High-resolution reconstruction positioning method and system based on sparse fingerprints

Abstract

The invention provides a high-resolution reconstruction positioning method and system based on sparse fingerprints, and the method comprises the steps: carrying out the collection and processing of received signal strength indication data of a target region through setting a reference point, constructing an offline fingerprint database, and designing a sparse fingerprint image reconstruction network which combines a historical fingerprint map and a real-time sparse fingerprint map. An offline fingerprint database is used for training; in the on-line stage, after a real-time sparse fingerprint map of a positioning target is collected, training is carried out by utilizing a historical fingerprint map and the real-time sparse fingerprint map, so that the purpose of realizing high-precision fingerprint map reconstruction and positioning in a dynamic and complex environment can be achieved. While the maintenance cost is reduced and the computing resources are saved, the reconstruction of the current high-resolution fingerprint map is realized, the positioning precision in the target area is improved, and the adaptability of a positioning system in a dynamic complex environment is greatly enhanced.

Description

Technical Field

[0001] The present invention relates to the field of information communication technology, and in particular to a high-resolution reconstruction positioning method and system based on sparse fingerprints. Background Art

[0002] With the development of wireless communication technology, indoor positioning systems are increasingly used in smart buildings, logistics management, precision medicine and other fields. As a common indoor positioning method, Wi-Fi fingerprint positioning builds a fingerprint library by collecting signal strength data at different locations offline, and matches them in the online stage to achieve accurate positioning. However, with the passage of time and changes in the environment (such as indoor layout adjustments, personnel flow, equipment differences, etc.), the timeliness of the fingerprint library gradually decreases, resulting in a decrease in positioning accuracy, and even complete failure in complex environments.

[0003] In the existing technology, in order to solve the timeliness problem of the fingerprint library, the fingerprint collection frequency is usually increased or environmental modeling is introduced, but these methods often require high computing costs or collection costs, and it is difficult to maintain high robustness in a dynamically changing environment. Therefore, how to achieve high-precision and high-robust positioning in a complex and changing environment has become a key challenge for current technology.

[0004] In recent years, deep learning technology has made significant progress in the fields of image processing and signal processing, especially in image super-resolution and feature reconstruction. However, existing deep learning technology is mostly applied to traditional image reconstruction tasks, and there is still insufficient research on the reconstruction of sparse fingerprint images and the improvement of positioning accuracy in dynamic environments. Summary of the invention

[0005] Purpose of the invention: The first purpose of the present invention is to provide a high-resolution reconstruction and positioning method based on sparse fingerprints that reduces energy consumption while ensuring accuracy. The second purpose is to provide a high-resolution reconstruction and positioning system based on sparse fingerprints.

[0006] Technical solution: A high-resolution reconstruction positioning method based on sparse fingerprints, comprising the following steps: S1, in the offline stage, reference points are evenly set in the target area, and the RSSI values ​​of all anchor nodes in the target area received at each reference point are collected to form the fingerprint of the reference point. The fingerprints of all reference points constitute a dense fingerprint library; S2, collecting a number of dense fingerprint libraries at timestamps at fixed time intervals to form an interval dense fingerprint library set; S3, extracting the data in the interval dense fingerprint library according to the anchor nodes, obtaining the dense fingerprint map corresponding to each anchor node at each timestamp, forming an interval dense fingerprint map set, setting the sparse coefficient to determine the sparse reference point, and extracting the sparse fingerprint map corresponding to each dense fingerprint map according to the sparse reference point, forming an interval sparse fingerprint map set; S4, respectively normalize the data in the interval sparse fingerprint atlas and the interval dense fingerprint atlas to obtain the corresponding interval sparse fingerprint pixel atlas and the interval dense fingerprint pixel atlas; extract the dense fingerprint pixel map corresponding to the earliest timestamp from the interval dense fingerprint pixel atlas to form a historical dense fingerprint pixel atlas, and the remaining dense fingerprint pixel maps constitute a real-time dense fingerprint pixel atlas; remove the sparse fingerprint pixel map corresponding to the earliest timestamp from the interval sparse fingerprint pixel atlas, and the remaining sparse fingerprint pixel maps constitute a real-time sparse fingerprint pixel atlas; S5, constructing a sparse fingerprint image reconstruction network, taking the historical dense fingerprint pixel atlas and the real-time sparse fingerprint pixel atlas as inputs of the sparse fingerprint image reconstruction network, and training the sparse fingerprint image reconstruction network using the real-time dense fingerprint pixel map for each anchor node to obtain the optimal sparse fingerprint image reconstruction network corresponding to each anchor node; S6, in the online stage, the target to be located collects RSSI values ​​on all anchor nodes in the target area to form a real-time fingerprint of the target to be located, normalizes the real-time fingerprint to obtain a real-time pixel fingerprint, and simultaneously receives RSSI values ​​on all anchor nodes at a sparse reference point to obtain an online sparse fingerprint atlas, which is converted into an online sparse fingerprint pixel atlas after normalization, and the online sparse fingerprint pixel atlas is matched with the historical dense fingerprint pixel atlas, and the matched historical dense fingerprint pixel atlas and the online sparse fingerprint pixel atlas are used as inputs of the optimal sparse fingerprint image reconstruction network, and a real-time dense pixel fingerprint set is output; S7. Calculate the Euclidean distance between the normalized real-time pixel fingerprint and each real-time dense pixel fingerprint in the real-time dense pixel fingerprint set, sort the calculated Euclidean distances, select several real-time dense pixel fingerprints with the smallest Euclidean distances, perform weighted calculation on the reference point coordinates corresponding to these several real-time dense pixel fingerprints, and obtain the coordinates of the target to be located.

[0007] Specifically, in step S1, reference points are evenly set at fixed intervals in the target area, and the number of reference points is: , Where: is the number of reference points, is the length of the target area, is the width of the target area, The fixed interval distance is set.

