Positioning method, device, equipment, medium and product

Abstract

The invention discloses a positioning method and device, equipment, a medium and a product, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining a received signal strength indication value of a to-be-positioned node; processing the received signal strength indication value based on a preset weighted hybrid filtering model to obtain a corrected received signal strength indication value; distance information is calculated according to the corrected received signal strength indication value, a positioning result corresponding to the to-be-positioned node is determined according to the distance information, and the received signal strength indication value is processed through a weighted hybrid filtering model, so that random noise and errors in received signal strength indication value data can be reduced, and the positioning accuracy of the to-be-positioned node is improved. The distance information is calculated according to the corrected received signal strength indication value, and the positioning result corresponding to the node to be positioned is determined according to the distance information, so that the accuracy of the distance information is improved, and the reliability and the real-time performance of the positioning result are improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a positioning method, device, equipment, medium and product. Background Art

[0002] In the construction of smart cities, smart parks, as an important part, are committed to improving the management efficiency of parks and the quality of life of residents. Among them, positioning technology plays a key role in smart parks, providing precise positioning and tracking services, supporting personnel and equipment management, logistics management, asset tracking, navigation safety and other applications within the park.

[0003] There are various types of positioning technologies based on device selection on the market. Among them, low-power Bluetooth technology is widely used due to its low cost, high anti-interference performance, low deployment cost, and easy integration and promotion. However, due to environmental restrictions and computational complexity, it is easy to affect the reliability and real-time performance of the positioning results.

[0004] Therefore, it is necessary to propose a solution to improve the reliability and real-time performance of positioning results.

[0005] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0006] The main purpose of this application is to provide a positioning method, device, equipment, medium and product, aiming to improve the reliability and real-time performance of positioning results.

[0007] To achieve the above object, the present application provides a positioning method, the method comprising:

[0008] Obtaining a received signal strength indicator value of the node to be located;

[0009] Processing the received signal strength indication value based on a preset weighted hybrid filtering model to obtain a modified received signal strength indication value;

[0010] Distance information is calculated according to the modified received signal strength indicator value, and a positioning result corresponding to the node to be located is determined according to the distance information.

[0011] In one embodiment, before the step of processing the received signal strength indication value based on a preset weighted hybrid filtering model to obtain a modified received signal strength indication value, the step further includes:

[0012] Acquire a received signal strength indication value sample data set, wherein the received signal strength indication value sample data set includes received signal strength indication value sample data corresponding to different fixed distances;

[0013] Respectively performing mean filtering, median filtering, Kalman filtering and Gaussian filtering operations on the received signal strength indication value sample data set to obtain each filtered received signal strength indication data;

[0014] A multivariate linear regression model based on ridge regression is established according to the received signal strength indication data after each filtering, hyperparameter tuning is performed using grid search, and optimal parameters are found using five-fold cross validation to obtain the weighted hybrid filtering model.

[0015] In one embodiment, the step of obtaining a received signal strength indicator value sample data set includes:

[0016] Collect received signal strength indicator value sample data corresponding to different fixed distances;

[0017] The interference outliers in the received signal strength indication value sample data are removed to obtain the received signal strength indication value sample data set.

[0018] In one embodiment, the step of calculating distance information according to the modified received signal strength indicator value, and determining a positioning result corresponding to the node to be located according to the distance information includes:

[0019] Predicting the corrected received signal strength indicator value based on a preset ranging model to determine the distance information;

[0020] A positioning calculation is performed according to the distance information in combination with at least three reference node coordinates to obtain a positioning result corresponding to the node to be positioned.

[0021] In one embodiment, the step of predicting the corrected received signal strength indicator value based on a preset ranging model and determining the distance information further includes:

[0022] Processing the received signal strength indication value sample data set by using the weighted hybrid filtering model to obtain a corrected received signal strength indication value sample data set;

[0023] Performing a positive number transformation on the modified received signal strength indicator value sample data set to obtain transformed received signal strength indicator value sample data, and performing a logarithmic transformation on a fixed distance of the received signal strength indicator value sample data to obtain a logarithmic distance;

[0024] Performing linear fitting on the transformed received signal strength indicator value sample data and the logarithmic distance through the constructed polynomial regression model to obtain a linear fitting result;

[0025] Model hyperparameters are selected using cross validation and grid search according to the linear fitting results, and training is performed according to the model hyperparameters to obtain the ranging model.

[0026] In one embodiment, the step of performing positioning calculation according to the distance information in combination with at least three reference node coordinates to obtain a positioning result corresponding to the node to be positioned includes:

[0027] Establishing a nonlinear equation group according to the distance information and the coordinates of the at least three reference nodes;

[0028] The nonlinear equation group is solved by the quasi-Newton method and the residual sum of squares is calculated, and the positioning result of the node to be located is obtained by performing iterative calculation based on the residual sum of squares.

[0029] In addition, to achieve the above-mentioned purpose, the present application also proposes a positioning device, which includes:

[0030] An acquisition module is used to acquire a received signal strength indicator value of a node to be located;

[0031] A processing module, used for processing the received signal strength indication value based on a preset weighted hybrid filtering model to obtain a modified received signal strength indication value;

[0032] The determination module is used to calculate the distance information according to the modified received signal strength indicator value, and determine the positioning result corresponding to the node to be located according to the distance information.

