal memory and a non-volatile storage medium, the non-volatile storage medium being used to store the operating system, computer programs, and database; The processor is configured to execute the computer program and call the internal memory to receive the original positioning parameter sequence through the communication interface, thereby implementing the wireless network positioning method as described in any one of claims 1-7.

9. A wireless network positioning system, characterized in that, include: The target node to be located is used to transmit or receive wireless signals; Multiple positioning reference nodes are used to wirelessly communicate with the target node to be positioned and to extract positioning parameters; The server device as described in claim 8 is configured to obtain positioning parameters from the positioning reference node and perform positioning calculations to output the position coordinates of the target node to be located.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements a wireless network positioning method according to any one of claims 1-7.

Images