[0008] Specifically, in step S3, setting the sparse coefficient to determine the sparse reference points includes: setting the sparse coefficient, and determining the number of sparse reference points selected for the long side and the wide side of the target area according to the sparse coefficient: , Where: The number of sparse reference points selected for the long side of the target area, The number of sparse reference points selected for the broad side of the target area; The number of reference points selected for the long side of the target area; The number of reference points selected for the broadside of the target area; is the sparse coefficient.

[0009] Specifically, in step S4, the normalization process includes: The RSSI value range is set to [-100dBm, 0dBm], and the RSSI value less than -100dBm is set to -100dBm. If the RSSI value of the corresponding anchor node is not collected at the reference point or sparse reference point, the missing RSSI value is also set to -100dBm, and then all RSSI values ​​are converted to the range of [0, 1] through normalization.

[0010] Specifically, in step S5: the sparse fingerprint image reconstruction network includes an upsampling layer, a channel merging layer, a convolutional coding layer, a feature fusion module and a convolutional decoding layer connected in sequence; the convolutional coding layer includes a first convolutional layer and a second convolutional layer connected in sequence; the feature fusion module includes four residual blocks connected in sequence, each residual block includes two convolutional layers; the convolutional decoding layer includes a third convolutional layer and a fourth convolutional layer connected in sequence.

[0011] Specifically, in the sparse fingerprint image reconstruction network, the upsampling layer upsamples the real-time sparse fingerprint pixel map to align the resolution of the real-time sparse fingerprint pixel map with the historical dense fingerprint pixel map; the channel merging layer merges the real-time sparse fingerprint pixel map with the historical dense fingerprint pixel map in the channel direction; the first convolution layer extracts shallow features from the merged pixel map; the second convolution layer extracts deep features in the pixel map; the feature fusion module fuses the deep features with the shallow features; the third convolution layer reduces the dimensionality of the high-dimensional features; and the fourth convolution layer generates a predicted real-time dense fingerprint pixel map.

[0012] Specifically, the convolution kernel width of the first convolution layer is 3, the height is 3, the number of channels is 2, the number of convolution kernels is 32, the padding is 1, and the stride is 1; the convolution kernel width of the second convolution layer is 3, the height is 3, the number of channels is 32, the number of convolution kernels is 64, the padding is 1, and the stride is 1; in the residual block, the convolution kernel width of the convolution layer is 3, the height is 3, the number of channels is 64, the number of convolution kernels is 64, the padding is 1, and the stride is 1; the convolution kernel width of the third convolution layer is 3, the height is 3, the number of channels is 64, the number of convolution kernels is 32, the padding is 1, and the stride is 1; the convolution kernel width of the fourth convolution layer is 3, the height is 3, the number of channels is 32, the number of convolution kernels is 1, the padding is 1, and the stride is 1.

[0013] Specifically, in step S5, the training of the sparse fingerprint image reconstruction network includes: The historical dense fingerprint pixel map and the real-time sparse fingerprint pixel map under the anchor node are input as input data into the sparse fingerprint image reconstruction network, and the real-time dense fingerprint pixel map is calculated and predicted; the loss value is calculated using the mean square error according to the predicted real-time dense fingerprint pixel map and the real-time dense fingerprint pixel map under the anchor node; the weights and biases of the sparse fingerprint image reconstruction network are gradient calculated by the loss value, and the gradient is propagated using the chain rule; based on the update rule of the optimizer, the parameters of the sparse fingerprint image reconstruction network are adjusted, and the cosine annealing learning rate scheduling is adopted, and the learning rate gradually decays during the training process; the real-time sparse fingerprint pixel map and the real-time dense fingerprint pixel map under the anchor node are used as feature values ​​and labels respectively, and are divided into training sets and validation sets according to the set ratio. Each round of iteration will be trained on the training set and the loss will be evaluated on the validation set; the batch size and the number of iterations are set; according to the change of the loss value, the model parameters with the best performance are saved to obtain the optimal sparse fingerprint image reconstruction network corresponding to the anchor node, and the above process is repeated to obtain the optimal sparse fingerprint image reconstruction network corresponding to all anchor nodes.

[0014] Specifically, in step S7, weighted calculation of the reference point coordinates corresponding to the three real-time dense pixel fingerprints includes: using the Euclidean distance to calculate the weights of the three reference point coordinates respectively: , Where: is the weight corresponding to the i-th reference point, is the Euclidean distance between the real-time dense pixel fingerprint of the i-th reference point and the real-time pixel fingerprint, It is a very small positive number greater than 0 to avoid the Euclidean distance being 0; Calculate the coordinates of the target to be located: , Where: is the horizontal coordinate of the target to be located, is the ordinate of the target to be located, is the horizontal coordinate of the i-th reference point, is the ordinate of the i-th reference point.