[0033] In addition, to achieve the above objectives, the present application also proposes a positioning device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the positioning method described above.

[0034] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the positioning method described above are implemented.

[0035] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the positioning method described above are implemented.

[0036] One or more technical solutions proposed in this application have at least the following technical effects:

[0037] By obtaining the received signal strength indication value of the node to be located; processing the received signal strength indication value based on a preset weighted hybrid filtering model to obtain a revised received signal strength indication value; calculating distance information according to the revised received signal strength indication value, and determining the positioning result corresponding to the node to be located according to the distance information, the received signal strength indication value is processed by the weighted hybrid filtering model, so that random noise and errors in the received signal strength indication value data can be reduced, and then the distance information is calculated according to the revised received signal strength indication value, and the positioning result corresponding to the node to be located is determined according to the distance information, thereby improving the accuracy of the distance information, thereby improving the reliability and real-time performance of the positioning result. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0040] Figure 1 It is a schematic diagram of the Bluetooth AOA ranging and positioning principle shown in the prior art;

[0041] Figure 2 This is a flow chart of the first embodiment of the positioning method of the present application;

[0042] Figure 3 This is a flow chart of the second embodiment of the positioning method of the present application;

[0043] Figure 4 A schematic diagram of a weighted hybrid filtering algorithm design process according to the second embodiment of the present application;

[0044] Figure 5 This is a flow chart of the third embodiment of the positioning method of the present application;

[0045] Figure 6 Schematic diagram of the RSSI ranging algorithm design process according to the third embodiment of the present application;

[0046] Figure 7 Schematic diagram of the RSSI positioning algorithm flow according to the third embodiment of the present application;

[0047] Figure 8 This is a schematic diagram of the module structure of the positioning device according to an embodiment of the present application;

[0048] Fig. 9 It is a schematic diagram of the overall process of the positioning method in the embodiment of the present application;

[0049] Fig.10 Schematic diagram of the device structure of the hardware operating environment involved in the positioning method in the embodiment of the present application.

[0050] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0051] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0052] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0053] The main solution of the embodiment of the present application is: by obtaining the received signal strength indication value of the node to be located; processing the received signal strength indication value based on a preset weighted hybrid filtering model to obtain a modified received signal strength indication value; calculating distance information according to the modified received signal strength indication value, and determining the positioning result corresponding to the node to be located according to the distance information. By processing the received signal strength indication value through the weighted hybrid filtering model, random noise and errors in the received signal strength indication value data can be reduced, and then the distance information is calculated according to the modified received signal strength indication value, and the positioning result corresponding to the node to be located is determined according to the distance information, thereby improving the accuracy of the distance information, thereby improving the reliability and real-time performance of the positioning result.

[0054] In this embodiment, for the convenience of description, the following description is made with the positioning device as the execution subject.

[0055] In the construction of smart cities, smart parks, as an important part, are committed to improving the management efficiency of parks and the quality of life of residents. Among them, positioning technology plays a key role in smart parks, providing precise positioning and tracking services, supporting personnel and equipment management, logistics management, asset tracking, navigation safety and other applications within the park.

[0056] In most outdoor positioning systems, GPS (Global Positioning System) is widely used because it is easy to implement and has an accuracy of up to 5 meters. However, due to the limited space for indoor positioning and the large number of obstacles indoors, GPS is not suitable. Solutions based on the Internet of Things (IOT) and wireless technology can achieve accurate and efficient positioning in indoor environments.

[0057] At present, the positioning technologies based on device selection on the market mainly include WIFI, wireless network Zigbee, low-power Bluetooth BLE, ultra-wideband UWB and other technologies. Among them, low-power Bluetooth technology is widely used due to its low cost, high anti-interference, low deployment cost, easy integration and promotion, etc. Low-power Bluetooth includes Bluetooth AOA (Angle of Arrival) positioning technology and Bluetooth RSSI (Received Signal Strength Indicator) positioning technology.

[0058] Bluetooth AOA is a technology that uses the angle of arrival of signals for positioning. It mainly uses multiple receivers to determine the direction of arrival of signals, thereby achieving precise positioning. Figure 1 , Figure 1 FIG. 1 is a schematic diagram of the Bluetooth AOA ranging and positioning principle according to the prior art. Figure 1 As shown in the figure, the sending device (such as Bluetooth tag MS) sends a signal, and the receiving device (such as Bluetooth base station) receives the signal through multiple antennas (BS1, BS2). According to the time difference and phase difference of the signal reaching each antenna, the arrival angle of the signal can be calculated, and then the specific location of the tag can be determined using the triangulation algorithm. The disadvantages of Bluetooth AOA positioning technology include:

[0059] (1) Environmental limitations: The accuracy of Bluetooth AOA positioning will be significantly reduced in complex environments or when signal interference is severe. Buildings, furniture, and other obstacles may cause signal reflection and attenuation, thus affecting the positioning results;

[0060] (2) Power consumption: Although Bluetooth itself is a low-power technology, when performing high-frequency signal analysis and processing, it may cause the battery life of the device to decrease, affecting the user experience;

[0061] (3) Antenna layout requirements: Effective AOA positioning requires multiple directional antennas to provide accurate direction information. However, the layout and configuration of antennas may be subject to space constraints in actual applications, leading to the complexity of equipment installation and maintenance;

[0062] (4) Computational complexity: The process of signal processing and calculating positioning results is relatively complex, especially in a dynamic environment, which may lead to high computational delay and affect real-time performance;

[0063] (5) Cost issue: Implementing a high-precision Bluetooth AOA positioning system requires more hardware investment (such as multi-antenna systems and high-performance processors), which increases the overall cost of the system.

