Long-distance WiFi signal positioning system and method

Abstract

The present invention discloses a long-distance WiFi signal positioning method and device, the method comprising: obtaining channel state information and round-trip time distance information; performing beam synthesis on the channel state information to obtain positioning feature parameter information; using the round-trip time distance information and the positioning feature parameter information to train a preset positioning model to obtain an optimized positioning model; using the optimized positioning model to process the channel state information and round-trip time distance information to be processed to obtain a positioning result. The method of the present invention comprehensively utilizes channel state information, round-trip time distance information and beam synthesis method to enhance the signal-to-noise ratio of the received signal, obtain a more accurate WiFi signal, effectively improve the detection distance of the WiFi signal, and improve the positioning accuracy.

Description

Technical Field

[0001] The present invention relates to the field of signal detection technology, and in particular to a long-distance WiFi signal positioning system and method. Background Art

[0002] Wireless communication has become an indispensable communication method in people's daily production and life, bringing endless convenience to people's production and life. In the process of upgrading and renovating the WiFi network, it is necessary to detect the WiFi network signal in the venue; the existing detection method is as follows: record the signal once every certain distance in the venue, and send all the records to the detection and analysis software for analysis, and finally analyze the overall WiFi network signal in the venue, which requires a lot of manpower and material resources and takes a long time. WiFi positioning technology has become a research hotspot due to the high popularity of wireless local area networks. Traditional WiFi positioning technology uses received signal strength indication (RSSI) as a feature parameter, but due to the coarse granularity and high variability of RSSI, it cannot achieve high-precision positioning in complex indoor environments, and the positioning distance is limited. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a long-distance WiFi signal positioning system and method, taking channel state information as the research object, and using the round-trip time ranging results provided by the 802.11 protocol to carry out positioning research based on the fusion of channel state information and round-trip time ranging information. Focusing on improving the hidden detection capability of near-field WiFi network signals, focusing on solving the problems of fast attenuation of WiFi signal transmission process and weak signal detection capability of traditional WiFi detection equipment, the radio frequency performance is greatly improved through beamforming technology, the security, convenience and success rate of WiFi detection and access are improved, and the positioning accuracy based on WiFi signal positioning is improved.

[0004] In order to solve the above technical problems, the first aspect of the embodiment of the present invention discloses a long-distance WiFi signal positioning method, the method comprising:

[0005] S1, obtain channel state information and round-trip time distance information;

[0006] S2, performing beam synthesis on the channel state information to obtain positioning feature parameter information;

[0007] S3, using the round-trip time distance information and the positioning feature parameter information, training a preset positioning model to obtain an optimized positioning model;

[0008] S4, using the optimized positioning model, processing the channel state information and round-trip time distance information to be processed to obtain a positioning result.

[0009] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the performing beamforming on the channel state information to obtain positioning feature parameter information includes:

[0010] S21, performing Hampel filtering on the channel state information to obtain filtering state information;

[0011] S22, performing phase extension on the filtering state information to obtain extended state information;

[0012] S23, performing a phase linear transformation on the extended state information to obtain linear phase state information;

[0013] S24, performing beam synthesis on the linear phase state information to obtain channel state information;

[0014] S25, extracting features from the channel state information to obtain positioning feature parameter information.

[0015] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the extracting features of the channel state information to obtain positioning feature parameter information includes:

[0016] S251, filtering the channel state information to obtain smoothed channel state information;

[0017] S252, enhancing the smoothed channel state information to obtain enhanced channel state information;

[0018] S253: extract features from the enhanced channel state information to obtain positioning feature parameter information.

[0019] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the enhancing the smoothed channel state information to obtain enhanced channel state information includes:

[0020] S2521, processing the smoothed channel state information to obtain smoothed channel state amplitude information and smoothed channel state phase information;

[0021] S2522, processing the smoothed channel state amplitude information and the smoothed channel state phase information to obtain reconstructed amplitude information and reconstructed phase information;

[0022] S2523: Process the reorganized amplitude information and the reorganized phase information to obtain enhanced channel state information.

[0023] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the step of extracting features from the enhanced channel state information to obtain positioning feature parameter information includes:

[0024] S2531, preprocessing the enhanced channel state information to obtain preprocessed enhanced channel state information;

[0025] S2532, processing the preprocessed enhanced channel state information to obtain fuzzy parameter information;

[0026] S2533, performing Hadamard product calculation on the fuzzy parameter information and a preset kernel function to obtain optimized fuzzy parameter information;

[0027] S2534, processing the optimized fuzzy parameter information to obtain positioning feature parameter information.

[0028] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the use of the round-trip time distance information and the positioning feature parameter information to train a preset positioning model to obtain an optimized positioning model includes:

[0029] S31, extracting features from the round-trip time distance information to obtain distance parameter information;

[0030] S32, fusing the distance parameter information and the positioning feature parameter information to obtain positioning parameter information;

[0031] S33, using the positioning parameter information, training a preset positioning model to obtain an optimized positioning model.