[0015] The present invention also provides a high-resolution reconstruction positioning system based on sparse fingerprints, comprising: Offline information collection module: used to evenly set reference points in the target area during the offline phase, collect the RSSI values ​​of all anchor nodes in the target area received at each reference point to form the fingerprint of the reference point, and the fingerprints of all reference points constitute a dense fingerprint library; Offline fingerprint library module: used to collect dense fingerprint libraries at several timestamps at fixed time intervals to form an interval dense fingerprint library set; Fingerprint map generation module: used to extract the data in the interval dense fingerprint library according to the anchor node, obtain the dense fingerprint map corresponding to each anchor node at each time stamp, form an interval dense fingerprint map set, set the sparse coefficient to determine the sparse reference point, and extract the sparse fingerprint map corresponding to each dense fingerprint map according to the sparse reference point to form an interval sparse fingerprint map set; Normalization processing module: used to normalize the data in the interval sparse fingerprint atlas and the interval dense fingerprint atlas respectively, to obtain the corresponding interval sparse fingerprint pixel atlas and the interval dense fingerprint pixel atlas; then extract the dense fingerprint pixel map corresponding to the earliest timestamp from the interval dense fingerprint pixel atlas to form the historical dense fingerprint pixel atlas, and the remaining dense fingerprint pixel maps form the real-time dense fingerprint pixel atlas; remove the sparse fingerprint pixel map corresponding to the earliest timestamp from the interval sparse fingerprint pixel atlas, and the remaining sparse fingerprint pixel maps form the real-time sparse fingerprint pixel atlas; Network construction and training module: used to construct a sparse fingerprint image reconstruction network. The historical dense fingerprint pixel atlas and the real-time sparse fingerprint pixel atlas are used as the input of the sparse fingerprint image reconstruction network. For each anchor node, the sparse fingerprint image reconstruction network is trained using the real-time dense fingerprint pixel map to obtain the optimal sparse fingerprint image reconstruction network corresponding to each anchor node. Online data acquisition module: used for collecting RSSI values ​​of all anchor nodes in the target area of ​​the target to be located in the online stage to form the real-time fingerprint of the target to be located, normalizing the real-time fingerprint to obtain the real-time pixel fingerprint, and receiving the RSSI values ​​of all anchor nodes at the sparse reference point to obtain the online sparse fingerprint atlas, which is converted into the online sparse fingerprint pixel atlas after normalization, and the online sparse fingerprint pixel atlas is matched with the historical dense fingerprint pixel atlas, and the matched historical dense fingerprint pixel atlas and the online sparse fingerprint pixel atlas are used as the input of the optimal sparse fingerprint image reconstruction network, and the real-time dense pixel fingerprint set is output; Positioning output module: used to calculate the Euclidean distance between the normalized real-time pixel fingerprint and each real-time dense pixel fingerprint in the real-time dense pixel fingerprint set, sort the calculated Euclidean distances, select several real-time dense pixel fingerprints with the smallest Euclidean distances, and perform weighted calculation on the reference point coordinates corresponding to these several real-time dense pixel fingerprints to obtain the coordinates of the target to be located.

[0016] Beneficial effects: Compared with the prior art, the present invention has the following significant effects: 1. The present invention designs a sparse fingerprint image reconstruction network, combines the historical dense fingerprint pixel map and the sparse fingerprint map obtained by current acquisition and processing for feature learning, realizes the reconstruction of the current high-resolution fingerprint map, improves the positioning accuracy in the target area, and greatly enhances the adaptability of the positioning system in dynamic and complex environments.

[0017] 2. After using the sparse fingerprint image reconstruction network, accurate positioning can be achieved using the historical dense fingerprint pixel atlas, which reduces the update frequency of the fingerprint database and thus reduces the cost of data collection and system maintenance.

[0018] 3. Combining sparse acquisition with deep learning optimizes computing resources, reduces resource consumption during the calculation process, and thus reduces computing resource dependence, making it suitable for lightweight deployment. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of the method of the present invention.

[0020] Figure 2 It is a schematic diagram of a sparse fingerprint pixel map of the present invention.

[0021] Figure 3 It is a flow chart of the sparse fingerprint image reconstruction network of the present invention.

[0022] Figure 4 This is a high-resolution reconstruction effect diagram of the sparse fingerprint image reconstruction network of the present invention.

[0023] Figure 5 It is a schematic diagram of predicting the position of a user to be located according to the present invention. DETAILED DESCRIPTION

[0024] A preferred embodiment of the present invention is further described below in conjunction with the accompanying drawings.

[0025] Example 1

[0026] See also Figure 1 As shown, this embodiment provides a high-resolution reconstruction positioning method based on sparse fingerprints, comprising the following steps: S1. In the offline stage, reference points are evenly set in the target area, and the received signal strength indication RSSI values ​​of all anchor nodes in the target area received at each reference point are collected to form the fingerprint Fp (Fingerprint) of the reference point. The fingerprints of all reference points constitute a dense fingerprint database FD (Fingerprint Database).

[0027] In this embodiment, reference points are evenly set at fixed intervals in the target area, and the number of reference points is: , Where: is the number of reference points, is the length of the target area, is the width of the target area, The fixed interval distance is set.

[0028] Take any reference point i as an example, at timestamp The fingerprint composed of the RSSI values ​​of p anchor nodes received in the positioning area is recorded as : , Where: Indicates the reference point i at timestamp The received signal strength indicator RSSI value of the anchor node j is received.

[0029] S2. Collect a number of dense fingerprint libraries at timestamps at fixed time intervals to form an interval dense fingerprint library set.

[0030] In this embodiment, all m reference points in the positioning area are time stamped The dense fingerprint library composed of fingerprints received from p anchor nodes in the positioning area is : , Where: Indicates the reference point k at timestamp Fingerprint.

[0031] Fingerprint database information is collected at fixed time intervals. The dense fingerprint library collected at x timestamps constitutes the interval dense fingerprint library set : , Where: It represents the dense fingerprint library under timestamp u, and the relationship between timestamps is as follows: .

[0032] S3. Extract the data in the interval dense fingerprint library according to the anchor nodes, obtain the dense fingerprint map corresponding to each anchor node at each timestamp, form an interval dense fingerprint map set, set the sparse coefficient to determine the sparse reference point, and extract the sparse fingerprint map corresponding to each dense fingerprint map according to the sparse reference point to form an interval sparse fingerprint map set.

[0033] In this embodiment, p anchor nodes are set in the target area, and the data in the interval dense fingerprint library is extracted according to the anchor nodes to form corresponding p dense fingerprint maps. For example, the fingerprint library can be extracted to form corresponding p dense fingerprint images. Dense fingerprint map at anchor node j as follows: , Where: The number of reference points set for the long side of the target area. The number of reference points to set for the wide side of the target area.

[0034] Fingerprint Library The p dense fingerprint images formed on the p anchor nodes are merged and used as the timestamp Dense fingerprint atlas under : .