[0064] RSSI is the received signal strength indication, which is an indicator used to measure the strength of the received wireless signal. It is usually used to evaluate the quality of communication signals between devices and estimate distance. The indoor ranging method based on Bluetooth RSSI needs to obtain the signal strength of the transmitting node. The receiving node collects RSSI and calculates the loss during the propagation process. Then, a series of propagation attenuation models are used to convert the loss of wireless signals into distance. The disadvantages of Bluetooth RSSI positioning technology include:

[0065] (1) Environmental limitations: RSSI positioning technology performs poorly in complex environments. Obstacles such as buildings and furniture may cause signal attenuation and multipath effects, thus affecting positioning accuracy. The signal propagation path is unstable, resulting in large fluctuations in the RSSI value at the same location, which in turn affects the reliability of the positioning results.

[0066] (2) Nonlinear distance relationship: There is a nonlinear relationship between RSSI and actual distance, which makes it difficult to accurately calculate the actual distance from the RSSI value measured in different environments, increasing the positioning error;

[0067] (3) Computational complexity: Although RSSI measurement is relatively simple, real-time positioning in a dynamic environment requires processing and filtering of a large amount of data, which may cause computational delays and affect real-time performance.

[0068] This application provides a new positioning technology that combines multiple signal processing methods to improve positioning accuracy and environmental adaptability, reduce calculation delay, and aims to enhance user experience and wide applicability, and achieve a more accurate and real-time indoor positioning solution. The main problems solved are as follows:

[0069] (1) Solve the cost issues of hardware investment and the complexity of equipment installation and maintenance in the positioning system;

[0070] (2) Solve the positioning errors caused by multipath effects, signal attenuation, environment and other factors during RSSI collection, and improve positioning accuracy;

[0071] (3) Solve the problem of adaptive ranging model in indoor environment and consider the calibration of nonlinear transformation relationship between RSSI signal value and distance;

[0072] (4) Solve the real-time processing of positioning coordinate calculation in dynamic environments, optimize algorithms and data processing processes to reduce delays, improve system response speed, and meet real-time application requirements.

[0073] It should be noted that the execution subject of this embodiment may be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a positioning device, etc. The following takes the positioning device as an example to illustrate this embodiment and the following embodiments.

[0074] Based on this, the present application embodiment provides a positioning method, referring to Figure 2 , Figure 2 This is a flow chart of the first embodiment of the positioning method of the present application.

[0075] In this embodiment, the positioning method includes steps S10 to S30:

[0076] Step S10, obtaining a received signal strength indicator value of the node to be located;

[0077] Specifically, the node to be located in the embodiment of the present application can be the location of the device to be located, and Bluetooth RSSI (Received Signal Strength Indicator) is a method for measuring the strength of radio signals. It represents the received signal power level, usually in dBm (decibel milliwatts). The RSSI value can be used to estimate the relative distance between two Bluetooth devices. Generally, RSSI signals are affected by multiple factors during propagation, such as multipath effects caused by obstacles, reflections, and refractions encountered during signal propagation, and electromagnetic interference and power supply noise generated by other irrelevant interference signals received by the receiver, resulting in the RSSI directly collected not being able to directly restore the true distance, affecting the ranging accuracy and stability. In the embodiment of the present application, after obtaining the received signal strength indicator value RSSI of the node to be located, the RSSI is corrected by a weighted hybrid filtering model to reduce interference such as noise in the RSSI data.

[0078] Step S20, processing the received signal strength indication value based on a preset weighted hybrid filtering model to obtain a modified received signal strength indication value;

[0079] Furthermore, after the received signal strength indication value of the node to be located is obtained, the received signal strength indication value may be processed based on a preset weighted hybrid filtering model to obtain a modified received signal strength indication value.

[0080] For example, traditional filtering algorithms often have certain limitations when processing RSSI data. For example, although the mean filtering algorithm is simple, it does not work well on sudden change data; the median filtering algorithm has a good suppression effect on pulse noise, but has poor adaptability to continuously changing data. In addition, a single filtering algorithm does not work well on RSSI. In order to solve these problems, a weighted hybrid filtering algorithm that integrates median filtering, mean filtering, Kalman filtering, and Gaussian filtering is designed in the embodiment of the present application. Combining the advantages of multiple filtering algorithms, the RSSI measurement data is smoothed, the noise in the RSSI data is removed, the accuracy of the data is improved, and accurate data support is provided for the subsequent ranging model.

[0081] Step S30: Calculate distance information according to the corrected received signal strength indicator value, and determine a positioning result corresponding to the node to be located according to the distance information.

[0082] Furthermore, after the received signal strength indication value is processed based on a preset weighted hybrid filtering model to obtain a corrected received signal strength indication value, distance information can be calculated based on the corrected received signal strength indication value, and the positioning result corresponding to the node to be located can be determined based on the distance information.