[0032] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the fusing the distance parameter information and the positioning feature parameter information to obtain the positioning parameter information includes:

[0033] S321, processing the distance parameter information to obtain a distance covariance matrix;

[0034] S322, processing the positioning feature parameter information to obtain a positioning covariance matrix;

[0035] S323, processing the distance parameter information and the positioning feature parameter information to obtain a first cross-covariance matrix and a second cross-covariance matrix;

[0036] S324, processing the distance covariance matrix, the positioning covariance matrix, the first cross-covariance matrix, and the second cross-covariance matrix to obtain a feature fusion matrix;

[0037] S325, processing the feature fusion matrix to obtain a first singular vector and a second singular vector;

[0038] S326, processing the first singular vector and the second singular vector to obtain a distance projection vector and a positioning projection vector;

[0039] S327: Process the distance projection vector and the positioning projection vector to obtain positioning parameter information.

[0040] A second aspect of an embodiment of the present invention discloses a long-distance WiFi signal positioning device, the device comprising:

[0041] An information acquisition module, used to acquire channel state information and round-trip time distance information;

[0042] A beamforming module, used to perform beamforming on the channel state information to obtain positioning feature parameter information;

[0043] A model training module, used to train a preset positioning model using the round-trip time distance information and the positioning feature parameter information to obtain an optimized positioning model;

[0044] The positioning module is used to process the channel state information and round-trip time distance information to be processed by using the optimized positioning model to obtain a positioning result.

[0045] As an optional implementation manner, in the second aspect of the embodiment of the present invention, the performing beamforming on the channel state information to obtain positioning feature parameter information includes:

[0046] S21, performing Hampel filtering on the channel state information to obtain filtering state information;

[0047] S22, performing phase extension on the filtering state information to obtain extended state information;

[0048] S23, performing a phase linear transformation on the extended state information to obtain linear phase state information;

[0049] S24, performing beam synthesis on the linear phase state information to obtain channel state information;

[0050] S25, extracting features from the channel state information to obtain positioning feature parameter information.

[0051] As an optional implementation manner, in the second aspect of the embodiment of the present invention, the extracting features of the channel state information to obtain positioning feature parameter information includes:

[0052] S251, filtering the channel state information to obtain smoothed channel state information;

[0053] S252, enhancing the smoothed channel state information to obtain enhanced channel state information;

[0054] S253: extract features from the enhanced channel state information to obtain positioning feature parameter information.

[0055] As an optional implementation manner, in the second aspect of the embodiment of the present invention, the enhancing the smoothed channel state information to obtain enhanced channel state information includes:

[0056] S2521, processing the smoothed channel state information to obtain smoothed channel state amplitude information and smoothed channel state phase information;

[0057] S2522, processing the smoothed channel state amplitude information and the smoothed channel state phase information to obtain reconstructed amplitude information and reconstructed phase information;

[0058] S2523: Process the reorganized amplitude information and the reorganized phase information to obtain enhanced channel state information.

[0059] As an optional implementation manner, in the second aspect of the embodiment of the present invention, the feature extraction of the enhanced channel state information to obtain the positioning feature parameter information includes:

[0060] S2531, preprocessing the enhanced channel state information to obtain preprocessed enhanced channel state information;

[0061] S2532, processing the preprocessed enhanced channel state information to obtain fuzzy parameter information;

[0062] S2533, performing Hadamard product calculation on the fuzzy parameter information and a preset kernel function to obtain optimized fuzzy parameter information;

[0063] S2534, processing the optimized fuzzy parameter information to obtain positioning feature parameter information.

[0064] As an optional implementation manner, in the second aspect of the embodiment of the present invention, the use of the round-trip time distance information and the positioning feature parameter information to train a preset positioning model to obtain an optimized positioning model includes:

[0065] S31, extracting features from the round-trip time distance information to obtain distance parameter information;

[0066] S32, fusing the distance parameter information and the positioning feature parameter information to obtain positioning parameter information;

[0067] S33, using the positioning parameter information, training a preset positioning model to obtain an optimized positioning model.

[0068] As an optional implementation manner, in the second aspect of the embodiment of the present invention, the fusing of the distance parameter information and the positioning feature parameter information to obtain the positioning parameter information includes:

[0069] S321, processing the distance parameter information to obtain a distance covariance matrix;

[0070] S322, processing the positioning feature parameter information to obtain a positioning covariance matrix;

[0071] S323, processing the distance parameter information and the positioning feature parameter information to obtain a first cross-covariance matrix and a second cross-covariance matrix;

[0072] S324, processing the distance covariance matrix, the positioning covariance matrix, the first cross-covariance matrix, and the second cross-covariance matrix to obtain a feature fusion matrix;

[0073] S325, processing the feature fusion matrix to obtain a first singular vector and a second singular vector;

[0074] S326, processing the first singular vector and the second singular vector to obtain a distance projection vector and a positioning projection vector;

[0075] S327: Process the distance projection vector and the positioning projection vector to obtain positioning parameter information.

[0076] The third aspect of the present invention discloses another long-distance WiFi signal positioning device, the device comprising:

[0077] A memory storing executable program code;

[0078] a processor coupled to the memory;

[0079] The processor calls the executable program code stored in the memory to execute part or all of the steps in the long-distance WiFi signal positioning method disclosed in the first aspect of the embodiment of the present invention.