[0035] Since there are p anchor nodes in the positioning area, there are p dense fingerprint maps formed for each timestamp. The dense fingerprint maps under all timestamps are combined to form an interval dense fingerprint map set composed of dense fingerprint maps at x timestamps. : , In order to save positioning energy consumption, the online stage adopts the method of collecting online sparse fingerprints at sparse reference points and reconstructing them into high-resolution dense fingerprints through a deep learning network, which includes: Let the sparse coefficient be , then the corresponding number of reference points selected for the long side and the wide side is: , Where: The number of sparse reference points selected for the long side of the target area, The number of sparse reference points selected for the broad side of the target area; The number of reference points selected for the long side of the target area; The number of reference points to pick for the broad side of the target area.

[0036] Each dense fingerprint image corresponds to a sparse fingerprint image. Taking the dense fingerprint map on anchor node j as an example, the corresponding anchor node j has a timestamp Sparse fingerprint map under as follows: , Where: Represents the total number of sparse reference points.

[0037] Fingerprint Library The p sparse fingerprints formed on the p anchor nodes are combined as timestamps Sparse Fingerprint Atlas : , With interval dense fingerprint collection The interval sparse fingerprint graph set composed of the corresponding sparse fingerprint graph sets at x timestamps : .

[0038] S4. Normalize the data in the interval sparse fingerprint atlas and the interval dense fingerprint atlas respectively to obtain the corresponding interval sparse fingerprint pixel atlas and the interval dense fingerprint pixel atlas.

[0039] In this embodiment, in order to obtain data that is more suitable for deep learning network optimization, it is necessary to collect interval sparse fingerprint graphs. and interval dense fingerprint collection The data in is further processed as follows: The RSSI value is normalized and mapped to the range of [0,1] through normalization processing, and the RSSI value range is set to the range of [-100dBm, 0dBm]. The RSSI value less than -100dBm is set to -100dBm. If the reference point does not collect the RSSI value of the corresponding anchor node, the RSSI value corresponding to the reference point is also set to the lowest RSSI value -100dBm. Through the normalization operation, all RSSI values ​​are converted to the range of [0,1]. The specific conversion formula is: , , , .

[0040] Where: Indicates the minimum element value in the dense fingerprint map of anchor node j at timestamp u, that is, the minimum RSSI value; Indicates the maximum element value in the dense fingerprint graph of anchor node j at timestamp u, that is, the maximum RSSI value; represents the normalized value of RSSI corresponding to anchor node j at dense reference point i at timestamp u; Represents the normalized value of RSSI corresponding to anchor node j at sparse reference point l at timestamp u; Indicates the maximum element value in the sparse fingerprint graph on anchor node j at timestamp u, that is, the maximum RSSI value; It represents the minimum element value in the sparse fingerprint map on anchor node j at timestamp u, that is, the minimum RSSI value; i represents the number of dense reference point sequences; j represents the number of anchor node sequences; l represents the number of sparse reference point sequences.

[0041] Through the above formula and method, the data of dense fingerprint map and sparse fingerprint map are processed to obtain the corresponding dense fingerprint pixel map and sparse fingerprint pixel map for: , , Where: represents the dense fingerprint pixel map under anchor node j at timestamp u; Represents the sparse fingerprint pixel map under anchor node j at timestamp u.

[0042] Correspondingly, the sparse fingerprint pixel maps and dense fingerprint pixel maps under all anchor nodes are integrated to form a sparse fingerprint pixel atlas and dense fingerprint pixel atlas : , Where: represents the dense fingerprint pixel atlas at timestamp u; Represents the sparse fingerprint pixel atlas at timestamp u.

[0043] Correspondingly, the dense fingerprint pixel atlas at x timestamps is composed of an interval dense fingerprint pixel map set and the interval sparse fingerprint pixel map set composed of the sparse fingerprint pixel map sets at x timestamps It can be represented by the following set: , Split the sparse fingerprint pixel map at x timestamps according to different anchor nodes: , Where: Represents the sparse fingerprint pixel atlas under anchor node j.

[0044] Split the dense fingerprint pixel map at x timestamps according to different anchor nodes: , Where: represents the dense fingerprint pixel atlas under anchor node j.

[0045] Please refer to Figure 2 As shown, Figure 2 For anchor node 2 at timestamp t 23 The sparse fingerprint pixel map under is 16×16 in size and the sparse coefficient is , , anchor node 2 at timestamp t 23 Sparse fingerprint map under and anchor node 2 at timestamp t 23 Sparse fingerprint pixel map under The matrix representation of is as follows: , .

[0046] Where: Indicates missing data.

[0047] It can be seen that through the normalization operation, the missing values ​​of the RSSI values ​​are completed, and all RSSI values ​​are normalized to the range of [0,1].

[0048] S5. Extract the dense fingerprint pixel map corresponding to the earliest timestamp from the interval dense fingerprint pixel map set to form a historical dense fingerprint pixel map set, and the remaining dense fingerprint pixel maps constitute a real-time dense fingerprint pixel map set; remove the sparse fingerprint pixel map corresponding to the earliest timestamp from the interval sparse fingerprint pixel map set, and the remaining sparse fingerprint pixel maps constitute a real-time sparse fingerprint pixel map set; construct a sparse fingerprint image reconstruction network (SFIR, Sparse Fingerprint Image Reproduction), use the historical dense fingerprint pixel map set and the real-time sparse fingerprint pixel map set as inputs of the sparse fingerprint image reconstruction network, and for each anchor node, use the real-time dense fingerprint pixel map to train the sparse fingerprint image reconstruction network to obtain the optimal sparse fingerprint image reconstruction network corresponding to each anchor node.

[0049] Taking anchor node j as an example, extract the dense fingerprint pixel atlas under anchor node j and the corresponding sparse fingerprint pixel atlas , select the dense fingerprint pixel map corresponding to the earliest timestamp among the x timestamps in the dense fingerprint pixel map set under anchor node j As the historical dense fingerprint pixel map under anchor node j, the data size is: ; The remaining x-1 dense fingerprint pixel images are used as the real-time dense fingerprint pixel images under anchor node j, and the data size is: ; Remove the earliest timestamp from the x timestamps in the sparse fingerprint pixel map under anchor node j, and use the remaining x-1 sparse fingerprint pixel maps as the real-time sparse fingerprint pixel maps under anchor node j. The data size is: .