[0083] For example, the relationship between RSSI and distance is usually nonlinear. The commonly used indoor ranging model based on RSSI is the attenuation factor model, which is expressed as follows:

[0084]

[0085] Where n represents the path attenuation factor, and A represents the RSSI value measured at 1 meter. This model is usually determined based on experimental data, empirical formulas, simple free space or obstacle models, but it cannot handle complex indoor environments, thus affecting the precision and accuracy of ranging. In addition, the attenuation factor model usually requires the establishment of a signal attenuation model in advance and the determination of model parameters based on experimental data, which requires an analysis in different indoor environments and application scenarios.

[0086] Exemplarily, an adaptive polynomial regression model is proposed in the embodiment of the present application to fit the relationship between RSSI and distance. By collecting the RSSI values ​​of multiple devices in each distance segment, the indoor environment is adapted to avoid the influence of ranging accuracy caused by manual selection of indoor ranging models for parameter setting. In addition, through the training and calculation of the polynomial regression model, and increasing the degree of the polynomial, the data fitting accuracy can be further improved, thereby more accurately predicting the distance.

[0087] For example, when determining the position of a node to be located in an actual application scenario, it is often necessary to use the distance information between multiple reference nodes and the node to be located. After obtaining the distances between multiple reference nodes and the node to be located, it is also necessary to calculate the coordinates of the node to be located through a certain positioning algorithm. However, traditional positioning algorithms often expose obvious limitations when dealing with nonlinear problems. For example, when facing complex environments or nonlinear relationships, triangulation may result in large errors, making it difficult to accurately determine the coordinates of the node to be located. Although the maximum likelihood estimation method can handle some uncertainty problems to a certain extent, its computational complexity is high in nonlinear situations, and its accuracy is difficult to guarantee. In order to overcome these difficulties, the nonlinear least squares method is used in the embodiment of the present application to solve the positioning coordinates. This method can deal with nonlinear problems more flexibly and determine the position of the node to be located more accurately.

[0088] This embodiment adopts the above scheme, specifically by obtaining the received signal strength indication value of the node to be located; processing the received signal strength indication value based on a preset weighted hybrid filtering model to obtain a modified received signal strength indication value; calculating distance information according to the modified received signal strength indication value, and determining the positioning result corresponding to the node to be located according to the distance information. By processing the received signal strength indication value through the weighted hybrid filtering model, the random noise and error in the received signal strength indication value data can be reduced, and then the distance information is calculated according to the modified received signal strength indication value, and the positioning result corresponding to the node to be located is determined according to the distance information, thereby improving the accuracy of the distance information, thereby improving the reliability and real-time performance of the positioning result.

[0089] Based on the first embodiment of the present application, a second embodiment of the present application is proposed. In the second embodiment of the present application, the same or similar contents as those of the above-mentioned embodiment 1 can be referred to the above introduction and will not be repeated later.

[0090] On this basis, please refer to Figure 3 Before step S20, the positioning method further includes steps S01 to S02:

[0091] Step S01: obtaining a received signal strength indicator value sample data set, wherein the received signal strength indicator value sample data set includes received signal strength indicator value sample data corresponding to different fixed distances;

[0092] Step S02: performing mean filtering, median filtering, Kalman filtering and Gaussian filtering operations on the received signal strength indication value sample data set respectively to obtain received signal strength indication data after filtering;

[0093] Step S03: A multivariate linear regression model based on ridge regression is established according to the received signal strength indication data after each filtering, hyperparameter tuning is performed using grid search, and optimal parameters are found using five-fold cross validation to obtain the weighted hybrid filtering model.

[0094] Exemplarily, the step of obtaining a received signal strength indicator value sample data set includes:

[0095] Collect received signal strength indicator value sample data corresponding to different fixed distances;

[0096] The interference outliers in the received signal strength indication value sample data are removed to obtain the received signal strength indication value sample data set.

[0097] Reference Figure 4 , Figure 4 FIG. 1 is a schematic diagram of a weighted hybrid filtering algorithm design process according to the second embodiment of the present application, as shown in FIG. Figure 4 As shown in the figure, first, the original RSSI data is preprocessed to remove interference outliers; secondly, the RSSI data is subjected to multiple filtering processes under the sliding window to obtain multiple filtering results; then, a multivariate linear regression model based on ridge regression is constructed according to the filtered data and the measured values, and the model is trained. The optimal parameters are searched through cross-validation and different weights are assigned to them; finally, the weighted filtering results are fused to obtain the final corrected RSSI weighted hybrid filtering data set.

[0098] For example, for mean filtering, the input data is assumed to be a sequence of RSSI values. By setting the window size, the data in the window is averaged to obtain the result after mean filtering. n}, the output signal after mean filtering is:

[0099]

[0100] where y n is the RSSI value after filtering, x n is the original signal value, and M is the sliding window size. Using a sliding window can effectively filter outliers and random noise, reduce data fluctuations, and capture dynamic change characteristics in the data to improve signal quality.

[0101] For example, for median filtering, a sliding window is used to perform median filtering on RSSI data. Assume that the input signal is {x1, x2, ..., x n}, the median filter arranges the data sampled by the sliding window in ascending order and selects the middle value of the window as the filtered result to effectively remove the impulse noise.