[0080] The fourth aspect of the present invention discloses a computer-storable medium, which stores computer instructions. When the computer instructions are called, they are used to execute some or all of the steps in the long-distance WiFi signal positioning method disclosed in the first aspect of an embodiment of the present invention.

[0081] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0082] The present invention discloses a long-distance WiFi signal positioning method, which effectively improves the detection distance of WiFi signals by performing beam synthesis on WiFi signals; enhances the signal-to-noise ratio of received signals by comprehensively utilizing channel state information, round-trip time distance information and a beam synthesis method, obtains more accurate WiFi signals, and effectively improves positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0084] Figure 1 It is a flowchart of a long-distance WiFi signal positioning method disclosed in an embodiment of the present invention;

[0085] Figure 2 It is a schematic diagram of a process of detecting WiFi signals using beamforming technology disclosed in an embodiment of the present invention;

[0086] Figure 3 is a schematic diagram of baseband signal processing disclosed in an embodiment of the present invention;

[0087] Figure 4 is a schematic diagram of a PCB stacking solution disclosed in an embodiment of the present invention;

[0088] Figure 5 It is a structural schematic diagram of a long-distance WiFi signal positioning device disclosed in an embodiment of the present invention;

[0089] Figure 6 It is a structural schematic diagram of another long-distance WiFi signal positioning device disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0090] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0091] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or equipment.

[0092] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0093] The present invention discloses a long-distance WiFi signal positioning method and device, the method comprising: obtaining channel state information and round-trip time distance information; performing beam synthesis on the channel state information to obtain positioning feature parameter information; using the round-trip time distance information and the positioning feature parameter information to train a preset positioning model to obtain an optimized positioning model; using the optimized positioning model to process the channel state information and round-trip time distance information to be processed to obtain a positioning result. The method of the present invention comprehensively utilizes channel state information, round-trip time distance information and beam synthesis method to enhance the signal-to-noise ratio of the received signal, obtain a more accurate WiFi signal, effectively improve the detection distance of the WiFi signal, and improve the positioning accuracy. The following are detailed descriptions.

[0094] Embodiment 1

[0095] See also Figure 1 , Figure 1 : is a flow chart of a long-distance WiFi signal positioning method disclosed in an embodiment of the present invention. Figure 1 The long-distance WiFi signal positioning method described is applied to indoor and outdoor positioning, effectively improving the positioning distance and positioning accuracy, which is not limited in the embodiments of the present invention. Figure 1 As shown, the long-distance WiFi signal positioning method may include the following operations:

[0096] S1, obtain channel state information and round-trip time distance information;

[0097] S2, performing beam synthesis on the channel state information to obtain positioning feature parameter information;

[0098] S3, using the round-trip time distance information and the positioning feature parameter information, training a preset positioning model to obtain an optimized positioning model;

[0099] S4, using the optimized positioning model, processing the channel state information and round-trip time distance information to be processed to obtain a positioning result.

[0100] The preset positioning model is a deep belief network optimized by an extreme learning machine. The traditional deep belief network uses the BP algorithm at the top layer to optimize the parameters of the entire network. The BP algorithm has good training efficiency and model accuracy for shallow networks, but as the number of hidden layers increases, the training efficiency of the algorithm will decrease significantly. In order to improve the training speed and the generalization ability of the network model, this embodiment uses an extreme learning machine instead of a BP network for supervised learning at the top layer of DBN. The positioning model training phase is divided into two processes, including pre-training of a deep belief network and a fast training process of an extreme learning machine. A deep belief network with four hidden layers is used for data pre-training, and each hidden layer consists of a different number of neurons. K1, K2, K3, and K4 represent hidden neurons from the first layer to the fourth layer, respectively, and K1>K2>K3>K4. The weights between the layers of the deep belief network are represented by W1, W2, W3, and W4, respectively. In the pre-training process of the deep belief network, a layer-by-layer greedy training method is used for fast training, that is, starting from the bottom network, only two layers in the entire network are trained each time, and the pre-training is simplified to the training of multiple RBMs. The layer-by-layer greedy algorithm first treats the input data x and the hidden layer h1 as RBM1 and trains the RBM1 parameters {b 0 ,b 1 ,W1} and fix them, then treat h1 as the visible layer of RBM2 and h2 as the hidden layer of RBM2, and learn the parameters of RBM2 {b 1 ,b 2 ,W2}, and so on, and finally the parameters of all RBMs are trained.

[0101] Using the CSI tool, a channel state information extraction tool for WiFi devices, in a computer with an Intel 5300 network card installed, use relevant commands to modify the Linux system kernel, and record the channel state information in a file by running the "ping" command to obtain the channel state information.

[0102] Channel state information is the known channel characteristics of the communication link, which contains fine-grained link information for each subcarrier. The signal will produce amplitude attenuation, phase shift and time delay on each path, so the signal propagation conditions of all paths can be used to characterize the channel characteristics, that is, the channel impulse response. The channel impulse response is Fourier transformed to obtain the channel frequency response. The channel frequency response is sampled to obtain the channel state information. This method is a prior art.