[0050] The following takes anchor node j as an example to explain the specific structure of the SFIR network: Please refer to Figure 3 As shown, the SFIR network includes an upsampling layer, a channel merging layer, a convolutional coding layer, a feature fusion module and a convolutional decoding layer connected in sequence; the convolutional coding layer includes a first convolutional layer and a second convolutional layer connected in sequence; the feature fusion module includes four residual blocks connected in sequence, each residual block includes two convolutional layers; the convolutional decoding layer includes a third convolutional layer and a fourth convolutional layer connected in sequence.

[0051] The input of the upsampling layer is: the historical dense fingerprint pixel map under anchor node j and the real-time sparse fingerprint pixel map under anchor node j; the output is: the real-time upsampled fingerprint pixel map under anchor node j after upsampling and the historical dense fingerprint pixel map under anchor node j; the input data size is: , the output data size is: The function of this layer is to upsample the real-time sparse fingerprint pixel map under anchor node j from low resolution Upsampling to high resolution , in order to align with the historical dense fingerprint pixel map.

[0052] The input of the channel merging layer is: the real-time upsampled fingerprint pixel map under the upsampled anchor node j and the historical dense fingerprint pixel map under the anchor node j; the input data size is: and ;The output data size is: ; The function of this layer is to merge the two pixel images in the channel direction to form a two-channel input tensor.

[0053] First convolutional layer: Input: , output: , convolution kernel size: , indicating that the width of the convolution kernel is 3, the height is 3, the number of channels is 2, and the number of convolution kernels is 32. This representation is used below, with a padding of 1, a step size of 1, and an activation function set to the ReLU function. The function of this layer is to extract shallow features and extract features from the merged image through the convolution kernel.

[0054] Second convolutional layer: Input: , output: , convolution kernel size: , the padding is 1, the step size is 1, and the activation function is set to the ReLU function. The function of this layer is to extract deeper features and deepen the network feature representation through the convolution kernel.

[0055] Feature fusion module: contains four residual blocks with the same structure, input: , output: , each residual block contains two convolutional layers, and the convolution kernel size is: , the padding is 1, the step size is 1, the activation function of the previous convolutional layer is the ReLU function, and no activation function is set in the subsequent convolutional layer, which is used to skip the residual block; repeat 4 residual blocks, and the input and output sizes are always kept ,The role of this layer is to retain the input features, while learning the residual information,,and fusing deep and shallow features.

[0056] The third convolutional layer: Input: , output: , convolution kernel size: , the padding is 1, the step size is 1, and the activation function is the ReLU function. The function of this layer is to reduce the high-dimensional features to low-dimensional features in preparation for the output stage.

[0057] Fourth convolutional layer: Input: , output: , convolution kernel size: , the filling is 1, the step size is 1, and the activation function is a linear function. The function of this layer is to generate the predicted real-time dense fingerprint pixel map under the final predicted anchor node j.

[0058] Output layer: Output size: ,The function of this layer is to output the predicted real-time dense fingerprint pixel map under the anchor node j.

[0059] SFIR network training phase: For p anchor nodes, p SFIR networks need to be trained specifically. The following takes anchor node j as an example to illustrate the training process: The loss function selects the mean square error loss function; the optimizer selects Adam, and the initial learning rate is set to 0.001; the historical dense fingerprint pixel map under anchor node j and the real-time sparse fingerprint pixel map under anchor node j are input as input data into the corresponding SFIR network of anchor node j, and the predicted output is calculated: the predicted real-time dense fingerprint pixel map under anchor node j; the loss value is calculated using the mean square error (MSE) based on the predicted output and the true value: the real-time dense fingerprint pixel map under anchor node j; the gradient of the network weights and biases is calculated through the loss value, and the gradient is propagated using the chain rule; based on the update rule of the optimizer (Adam), the network parameters are adjusted, the learning rate gradually decays during the training process, and the cosine annealing learning rate scheduling is adopted.

[0060] The real-time sparse fingerprint pixel map under anchor node j and the real-time dense fingerprint pixel map under anchor node j are used as feature values ​​and labels respectively, and are divided into training set and validation set in a ratio of 8:2. Each round of iteration (Epoch) will be trained on the training set and the loss will be evaluated on the validation set. The batch size is set to 32 and the number of iterations (Epoch) is set to 30. According to the change of the validation loss, the model parameters with the best performance are saved, and the early stopping strategy is used to avoid overfitting.

[0061] Using the same method as above, we can get the SFIR network corresponding to all anchor nodes to determine the optimal model parameters, denoted as : .

[0062] Where: Represents the SFIR network that determines the optimal model parameters corresponding to the j-th anchor node.

[0063] Please refer to Figure 4 As shown, Figure 2 The corresponding anchor node 2 is at timestamp t 23 The sparse fingerprint pixel map under the SFIR network is reconstructed with high resolution to obtain the predicted real-time dense fingerprint pixel map, and the size is restored to 64×64.

[0064] S6, in the online stage, the target to be located collects RSSI values ​​on all anchor nodes in the target area to form a real-time fingerprint of the target to be located, normalizes the real-time fingerprint to obtain a real-time pixel fingerprint, and simultaneously receives RSSI values ​​on all anchor nodes at a sparse reference point to obtain an online sparse fingerprint atlas, which is converted into an online sparse fingerprint pixel atlas after normalization, and the online sparse fingerprint pixel atlas is matched with the historical dense fingerprint pixel atlas, and the matched historical dense fingerprint pixel atlas and online sparse fingerprint pixel atlas are used as inputs of the optimal sparse fingerprint image reconstruction network, and the real-time dense pixel fingerprint set is output.

[0065] In this embodiment, the user to be located collects the RSSI values ​​corresponding to p anchor nodes in the positioning area as the real-time fingerprint. : , Where: Indicates the RSSI value of anchor node j received by the user to be located.