[0102] After median filtering, the output signal is

[0103] yi=Mid{sort{x i-k ,x i-k+1 ,…,x i+k-1 ,x i+k}}

[0104] Here, k represents the sliding window radius, generally k=M / 2, and M represents the sliding window size.

[0105] Exemplarily, for Kalman filtering processing, the RSSI estimate at the previous moment and the RSSI measurement at the current moment are used to update the state variable estimate, and the best estimate at the current moment is obtained. The state estimate and the error covariance matrix can be dynamically updated, with high real-time and computational efficiency, and good robustness to noise and uncertainty.

[0106] The calculation formula of the state prediction value is as follows:

[0107]

[0108] Among them: F k μ is the state transfer matrix, which describes the state change from time k-1 to time k. It maps the state of the previous moment to the current moment. k B is the control input, which represents the external influence on the system at time k. k is the control input matrix, and the control input μ k Translated to the effect on the state.

[0109] Error covariance prediction calculation formula:

[0110]

[0111] Where Q k represents the process noise covariance matrix.

[0112] Kalman gain calculation formula:

[0113]

[0114] Where: H k P is the measurement matrix, which maps the state vector to the measurement space so that the state can be compared with the actual measurement value. k|k Represents the error covariance matrix of the previous time point k-1, indicating the confidence of the state estimate, and a smaller value indicates a higher confidence. R k K represents the measurement noise covariance and represents the uncertainty of the data. k represents the Kalman gain, which is used to balance the weights of prediction and measurement and to update the state estimate.

[0115] Status Update:

[0116]

[0117] Where: z avg Represents the average of the measurements in the sliding window.

[0118] Error covariance update:

[0119] P k|k =(IK k H k ) k|k-1

[0120] Where: I is the identity matrix, P k|k Represents the error covariance matrix at the current time point k.

[0121] For example, for Gaussian filtering, the RSSI data collected generally follows a normal distribution, and the next RSSI value data collected by the receiving node will vary within a reasonable range, and the probability of abnormal value data generation is small. Based on the concept of Gaussian function, high-probability data can be screened out, low-probability data can be removed, and high-frequency noise can be effectively removed through weighted averaging to retain the main features of the data and make the data smoother. The calculation steps are as follows:

[0122]

[0123] Among them G norm(k) Represents the normalized Gaussian weight, and the calculation formula is:

[0124]

[0125] in represents the sum of all weights, represents Gaussian weight, and σ represents standard deviation.

[0126] Exemplarily, in the embodiment of the present application, based on the four RSSI data sets after the above filtering processing and the RSSI measurement values ​​at different fixed distances, a multivariate linear regression model based on ridge regression is established, grid search is used for hyperparameter tuning, and five-fold cross validation is used to find the optimal parameters, the optimal model is trained, and the various weights of the weighted hybrid filter are derived, thereby combining the advantages of various filters to greatly improve the accuracy of ranging.

[0127] The constructed multiple linear regression model is as follows:

[0128] Y=β0+β1X1+β2X2+β3X3+β4X4+ε

[0129] Among them: X1 is the mean filter processing value, X2 is the median filter processing value, X3 is the Kalman filter processing value, X4 is the Gaussian filter processing value, ε is the error term, and Y is the actual RSSI measurement value at different fixed distances, which is expressed in matrix form as follows:

[0130] Y=βX+E

[0131] Where: Y is the dependent variable, X is the independent variable matrix (including X1, X2, X3, X4), β is the regression coefficient vector, and E is the error term.

[0132] For example, the ridge regression model adds a regularization term to control the size of the regression coefficient while minimizing the loss function. Its objective function is expressed as:

[0133] min(||Y-Xβ|| 2 +θ||β|| 2 )

[0134] Among them, ||Y-Xβ|| 2 is the residual sum of squares, ||β|| 2 is the L2 norm of the regression coefficient, and θ is the regularization parameter that controls the strength of the penalty term.

[0135] For example, since the features of RSSI data collection are far less than the number of samples, and the input features of the model are mainly four filter inputs, if L1 regularization is used, that is, LASSO regression is used, some important feature coefficients will be compressed to 0, and the information of all four filter inputs cannot be fully utilized, resulting in an increase in the deviation of the model. Therefore, the main feature of using L2 regularization is to prevent overfitting, better process the features composed of four filter inputs without eliminating eigenvalues, and improve the stability and prediction ability of the model.

[0136] For example, using real measured RSSI values ​​can enable the model to better adapt to factors such as multipath effects and signal attenuation in actual environments. At the same time, data containing real environmental noise enhances the robustness of the model and can better handle various situations in actual use.