[0103] The frequency domain model of the OFDM system is:

[0104] Y=HX+N

[0105] Where Y is the received signal, X is the transmitted signal, H is the channel state information matrix, and N is the additional noise. If the noise N is negligible, the channel state information can be estimated by the following formula:

[0106]

[0107] Among them, H1 is the estimated matrix of the channel state information matrix H. The channel state information describes the channel gain between the transmitting and receiving antennas and provides more fine-grained physical layer information. The channel state information of 30 subcarriers can be obtained by modifying the network card driver. Therefore, H can be expressed as:

[0108] H=[H1, x2 , …, H 30 ] T

[0109] Among them, H i represents the channel state information on the i-th subcarrier, i = 1, 2, ..., 30, which can be expressed as:

[0110]

[0111] Among them, |H i | and ∠H i are the amplitude and phase of the channel state information respectively. In the IEEE 802.11n standard MIMO system, multiple receiving antennas and multiple transmitting antennas are supported, so H i It can also be expressed as:

[0112]

[0113] Where t and r are the number of transmitting and receiving antennas respectively, H tr is the channel state information between the corresponding transmitting and receiving antennas. It can be seen that the channel state information matrix H is a t×r×30 matrix.

[0114] Optionally, the performing beamforming on the channel state information to obtain positioning feature parameter information includes:

[0115] S21, performing Hampel filtering on the channel state information to obtain filtering state information;

[0116] The Hampel filter method is used to remove abnormal data: the amplitude of a data packet is in the interval [MED- γ .MAD, MED+ γ Subcarriers other than .MAD] are considered abnormal subcarriers. If the number of such subcarriers exceeds the threshold, the data packet will be considered abnormal data and will be removed. MED is the median of each group of 30 subcarriers. γ is the strength of the filter, and MAD is the median absolute deviation, which is calculated as

[0117] MAD=median(|x1-MED|,…,|x N -MED|)

[0118] In the formula, x i is the amplitude of the i-th subcarrier, N is the total number of subcarriers, i = 1, 2, ..., N. For the CSI data packet extracted by the CSI tool, N = 30. Set γ to 1.2 and the threshold to 15, that is, when more than 15 of the 30 subcarriers are outside the interval [MED-γ·MAD, MED+γ·MAD], they are considered as outliers, and median means taking the median.

[0119] S22, performing phase extension on the filtering state information to obtain extended state information;

[0120] The CSI (channel state information) phase collected by the CSI tool is folded into the interval [-π, π], which differs from the actual phase by an integer multiple of 2π. In order to make the positioning result more accurate, the collected CSI data needs to be phase extended to restore its actual phase. The specific method is: calculate the phase difference between the i-th (2≤i≤N) subcarrier and the i-1-th subcarrier in the same data packet. If it is greater than π, it means that a positive phase transition has occurred at the i-th subcarrier, and the i-th and subsequent subcarriers should be compensated for the phase of -2π; if the subcarrier phase difference is less than -π, it means that a negative phase transition has occurred at the i-th subcarrier, and the i-th and subsequent subcarriers should be compensated for the phase of 2π. It can be expressed as:

[0121]

[0122] In the formula represents the measured value of the phase of the i-th subcarrier, φ i represents the true phase of the ith subcarrier, h pis the number of positive transitions of the ith and subsequent subcarriers, h n is the number of negative transitions that occur in the i-th and subsequent subcarriers.

[0123] S23, performing a phase linear transformation on the extended state information to obtain linear phase state information;

[0124] The received phase data often contains unknown phase error β and time lag Δt. The measured phase of the i-th subcarrier is expressed as:

[0125]

[0126] Among them, φ i is the true phase of the ith subcarrier, Z i is the measurement noise, f i is the transmission frequency of the ith subcarrier:

[0127]

[0128] Among them, k i is the number of the ith subcarrier, f0 and A are constant coefficients, and Z is ignored. i ,have:

[0129]

[0130] Among them, δ and β′ are unknown phase interferences, and the synchronization error composed of δ and β′ is linearly correlated with the subcarrier number.

[0131] For the 30 subcarriers in a group of data packets, the values ​​of δ and β′ are the same, so the phase error can be eliminated by linear transformation. For each group of received data packets,

[0132]

[0133] Where a is the slope and b is the offset. Subtracting the linear error can get the true phase of CSI, that is, According to the above analysis, the true phase of CSI after linear transformation is:

[0134]

[0135] S24, performing beam synthesis on the linear phase state information to obtain channel state information;

[0136] Set N access points (APs), repeat steps S21 to S24 to obtain N linear phase state information, perform beamforming on the N linear phase state information to obtain channel state information;

[0137] Beamforming is a prior art in this field, and its purpose is to use the signals in the weighted synthesis array antenna to suppress interference and noise and enhance the signal. Common methods include beam space filtering, minimum variance distortion-free response, linear constrained least squares, etc., which are not limited in this embodiment.

[0138] S25, extracting features from the channel state information to obtain positioning feature parameter information.