[0066] Normalize the real-time fingerprint to obtain the corresponding normalized real-time fingerprint : , Where: Indicates the normalized value of the RSSI of anchor node j received by the user to be located.

[0067] At the same time, a positioning request is sent to the positioning system, and the sparse reference point receives the RSSI values ​​of all anchor nodes to form p online sparse fingerprint maps, which constitute the online sparse fingerprint atlas. : , Where: Represents the online sparse fingerprint graph under the anchor node p.

[0068] By using the sparse fingerprint pixel map conversion method recorded in step S4, the online sparse fingerprint map is converted into an online sparse fingerprint pixel map. Taking anchor node j as an example, the online sparse fingerprint pixel map under anchor node j is expressed as: ; Then the online sparse fingerprint pixel images under the corresponding p anchor nodes constitute the online sparse fingerprint pixel atlas : .

[0069] The online sparse fingerprint pixel atlas and The historical dense fingerprint pixel atlas saved in the network is matched, and the corresponding historical dense fingerprint pixel atlas is : , Where: Represents the historical dense fingerprint pixel map under the anchor node p at the earliest timestamp.

[0070] The matched data is used as online network input : , Where: Represents the input data of the SFIR network corresponding to the jth anchor node.

[0071] Will Bring in A real-time dense pixel fingerprint set can be obtained : , Where: represents the real-time pixel fingerprint at the reference point r.

[0072] S7. Calculate the Euclidean distance between the normalized real-time pixel fingerprint and each real-time dense pixel fingerprint in the real-time dense pixel fingerprint set, sort the calculated Euclidean distances, select several real-time dense pixel fingerprints with the smallest Euclidean distances, perform weighted calculation on the reference point coordinates corresponding to these several real-time dense pixel fingerprints, and obtain the coordinates of the target to be located.

[0073] In this embodiment, the normalized real-time pixel fingerprint and each real-time pixel fingerprint in the real-time dense pixel fingerprint set are calculated using the following formula: , Where: It represents the Euclidean distance between the real-time pixel fingerprint at the reference point r and the normalized real-time pixel fingerprint.

[0074] The calculated Euclidean distances are sorted. In this embodiment, the reference points corresponding to the three real-time pixel fingerprints with the smallest Euclidean distances are selected: , The corresponding reference point coordinates are , , ,in, is the index of the reference point in the real-time dense pixel fingerprint set; calculate each distance Weight : , Where: Represents a very small positive number to avoid the division by zero problem when the distance is zero.

[0075] The corresponding coordinates of the user to be located are calculated as : , Where: is the horizontal coordinate of the target to be located, is the ordinate of the target to be located, is the horizontal coordinate of the i-th reference point, is the ordinate of the i-th reference point.

[0076] Please refer to Figure 5As shown in FIG. 1 , a position prediction map of a user to be located is obtained by applying the above solution in a specific scenario of the present invention. The normalized real-time pixel fingerprint of the user to be located is: [0.2, 0.3, 0.4, 0.1]; a partial real-time pixel fingerprint set of the normalized reference point is shown in Table 1 below: Table 1: Real-time pixel fingerprint set (normalized) Reference Points Reference point A Reference point B Reference point C Reference point D Reference point E Real-time pixel fingerprint [0.1,0.3,0.5,0.0] [0.2,0.4,0.3,0.2] [0.3,0.2,0.4,0.1] [0.0,0.3,0.6, 0.1] [0.2,0.3,0.5,0.0] …… coordinate (2,3) (5,8) (8,6) (1,1) (3,7) …… Use the above formula to calculate the Euclidean distance point by point : .

[0077] The three reference points with the smallest Euclidean distance are sorted from small to large as reference point C, reference point E, and reference point A. The weights are calculated according to the above formula. Let , and the corresponding weights are: .

[0078] According to the above formula for solving the coordinates of the user to be located, the coordinates of the user to be located are: (x, y) = (4.49, 5.55). The specific position relationship is as follows: Figure 5 The method provided by the present invention can complete high-precision positioning tasks with low energy consumption and high robustness, and has far-reaching research significance and broad application prospects.

[0079] Example 2 This embodiment provides a high-resolution reconstruction positioning system based on sparse fingerprints corresponding to the high-resolution reconstruction positioning method described in Embodiment 1, including: Offline information collection module: used to evenly set reference points in the target area during the offline phase, collect the RSSI values ​​of all anchor nodes in the target area received at each reference point to form the fingerprint of the reference point, and the fingerprints of all reference points constitute a dense fingerprint library; Offline fingerprint library module: used to collect dense fingerprint libraries at several timestamps at fixed time intervals to form an interval dense fingerprint library set; Fingerprint map generation module: used to extract the data in the interval dense fingerprint library according to the anchor node, obtain the dense fingerprint map corresponding to each anchor node at each time stamp, form an interval dense fingerprint map set, set the sparse coefficient to determine the sparse reference point, and extract the sparse fingerprint map corresponding to each dense fingerprint map according to the sparse reference point to form an interval sparse fingerprint map set; Normalization processing module: used to normalize the data in the interval sparse fingerprint atlas and the interval dense fingerprint atlas respectively, to obtain the corresponding interval sparse fingerprint pixel atlas and the interval dense fingerprint pixel atlas; Network construction and training module: used to extract the dense fingerprint pixel map corresponding to the earliest timestamp from the interval dense fingerprint pixel map set to form a historical dense fingerprint pixel map set, and the remaining dense fingerprint pixel maps constitute a real-time dense fingerprint pixel map set; remove the sparse fingerprint pixel map corresponding to the earliest timestamp from the interval sparse fingerprint pixel map set, and the remaining sparse fingerprint pixel maps constitute a real-time sparse fingerprint pixel map set; construct a sparse fingerprint image reconstruction network, use the historical dense fingerprint pixel map set and the real-time sparse fingerprint pixel map set as inputs of the sparse fingerprint image reconstruction network, and for each anchor node, use the real-time dense fingerprint pixel map to train the sparse fingerprint image reconstruction network to obtain the optimal sparse fingerprint image reconstruction network corresponding to each anchor node; Online data acquisition module: used for collecting RSSI values ​​of all anchor nodes in the target area of ​​the target to be located in the online stage to form the real-time fingerprint of the target to be located, normalizing the real-time fingerprint to obtain the real-time pixel fingerprint, and receiving the RSSI values ​​of all anchor nodes at the sparse reference point to obtain the online sparse fingerprint atlas, which is converted into the online sparse fingerprint pixel atlas after normalization, and the online sparse fingerprint pixel atlas is matched with the historical dense fingerprint pixel atlas, and the matched historical dense fingerprint pixel atlas and the online sparse fingerprint pixel atlas are used as the input of the optimal sparse fingerprint image reconstruction network, and the real-time dense pixel fingerprint set is output; Positioning output module: used to calculate the Euclidean distance between the normalized real-time pixel fingerprint and each real-time dense pixel fingerprint in the real-time dense pixel fingerprint set, sort the calculated Euclidean distances, select several real-time dense pixel fingerprints with the smallest Euclidean distances, and perform weighted calculation on the reference point coordinates corresponding to these several real-time dense pixel fingerprints to obtain the coordinates of the target to be located.