[0137] Based on any of the above embodiments of the present application, a third embodiment of the present application is proposed. In the third embodiment of the present application, the same or similar contents as any of the above embodiments can be referred to the above introduction and will not be repeated later. Figure 5 , step S30 also includes steps S301-S302:

[0138] Step S301: predicting the corrected received signal strength indicator value based on a preset distance measurement model to determine the distance information;

[0139] Exemplarily, the step of predicting the corrected received signal strength indicator value based on a preset ranging model and determining the distance information further includes:

[0140] Processing the received signal strength indication value sample data set by using the weighted hybrid filtering model to obtain a corrected received signal strength indication value sample data set;

[0141] Performing a positive transformation on the modified received signal strength indicator value sample data set to obtain transformed received signal strength indicator value sample data, and performing a logarithmic transformation on the fixed distance of the received signal strength indicator value sample data to obtain a logarithmic distance;

[0142] Performing linear fitting on the transformed received signal strength indicator value sample data and the logarithmic distance through the constructed polynomial regression model to obtain a linear fitting result;

[0143] Model hyperparameters are selected using cross validation and grid search according to the linear fitting results, and training is performed according to the model hyperparameters to obtain the ranging model.

[0144] Reference Figure 6 , Figure 6 FIG. 1 is a schematic diagram of a design flow of an RSSI ranging algorithm according to the third embodiment of the present application. Figure 6 As shown in the figure, first, the corrected RSSI data set obtained after the original RSSI data is processed by the weighted hybrid filtering model is used as input; secondly, the RSSI corrected data is transformed into a positive number, and the fixed sampling distance of the RSSI data is logarithmically transformed to linearize the relationship between the RSSI value and the logarithmic distance, so that the polynomial regression model can effectively fit the data; a polynomial regression model is constructed to fit the relationship between the RSSI value and the logarithmic distance, and the best model hyperparameters are selected using methods such as cross-validation and grid search to ensure the adaptability of the model on different data sets; the RSSI ranging model training is completed based on the optimal parameters, and the logarithmic distance is predicted and converted back to the actual distance.

[0145] For example, RSSI refers to the received signal strength, which is generally a negative number, so it needs to be converted to a positive number. The formula is as follows:

[0146] R=-RSSI

[0147] The actual RSSI sampling distance is logarithmically transformed to linearize the relationship between RSSI and distance. The formula is as follows:

[0148] D = log(sampling distance)

[0149] Exemplarily, a polynomial regression model is used to fit the nonlinear relationship between RSSI value and logarithmic distance. The model formula can be expressed as:

[0150] D=β0+β1R+β2R 2 +∈

[0151] Where: D represents the logarithmic distance, R represents the processed RSSI value, β0, β1, β2 are model parameters, and ∈ represents the error term.

[0152] For example, after obtaining the polynomial regression model with the best parameters based on cross-validation and grid search, the logarithmic distance is predicted by the model, and then converted back to the calculated distance d by an exponential function:

[0153] d=e D

[0154] Exemplarily, based on the above-mentioned model training process, an adaptive dynamic model update strategy based on the idea of ​​incremental learning is further proposed in the embodiment of the present application. By comparing the mean square error MSE of the model with the set error threshold, it is determined whether the model performance has declined, thereby deciding whether to update the model. In addition, when the model obtains new data, it is not necessary to retrain the entire model every time. Instead, the existing model is updated after obtaining the new data, which greatly saves computing resources and time. In addition, it can quickly adapt to new data and maintain the accuracy of the model.

[0155] Step S302: performing positioning calculation according to the distance information in combination with coordinates of at least three reference nodes to obtain a positioning result corresponding to the node to be positioned.

[0156] Exemplarily, the step of performing positioning calculation according to the distance information in combination with at least three reference node coordinates to obtain a positioning result corresponding to the node to be positioned includes:

[0157] Establishing a nonlinear equation group according to the distance information and the coordinates of the at least three reference nodes;

[0158] The nonlinear equation group is solved by the quasi-Newton method and the residual sum of squares is calculated, and the positioning result of the node to be located is obtained by performing iterative calculation based on the residual sum of squares.

[0159] Reference Figure 7 , Figure 7 FIG. 1 is a flow chart of an RSSI positioning algorithm according to the third embodiment of the present application. Figure 7As shown in the figure, first, according to the obtained RSSI value, the weighted hybrid filtering algorithm is corrected and the adaptive polynomial regression model is used to predict the distance between the node to be located and the reference node; according to the known reference node coordinates and the predicted distance, a nonlinear equation group is established; the equation group is solved by the optimization algorithm, and the coordinates of the node to be located are continuously adjusted during the solution process to minimize the residual square sum of the equation group and improve the positioning accuracy and stability; through continuous iteration and optimization, the precise coordinates of the node to be located are obtained.

[0160] For example, in the process of establishing the nonlinear equation system, based on the three known reference node coordinates (x i ,y i )(i=1,2,3), and a node to be located (x,y). The distance d between the node to be located and the reference node is predicted by the adaptive model. i , we get the following relationship:

[0161]

[0162] In order to transform the above relationship into a nonlinear system of equations, the residual r is introduced i , which represents the difference between the predicted distance and the distance calculated based on the coordinates of the node to be located:

[0163]

[0164] At this time, the calculation of positioning coordinates needs to ensure positioning accuracy and minimize the sum of squares of residuals, so the following objective function is established:

[0165]

[0166] For example, Newton's method or gradient descent method is generally used to solve nonlinear equations. Newton's method has the characteristics of fast convergence, but it is sensitive to the selection of initial values ​​and has a large amount of calculation; if the initial value is not selected properly, the algorithm may not converge. Although the gradient descent method is relatively simple to calculate, the convergence speed may be slow. During the solution process, the coordinates of the nodes to be located need to be continuously adjusted, and considering the real-time impact, the quasi-Newton method is chosen.