[0139] Optionally, the extracting features from the channel state information to obtain positioning feature parameter information includes:

[0140] S251, filtering the channel state information to obtain smoothed channel state information;

[0141] Optionally, the filtering method is a low-pass filter, the purpose of which is to filter out high-frequency noise. The parameters of the filter are set by experiment.

[0142] S252, enhancing the smoothed channel state information to obtain enhanced channel state information;

[0143] S253: extract features from the enhanced channel state information to obtain positioning feature parameter information.

[0144] Optionally, the enhancing the smoothed channel state information to obtain enhanced channel state information includes:

[0145] S2521, processing the smoothed channel state information to obtain smoothed channel state amplitude information and smoothed channel state phase information;

[0146] The method is to perform Fourier transform on the smoothed channel state information to obtain smoothed channel state amplitude information and smoothed channel state phase information;

[0147] S2522, processing the smoothed channel state amplitude information and the smoothed channel state phase information to obtain reconstructed amplitude information and reconstructed phase information;

[0148] Calculate the median and median absolute deviation of the smoothed channel state amplitude information in each packet;

[0149] Calculate the number of outliers in each data packet, the variation, and the discrimination threshold. The discrimination threshold is set by experiment. If variation < threshold, the data packet is retained, otherwise the data packet is discarded. If the phase of subcarrier k in each data packet is k The phase of the previous subcarrier in the packet k-1If the difference is greater than π, then the phase i →phase i -2π(i≥k); if the phase of subcarrier k in each data packet is k The phase of the previous subcarrier in the packet k-1 If the difference is less than -π, then the phase i →phase i +2π(i≥k);

[0150] Calculate the amplitude information and phase information of the retained data packet as reconstructed amplitude information and reconstructed phase information;

[0151] S2523: Process the reorganized amplitude information and the reorganized phase information to obtain enhanced channel state information.

[0152] Perform inverse Fourier transform on the reorganized amplitude information and the reorganized phase information to obtain enhanced channel state information.

[0153] Optionally, the extracting features of the enhanced channel state information to obtain positioning feature parameter information includes:

[0154] S2531, preprocessing the enhanced channel state information to obtain preprocessed enhanced channel state information;

[0155] Performing wavelet transform on the enhanced channel state information, and obtaining wavelet parameter information as preprocessed enhanced channel state information;

[0156] S2532, processing the preprocessed enhanced channel state information to obtain fuzzy parameter information;

[0157] The preprocessing enhanced channel state information is calculated as follows:

[0158]

[0159]

[0160] Among them, x(t) is the pre-processed enhanced channel state information, A x (θ, τ) is the fuzzy parameter information, t is the time variable, τ is the displacement variable, θ is the frequency variable corresponding to τ, and * indicates taking the conjugate.

[0161] S2533, performing Hadamard product calculation on the fuzzy parameter information and a preset kernel function to obtain optimized fuzzy parameter information;

[0162] The Hadamard product calculation formula is:

[0163]

[0164] Among them, C x To optimize the fuzzy parameter information, the integral range is -∞~∞, g(θ, τ) is the preset kernel function, a=0.001,β=50,u,t are time variables, τ is the time shift variable, Ω is the frequency variable corresponding to u, θ is the frequency variable corresponding to τ;

[0165] S2534, processing the optimized fuzzy parameter information to obtain positioning feature parameter information.

[0166] The optimized fuzzy parameter information is clustered to obtain positioning feature parameter information. Clustering is a prior art in the field and is not limited in this embodiment.

[0167] Optionally, the using the round-trip time distance information and the positioning feature parameter information to train a preset positioning model to obtain an optimized positioning model includes:

[0168] S31, extracting features from the round-trip time distance information to obtain distance parameter information;

[0169] The calculation formula for the round-trip time distance information is:

[0170] d RTT =(t RTT ·c) / 2=((t4-t1)-(t3-t2))·c / 2

[0171] Among them, d RTT is the round-trip time distance information, t RTT =(t4-t1)-(t3-t2), the starting time of the signal from the access point to the target terminal is t1 and t2 respectively, and the starting time of the signal from the target terminal back to the access point is t3 and t4 respectively, and c is the speed of light.

[0172] Calculate the average value, standard deviation, maximum value, and minimum value of the round-trip time distance information obtained from each set of measurements to form distance parameter information;

[0173] S32, fusing the distance parameter information and the positioning feature parameter information to obtain positioning parameter information;

[0174] S33, using the positioning parameter information, training a preset positioning model to obtain an optimized positioning model.