Claims

1. A high-resolution reconstruction positioning method based on sparse fingerprints, characterized in that: The following steps are involved: S1, in the offline stage, reference points are evenly set in the target area and the RSSI values ​​of all anchor nodes are received to obtain the fingerprints of the reference points. The fingerprints of all reference points constitute a dense fingerprint library; S2, collecting dense fingerprint libraries at several timestamps to form an interval dense fingerprint library set; S3, extracting data from the interval dense fingerprint library, obtaining a dense fingerprint map of each anchor node, forming an interval dense fingerprint map set, setting a sparse reference point, and extracting a sparse fingerprint map corresponding to the dense fingerprint map, forming an interval sparse fingerprint map set; S4, normalizing the interval sparse fingerprint atlas and the interval dense fingerprint atlas to obtain the interval sparse fingerprint pixel atlas and the interval dense fingerprint pixel atlas, respectively, and extracting the historical dense fingerprint pixel atlas, the real-time dense fingerprint pixel atlas and the real-time sparse fingerprint pixel atlas; S5, constructing a sparse fingerprint image reconstruction SFIR network, taking the historical dense fingerprint pixel atlas and the real-time sparse fingerprint pixel atlas as input, using the real-time dense fingerprint pixel atlas to train the SFIR network, and obtaining the optimal SFIR network corresponding to each anchor node; S6, in the online stage, the real-time pixel fingerprint of the target to be located is obtained, and the online sparse fingerprint pixel atlas corresponding to the sparse reference point is obtained, the online sparse fingerprint pixel atlas is matched with the historical dense fingerprint pixel atlas, and the atlas is used as the input of the corresponding optimal SFIR network, and the real-time dense pixel fingerprint set is output; S7. Calculate the Euclidean distance between the real-time pixel fingerprint and each real-time dense pixel fingerprint in the real-time dense pixel fingerprint set, select several real-time dense pixel fingerprints with the smallest Euclidean distance, perform weighted calculation on the corresponding reference point coordinates, and obtain the coordinates of the target to be located.

2. The high-resolution reconstruction positioning method based on sparse fingerprint according to claim 1 is characterized in that: In step S1, reference points are evenly set in the target area, and the number of reference points is: ， Where: is the number of reference points, is the length of the target area, is the width of the target area, The fixed interval distance is set.

3. The high-resolution reconstruction positioning method based on sparse fingerprint according to claim 1, characterized in that: In step S3, setting the sparse reference points includes: setting a sparse coefficient, and determining the number of sparse reference points selected for the long side and the wide side of the target area according to the sparse coefficient: ， Where: The number of sparse reference points selected for the long side of the target area, The number of sparse reference points selected for the broad side of the target area; The number of reference points selected for the long side of the target area; The number of reference points selected for the broadside of the target area; is the sparse coefficient.

4. The high-resolution reconstruction positioning method based on sparse fingerprint according to claim 1, characterized in that: In step S4, the normalization process includes: The RSSI value range is set to [-100dBm, 0dBm], and the RSSI value less than -100dBm is set to -100dBm. If the RSSI value of the corresponding anchor node is not collected at the reference point or sparse reference point, the missing RSSI value is also set to -100dBm, and then all RSSI values ​​are converted to the range of [0, 1] through normalization.

5. The high-resolution reconstruction positioning method based on sparse fingerprint according to claim 1, characterized in that: In step S5: the sparse fingerprint image reconstruction network includes an upsampling layer, a channel merging layer, a convolutional coding layer, a feature fusion module and a convolutional decoding layer connected in sequence; the convolutional coding layer includes a first convolutional layer and a second convolutional layer connected in sequence; the feature fusion module includes four residual blocks connected in sequence, each residual block includes two convolutional layers; the convolutional decoding layer includes a third convolutional layer and a fourth convolutional layer connected in sequence.

6. The high-resolution reconstruction positioning method based on sparse fingerprint according to claim 5, characterized in that: In the sparse fingerprint image reconstruction network, the upsampling layer performs upsampling processing on the real-time sparse fingerprint pixel map, and aligns the resolution of the real-time sparse fingerprint pixel map with the historical dense fingerprint pixel map; the channel merging layer merges the real-time sparse fingerprint pixel map with the historical dense fingerprint pixel map in the channel direction; the first convolution layer extracts shallow features from the merged pixel map; the second convolution layer extracts deep features in the pixel map; the feature fusion module fuses deep features with shallow features; the third convolution layer reduces the dimension of high-dimensional features; and the fourth convolution layer generates a predicted real-time dense fingerprint pixel map.