[0167] For example, the quasi-Newton method does not need to calculate the Jacobian matrix, the amount of calculation is relatively small, and it has good convergence. The idea of ​​the quasi-Newton method is to update the approximate Jacobian matrix according to the information of the current iteration point in each iteration step, so as to determine the next iteration point. Its core goal is to minimize the residual sum of squares of the equation system. By continuously adjusting and optimizing the solution process, nonlinear problems are effectively handled, thereby significantly improving the accuracy and stability of positioning. The calculation process is as follows:

[0168] Based on the objective function of the above nonlinear equations, its gradient is calculated as follows:

[0169]

[0170] Expressed as a vector:

[0171]

[0172] Update rules:

[0173] In each iteration, the coordinates (x i ,y i ), as follows:

[0174]

[0175] And update the approximate Hessian matrix, the formula is as follows:

[0176]

[0177] Among them, H k is the current approximate Hessian matrix, represents the current step vector, Represents the gradient change vector.

[0178] For example, by applying the quasi-Newton method, the coordinates of the node to be located can be effectively solved by continuous iterative optimization, thereby improving the accuracy and stability of positioning. In each iteration, the Hessian matrix is ​​optimized using current information, thereby accelerating the convergence process.

[0179] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the positioning method of the present application. More simple transformations based on this technical concept are all within the protection scope of the present application.

[0180] This application also provides a positioning device, please refer to Figure 8 , the positioning device comprises:

[0181] An acquisition module is used to acquire a received signal strength indicator value of a node to be located;

[0182] A processing module, used for processing the received signal strength indication value based on a preset weighted hybrid filtering model to obtain a modified received signal strength indication value;

[0183] The determination module is used to calculate the distance information according to the modified received signal strength indicator value, and determine the positioning result corresponding to the node to be located according to the distance information.

[0184] Reference Fig. 9 , Fig. 9 FIG. 1 is a schematic diagram of the overall flow of the positioning method in the embodiment of the present application. Fig. 9As shown, an RSSI-based indoor positioning method is provided in the embodiment of the present application to improve the accuracy and stability of indoor positioning. The method mainly includes three parts: RSSI weighted hybrid filtering algorithm design, RSSI ranging algorithm design and RSSI positioning algorithm design. The basic steps include:

[0185] 1: Design a weighted hybrid filtering algorithm and train a regression model based on the measured data, which contains weighted coefficients for multiple filtering algorithms;

[0186] 2: The RSSI measurement data set is passed through a weighted hybrid filtering algorithm to obtain a corrected RSSI data set, thereby reducing the random noise and error of RSSI;

[0187] 3: Take the corrected RSSI data set as input, establish an adaptive ranging model for model training, and reduce the nonlinear error of RSSI-path distance;

[0188] 4: The three RSSI values ​​collected by the device to be located are input into the ranging model to obtain the predicted distances d1, d2, and d3

[0189] 5: Using the three distances and the known device location, the three-point positioning algorithm is used to calculate the coordinates (x, y) of the device to be positioned.

[0190] Exemplarily, the key technical points involved in the positioning method in the embodiment of the present application include:

[0191] (1) Design a weighted hybrid filtering algorithm and ridge regression model to solve the positioning errors caused by multipath effects, signal attenuation, and environmental factors during RSSI collection, and improve positioning accuracy;

[0192] (2) Considering the calibration of the nonlinear transformation relationship between RSSI signal value and distance, a polynomial regression model is designed to solve the adaptability of the ranging model in indoor environments;

[0193] (3) To solve the real-time processing of positioning coordinate calculation in a dynamic environment, the algorithm and data processing flow are optimized to reduce delays, improve the system's response speed, and meet the requirements of real-time applications. An adaptive dynamic model update strategy based on incremental learning is further proposed for updating the adaptive polynomial regression ranging model.

[0194] (4) To solve the problem that traditional calculation methods are difficult to ensure accuracy under nonlinear calculations, the idea of ​​nonlinear least squares method is innovatively adopted and the quasi-Newton method is used to accurately determine the position of the node to be located.

[0195] This embodiment, through the above scheme, specifically by adopting technologies such as weighted hybrid filtering algorithm, adaptive polynomial regression model and nonlinear least squares method, effectively improves the accuracy of indoor positioning and can meet the needs of practical applications. The positioning method in the embodiment of the present application has strong adaptability to changes in the indoor environment and can maintain high positioning stability under different signal propagation conditions. The positioning method in the embodiment of the present application is based on existing wireless communication technology and equipment, does not require additional hardware investment, and is easy to implement and promote. Compared with other indoor positioning technologies, the positioning method in this embodiment is low in cost and has a high cost performance.

[0196] The positioning device provided by the present application adopts the positioning method in the above embodiment to solve the technical problem of positioning. Compared with the prior art, the beneficial effects of the positioning device provided by the present application are the same as the beneficial effects of the positioning method provided by the above embodiment, and other technical features in the positioning device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0197] The present application provides a positioning device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the positioning method in the above-mentioned embodiment 1.