[0175] Optionally, the fusing the distance parameter information and the positioning feature parameter information to obtain the positioning parameter information includes:

[0176] S321, processing the distance parameter information to obtain a distance covariance matrix;

[0177] The distance parameter information is X, and the distance covariance matrix is ​​S XX ;

[0178] S XX =cov(X,X)

[0179] cov means covariance calculation;

[0180] S322, processing the positioning feature parameter information to obtain a positioning covariance matrix;

[0181] The positioning feature parameter information is Y, and the positioning covariance matrix is ​​S YY ;

[0182] S YY =cov(Y,Y)

[0183] S323, processing the distance parameter information and the positioning feature parameter information to obtain a first cross-covariance matrix and a second cross-covariance matrix;

[0184] The first cross-covariance matrix is ​​S XY ;

[0185] S XY =cov(X,Y)

[0186] The second cross-covariance matrix is ​​S YX ;

[0187] S YX =cov(Y,X)

[0188] S XY =S YX T ;

[0189] S324, processing the distance covariance matrix, the positioning covariance matrix, the first cross-covariance matrix, and the second cross-covariance matrix to obtain a feature fusion matrix;

[0190]

[0191] M is the feature fusion matrix;

[0192] S325, processing the feature fusion matrix to obtain a first singular vector and a second singular vector;

[0193] Solve the singular values ​​of M, and obtain the largest singular value and its preceding and following singular vectors as the first singular vector u and the second singular vector v;

[0194] S326, processing the first singular vector and the second singular vector to obtain a distance projection vector and a positioning projection vector;

[0195]

[0196] Among them, a is the distance projection vector, b is the positioning projection vector;

[0197] S327: Process the distance projection vector and the positioning projection vector to obtain positioning parameter information.

[0198]

[0199] Among them, Z is the positioning parameter information.

[0200] It can be seen that the present invention discloses a long-distance WiFi signal positioning method, which effectively improves the detection distance of the WiFi signal by performing beam synthesis on the WiFi signal; enhances the signal-to-noise ratio of the received signal by comprehensively utilizing the channel state information, the round-trip time distance information and the beam synthesis method, obtains a more accurate WiFi signal, and effectively improves the positioning accuracy.

[0201] Embodiment 2

[0202] This embodiment discloses another long-distance WiFi signal positioning method, which includes a WiFi signal detection module, a WiFi signal enhancement module, a signal transmission module and a signal processing module;

[0203] The WiFi signal detection module is data-connected to the WiFi signal enhancement module, and is used to detect WiFi signals and send the WiFi signals to the WiFi signal enhancement module;

[0204] The WiFi signal enhancement module is data-connected to the signal transmission module, and is used to enhance the received WiFi signal and send the enhanced WiFi signal to the signal processing module;

[0205] The signal processing module is used to process the received WiFi signal to achieve positioning.

[0206] The WiFi signal detection module includes an antenna array, a filtering unit, an amplification unit, an analog-to-digital conversion unit and a beamforming unit;

[0207] The antenna array is data-connected to the filter unit for receiving WiFi signals;

[0208] The filtering unit is data-connected to the amplifying unit and is used to filter the WiFi signal;

[0209] The amplification unit is data-connected to the analog-to-digital conversion unit, and is used to amplify the filtered WiFi signal;

[0210] The analog-to-digital conversion unit is data-connected to the beamforming unit for performing data conversion;

[0211] The beamforming unit is used for performing beamforming.

[0212] Use the WiFi signal detection module to detect and obtain the WiFi signal;

[0213] The WiFi signal is enhanced by using a WiFi signal enhancement module to obtain an enhanced WiFi signal, including:

[0214] Filtering the WiFi signal to obtain a smoothed WiFi signal;

[0215] The smoothed WiFi signal is enhanced to obtain an enhanced WiFi signal, comprising:

[0216] Processing the smoothed WiFi signal to obtain smoothed WiFi signal amplitude information and smoothed WiFi signal phase information;

[0217] Processing the smoothed WiFi signal amplitude information and the smoothed WiFi signal phase information to obtain reconstructed amplitude information and reconstructed phase information;

[0218] Processing the reorganized amplitude information and the reorganized phase information to obtain an enhanced WiFi signal;

[0219] Using a signal transmission module, the enhanced WiFi signal is sent to a signal processing module;

[0220] The enhanced WiFi signal is processed by a signal processing module to obtain a positioning result, including:

[0221] Get round trip time and distance information;

[0222] Using the enhanced WiFi signal and the round-trip time distance information, a preset positioning model is trained to obtain an optimized positioning model, including:

[0223] Performing outlier processing on the enhanced WiFi signal to obtain a first WiFi signal;

[0224] Extracting features of the first WiFi signal to obtain location feature information;

[0225] Using the location feature information and the round-trip time distance information, a preset positioning model is trained to obtain an optimized positioning model;

[0226] The optimized positioning model is used to process the enhanced WiFi signal to be processed to obtain a positioning result.

[0227] The calculation formula of the round-trip time distance information is:

[0228] d RTT =(t RTT ·c) / 2=((t4-t1)-(t3-t2))·c / 2

[0229] Among them, d RTT is the round-trip time distance information, t RTT =(t4-t1)-(t3-t2), the starting time of the signal from the access point to the target terminal is t1 and t2 respectively, and the starting time of the signal from the target terminal back to the access point is t3 and t4 respectively, and c is the speed of light.

[0230] It can be seen that the present invention discloses a long-distance WiFi signal positioning method, which effectively improves the detection distance of the WiFi signal by performing beam synthesis on the WiFi signal; enhances the signal-to-noise ratio of the received signal by comprehensively utilizing the channel state information, the round-trip time distance information and the beam synthesis method, obtains a more accurate WiFi signal, and effectively improves the positioning accuracy.