7. The high-resolution reconstruction positioning method based on sparse fingerprint according to claim 1, characterized in that: In the step S4, respectively extracting the historical dense fingerprint pixel atlas, the real-time dense fingerprint pixel atlas and the real-time sparse fingerprint pixel atlas includes: extracting the dense fingerprint pixel map corresponding to the earliest timestamp from the interval dense fingerprint pixel atlas to form the historical dense fingerprint pixel atlas, and the remaining dense fingerprint pixel maps constitute the real-time dense fingerprint pixel atlas; removing the sparse fingerprint pixel map corresponding to the earliest timestamp from the interval sparse fingerprint pixel atlas, and the remaining sparse fingerprint pixel maps constitute the real-time sparse fingerprint pixel atlas.

8. The high-resolution reconstruction positioning method based on sparse fingerprint according to claim 1, characterized in that: In step S5, the training of the sparse fingerprint image reconstruction network includes: The historical dense fingerprint pixel map and the real-time sparse fingerprint pixel map under the anchor node are input as input data into the sparse fingerprint image reconstruction network, and the real-time dense fingerprint pixel map is calculated and predicted; the loss value is calculated using the mean square error according to the predicted real-time dense fingerprint pixel map and the real-time dense fingerprint pixel map under the anchor node; the weights and biases of the sparse fingerprint image reconstruction network are gradient calculated by the loss value, and the gradient is propagated using the chain rule; based on the update rule of the optimizer, the parameters of the sparse fingerprint image reconstruction network are adjusted, and the cosine annealing learning rate scheduling is adopted, and the learning rate gradually decays during the training process; the real-time sparse fingerprint pixel map and the real-time dense fingerprint pixel map under the anchor node are used as feature values ​​and labels respectively, and are divided into training sets and validation sets according to the set ratio. Each round of iteration will be trained on the training set and the loss will be evaluated on the validation set; the batch size and the number of iterations are set; according to the change of the loss value, the model parameters with the best performance are saved to obtain the optimal sparse fingerprint image reconstruction network corresponding to the anchor node, and the above process is repeated to obtain the optimal sparse fingerprint image reconstruction network corresponding to all anchor nodes.

9. The high-resolution reconstruction positioning method based on sparse fingerprint according to claim 1, characterized in that: In step S7, weighted calculation of the reference point coordinates corresponding to the three real-time dense pixel fingerprints includes: using Euclidean distance to calculate the weights of the three reference point coordinates respectively: ， Where: is the weight corresponding to the i-th reference point, is the Euclidean distance between the real-time dense pixel fingerprint of the i-th reference point and the real-time pixel fingerprint, It is a very small positive number greater than 0, which avoids the Euclidean distance being 0; Calculate the coordinates of the target to be located: ， Where: is the horizontal coordinate of the target to be located, is the ordinate of the target to be located, is the horizontal coordinate of the i-th reference point, is the ordinate of the i-th reference point.

10. A high-resolution reconstruction positioning system based on sparse fingerprints, characterized in that: include: Offline information collection module: used to evenly set reference points in the target area during the offline phase, collect the RSSI values ​​of all anchor nodes in the target area received at each reference point to form the fingerprint of the reference point, and the fingerprints of all reference points constitute a dense fingerprint library; Offline fingerprint library module: used to collect dense fingerprint libraries at several timestamps at fixed time intervals to form an interval dense fingerprint library set; Fingerprint map generation module: used to extract the data in the interval dense fingerprint library according to the anchor node, obtain the dense fingerprint map corresponding to each anchor node at each time stamp, form an interval dense fingerprint map set, set the sparse coefficient to determine the sparse reference point, and extract the sparse fingerprint map corresponding to each dense fingerprint map according to the sparse reference point to form an interval sparse fingerprint map set; Normalization processing module: used to normalize the data in the interval sparse fingerprint atlas and the interval dense fingerprint atlas respectively, to obtain the corresponding interval sparse fingerprint pixel atlas and the interval dense fingerprint pixel atlas; then extract the dense fingerprint pixel map corresponding to the earliest timestamp from the interval dense fingerprint pixel atlas to form the historical dense fingerprint pixel atlas, and the remaining dense fingerprint pixel maps form the real-time dense fingerprint pixel atlas; remove the sparse fingerprint pixel map corresponding to the earliest timestamp from the interval sparse fingerprint pixel atlas, and the remaining sparse fingerprint pixel maps form the real-time sparse fingerprint pixel atlas; Network construction and training module: used to construct a sparse fingerprint image reconstruction network. The historical dense fingerprint pixel atlas and the real-time sparse fingerprint pixel atlas are used as the input of the sparse fingerprint image reconstruction network. For each anchor node, the sparse fingerprint image reconstruction network is trained using the real-time dense fingerprint pixel map to obtain the optimal sparse fingerprint image reconstruction network corresponding to each anchor node. Online data acquisition module: used for collecting RSSI values ​​of all anchor nodes in the target area of ​​the target to be located in the online stage to form the real-time fingerprint of the target to be located, normalizing the real-time fingerprint to obtain the real-time pixel fingerprint, and receiving the RSSI values ​​of all anchor nodes at the sparse reference point to obtain the online sparse fingerprint atlas, which is converted into the online sparse fingerprint pixel atlas after normalization, and the online sparse fingerprint pixel atlas is matched with the historical dense fingerprint pixel atlas, and the matched historical dense fingerprint pixel atlas and the online sparse fingerprint pixel atlas are used as the input of the optimal sparse fingerprint image reconstruction network, and the real-time dense pixel fingerprint set is output; Positioning output module: used to calculate the Euclidean distance between the normalized real-time pixel fingerprint and each real-time dense pixel fingerprint in the real-time dense pixel fingerprint set, sort the calculated Euclidean distances, select several real-time dense pixel fingerprints with the smallest Euclidean distances, and perform weighted calculation on the reference point coordinates corresponding to these several real-time dense pixel fingerprints to obtain the coordinates of the target to be located.

Images