[0198] Reference below Fig.10 , which shows a schematic diagram of the structure of a positioning device suitable for implementing the embodiments of the present application. The positioning device in the embodiments of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Fig.10 The positioning device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0199] like Fig.10As shown, the positioning device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the positioning device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the positioning device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a positioning device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0200] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0201] The positioning device provided by the present application adopts the positioning method in the above embodiment to solve the technical problem of positioning. Compared with the prior art, the beneficial effects of the positioning device provided by the present application are the same as the beneficial effects of the positioning method provided by the above embodiment, and the other technical features in the positioning device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0202] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0203] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0204] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the positioning method in the above-mentioned embodiment.

[0205] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0206] The computer-readable storage medium may be included in the positioning device; or may exist independently without being assembled into the positioning device.

[0207] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the positioning device, the positioning device: obtains the received signal strength indication value of the node to be located; processes the received signal strength indication value based on a preset weighted hybrid filtering model to obtain a modified received signal strength indication value; calculates distance information according to the modified received signal strength indication value, and determines the positioning result corresponding to the node to be located according to the distance information. By processing the received signal strength indication value by the weighted hybrid filtering model, random noise and errors in the received signal strength indication value data can be reduced, and then the distance information is calculated according to the modified received signal strength indication value, and the positioning result corresponding to the node to be located is determined according to the distance information, thereby improving the accuracy of the distance information, thereby improving the reliability and real-time performance of the positioning result.

[0208] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0209] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0210] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0211] The readable storage medium provided in this application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned positioning method, and can solve the technical problem of positioning. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the positioning method provided in the above-mentioned embodiment, and will not be repeated here.

[0212] The present application also provides a computer program product, including a computer program, which implements the steps of the positioning method as described above when executed by a processor.

[0213] The computer program product provided by the present application can solve the technical problem of positioning. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the positioning method provided by the above embodiment, which will not be repeated here.

[0214] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A positioning method, characterized in that: The method comprises: Obtaining a received signal strength indicator value of the node to be located; Processing the received signal strength indication value based on a preset weighted hybrid filtering model to obtain a modified received signal strength indication value; Distance information is calculated according to the modified received signal strength indicator value, and a positioning result corresponding to the node to be located is determined according to the distance information.

2. The method according to claim 1, characterized in that Before the step of processing the received signal strength indication value based on a preset weighted hybrid filtering model to obtain a modified received signal strength indication value, the step further includes: Acquire a received signal strength indication value sample data set, wherein the received signal strength indication value sample data set includes received signal strength indication value sample data corresponding to different fixed distances; Respectively performing mean filtering, median filtering, Kalman filtering and Gaussian filtering operations on the received signal strength indication value sample data set to obtain each filtered received signal strength indication data; A multivariate linear regression model based on ridge regression is established according to the received signal strength indication data after each filtering, hyperparameter tuning is performed using grid search, and the optimal parameters are found using five-fold cross validation to obtain the weighted hybrid filtering model.

3. The method according to claim 2, characterized in that The step of obtaining a received signal strength indicator value sample data set comprises: Collect received signal strength indicator value sample data corresponding to different fixed distances; The interference outliers in the received signal strength indication value sample data are removed to obtain the received signal strength indication value sample data set.

4. The method according to claim 2, characterized in that The step of calculating distance information according to the modified received signal strength indicator value, and determining a positioning result corresponding to the node to be located according to the distance information comprises: Predicting the corrected received signal strength indicator value based on a preset ranging model to determine the distance information; A positioning calculation is performed according to the distance information in combination with at least three reference node coordinates to obtain a positioning result corresponding to the node to be positioned.

5. The method according to claim 4, characterized in that Before the step of predicting the corrected received signal strength indicator value based on a preset ranging model and determining the distance information, the step further includes: Processing the received signal strength indication value sample data set by using the weighted hybrid filtering model to obtain a corrected received signal strength indication value sample data set; Performing a positive number transformation on the modified received signal strength indicator value sample data set to obtain transformed received signal strength indicator value sample data, and performing a logarithmic transformation on the fixed distance of the received signal strength indicator value sample data to obtain a logarithmic distance; Performing linear fitting on the transformed received signal strength indicator value sample data and the logarithmic distance through the constructed polynomial regression model to obtain a linear fitting result; Model hyperparameters are selected using cross validation and grid search according to the linear fitting results, and training is performed according to the model hyperparameters to obtain the ranging model.

6. The method according to claim 4, characterized in that The step of performing positioning calculation according to the distance information in combination with at least three reference node coordinates to obtain a positioning result corresponding to the node to be positioned comprises: Establishing a nonlinear equation group according to the distance information and the coordinates of the at least three reference nodes; The nonlinear equation group is solved by the quasi-Newton method and the residual sum of squares is calculated, and the positioning result of the node to be located is obtained by iterative calculation based on the residual sum of squares.

7. A positioning device, characterized in that: The device comprises: An acquisition module is used to acquire a received signal strength indicator value of a node to be located; A processing module, used for processing the received signal strength indication value based on a preset weighted hybrid filtering model to obtain a modified received signal strength indication value; The determination module is used to calculate the distance information according to the modified received signal strength indicator value, and determine the positioning result corresponding to the node to be located according to the distance information.

8. A positioning device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the positioning method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the positioning method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the positioning method according to any one of claims 1 to 6 are implemented.

Images