[0231] Embodiment 3

[0232] Aiming at improving the hidden detection capability of near-field WiFi network signals and focusing on solving the problems of fast attenuation of WiFi signal transmission process and weak signal detection capability of traditional WiFi detection equipment, this embodiment discloses a portable high-performance WiFi detection device based on digital beamforming technology. The beamforming technology is used to greatly improve the radio frequency performance of the device. Combined with the high-integration PCB stacking design, the device achieves the unity of high performance and miniaturization, greatly improving the security, convenience and success rate of WiFi detection and access.

[0233] First, by using digital beamforming technology and combining it with the front-end high-performance array antenna design, the received RF signal is filtered, amplified, frequency-converted, analog-to-digital converted, and digitally phase-shifted and signal-weighted amplified, as well as multi-channel phase calibration and synchronization is performed to greatly enhance the signal-to-noise ratio of the received signal, obtain a more accurate signal incident direction, and effectively suppress signal clutter interference. For details, see Figure 2 .

[0234] Second, by implementing WiFi baseband protocol processing and beam direction control in the FPGA chip, the transmit signal gain is greatly enhanced, and directional transmission and long-distance access of RF signals are achieved. For details, see Figure 3 .

[0235] Third, in the RF front-end design, microstrip antennas and coupled feeding technology are used to increase the array antenna density and improve the antenna gain.

[0236] Fourth, through the zero intermediate frequency technology, the PCB board area is reduced by nearly half, and the power consumption of the equipment is reduced; at the same time, the phased array technology based on electronic scanning is used, combined with the highly integrated PCB stacking solution, to further significantly reduce the size and weight of the equipment. For details, see Figure 4 .

[0237] Through the application of the above technologies, the high performance and portability of WiFi devices are effectively achieved. The device has a detection distance of 3,000 meters in an open environment and an access distance of 1,500 meters. The device weighs less than 5kg and consumes only 30W of power. In the simulation test, the WiFi portable device can detect the proximity of office buildings and residential areas, crack passwords, and access them.

[0238] Embodiment 4

[0239] See also Figure 5 , Figure 5 : is a schematic diagram of a long-distance WiFi signal positioning device disclosed in an embodiment of the present invention. Figure 5 The long-distance WiFi signal positioning device described is applied to indoor and outdoor positioning, effectively improving the positioning distance and positioning accuracy, which is not limited in the embodiments of the present invention. Figure 5 As shown, the long-distance WiFi signal positioning device may include the following operations:

[0240] S301, an information acquisition module, used to acquire channel state information and round-trip time distance information;

[0241] S302, a beamforming module, configured to perform beamforming on the channel state information to obtain positioning feature parameter information;

[0242] S303, a model training module, used to train a preset positioning model using the round-trip time distance information and the positioning feature parameter information to obtain an optimized positioning model;

[0243] S304, a positioning module is used to use the optimized positioning model to process the channel state information and round-trip time distance information to be processed to obtain a positioning result.

[0244] Embodiment 6

[0245] See also Figure 6 , Figure 6 : is a schematic diagram of the structure of another long-distance WiFi signal positioning device disclosed in an embodiment of the present invention. Figure 6The long-distance WiFi signal positioning device described is applied to indoor and outdoor positioning, effectively improving the positioning distance and positioning accuracy, which is not limited in the embodiments of the present invention.

[0246] like Figure 6 As shown, the long-distance WiFi signal positioning device may include the following operations:

[0247] A memory 401 storing executable program codes;

[0248] a processor 402 coupled to the memory 401;

[0249] The processor 402 calls the executable program code stored in the memory 401 to execute the steps in the long-distance WiFi signal positioning method described in the first and second embodiments.

[0250] Embodiment 7

[0251] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps in the long-distance WiFi signal positioning method described in Embodiment 1 and Embodiment 2.

[0252] The device embodiments described above are only illustrative, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, i.e., they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0253] Through the specific description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution can be essentially or partly contributed to the prior art in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0254] Finally, it should be noted that the long-distance WiFi signal positioning method and device disclosed in the embodiments of the present invention disclose only the preferred embodiments of the present invention, which are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A long-distance WiFi signal positioning method, characterized in that: The method comprises: S1, obtain channel state information and round-trip time distance information; S2, performing beam synthesis on the channel state information to obtain positioning feature parameter information, including: S21, performing Hampel filtering on the channel state information to obtain filtering state information; The Hampel filter method is used to remove abnormal data: the subcarriers in a data packet whose amplitude is outside the interval [MED-γ·MAD, MED+γ·MAD] are regarded as abnormal subcarriers. If the number of such subcarriers exceeds the threshold, the data packet will be regarded as abnormal data and removed, where MED is the median of each group of 30 subcarriers, γ is the filtering strength, and MAD is the median absolute deviation. The calculation method is MAD=median(|x1-MED|, ..., |x N -MED|) In the formula, x i is the amplitude of the i-th subcarrier, N is the total number of subcarriers, i=1, 2, ..., N. For the CSI data packet extracted by the CSI tool, N=30, γ is set to 1.2, and the threshold is set to 15. That is, when more than 15 of the 30 subcarriers are outside the interval [MED-γ·MAD, MED+γ·MAD], they are regarded as outliers. Median means taking the median; S22, performing phase extension on the filtering state information to obtain extended state information; S23, performing a phase linear transformation on the extended state information to obtain linear phase state information; S24, performing beam synthesis on the linear phase state information to obtain channel state information; S25, extracting features from the channel state information to obtain positioning feature parameter information, including: S251, filtering the channel state information to obtain smoothed channel state information; S252, enhancing the smoothed channel state information to obtain enhanced channel state information, including: S2521, processing the smoothed channel state information to obtain smoothed channel state amplitude information and smoothed channel state phase information; S2522, processing the smoothed channel state amplitude information and the smoothed channel state phase information to obtain reconstructed amplitude information and reconstructed phase information; Calculate the median and median absolute deviation of the smoothed channel state amplitude information in each packet; Calculate the number of outliers in each data packet, the variation, and the discrimination threshold. The discrimination threshold is set by experiment. If variation < threshold, the data packet is retained, otherwise the data packet is discarded. If the phase of subcarrier k in each data packet is k The phase of the previous subcarrier in the packet k-1 If the difference is greater than π, then the phase i →phase i -2π(i≥k); if the phase of subcarrier k in each data packet is k The phase of the previous subcarrier in the packet k-1 If the difference is less than -π, then the phase i →phase i +2π(i≥k); Calculate the amplitude information and phase information of the retained data packet as reconstructed amplitude information and reconstructed phase information; S2523, processing the reorganized amplitude information and the reorganized phase information to obtain enhanced channel state information; S253, extracting features from the enhanced channel state information to obtain positioning feature parameter information, including: S2531, preprocessing the enhanced channel state information to obtain preprocessed enhanced channel state information; S2532, processing the preprocessed enhanced channel state information to obtain fuzzy parameter information; The preprocessing enhanced channel state information is calculated as follows: Among them, x(t) is the pre-processed enhanced channel state information, A x (θ, τ) is the fuzzy parameter information, t is the time variable, τ is the displacement variable, θ is the frequency variable corresponding to τ, and * indicates taking the conjugate; S2533, performing Hadamard product calculation on the fuzzy parameter information and a preset kernel function to obtain optimized fuzzy parameter information; The Hadamard product calculation formula is: Among them, C x To optimize the fuzzy parameter information, the integral range is -∞~∞, g(θ,τ) is the preset kernel function, a=0.001,β=50,u,t are time variables, τ is the time shift variable, Ω is the frequency variable corresponding to u, θ is the frequency variable corresponding to τ; S2534, processing the optimized fuzzy parameter information to obtain positioning feature parameter information; S3, using the round-trip time distance information and the positioning feature parameter information, training a preset positioning model to obtain an optimized positioning model, including: S31, extracting features from the round-trip time distance information to obtain distance parameter information; S32, fusing the distance parameter information and the positioning feature parameter information to obtain positioning parameter information, including: S321, processing the distance parameter information to obtain a distance covariance matrix; S322, processing the positioning feature parameter information to obtain a positioning covariance matrix; S323, processing the distance parameter information and the positioning feature parameter information to obtain a first cross-covariance matrix and a second cross-covariance matrix; S324, processing the distance covariance matrix, the positioning covariance matrix, the first cross-covariance matrix, and the second cross-covariance matrix to obtain a feature fusion matrix; S325, processing the feature fusion matrix to obtain a first singular vector and a second singular vector; S326, processing the first singular vector and the second singular vector to obtain a distance projection vector and a positioning projection vector; S327, processing the distance projection vector and the positioning projection vector to obtain positioning parameter information; S33, using the positioning parameter information, training a preset positioning model to obtain an optimized positioning model; The preset positioning model is a deep belief network optimized by an extreme learning machine. The extreme learning machine is used instead of the BP network for the top-level supervised learning of the DBN. The positioning model training phase is divided into two processes, including the pre-training of the deep belief network and the fast training process of the extreme learning machine. The deep belief network with 4 hidden layers is used for data pre-training, and each hidden layer consists of a different number of neurons. S4, using the optimized positioning model, processing the channel state information and round-trip time distance information to be processed to obtain a positioning result.

2. A long-distance WiFi signal positioning device, characterized in that: The device comprises: An information acquisition module, used to acquire channel state information and round-trip time distance information; A beamforming module, used to perform beamforming on the channel state information to obtain positioning feature parameter information, including: S21, performing Hampel filtering on the channel state information to obtain filtering state information; The Hampel filter method is used to remove abnormal data: the amplitude of a data packet is in the interval [MED- γ MAD, MED+ γ ·MAD] are considered abnormal subcarriers. If the number of such subcarriers exceeds the threshold, the data packet will be considered abnormal data and removed, where MED is the median of each group of 30 subcarriers, γ is the filtering strength, and MAD is the median absolute deviation, which is calculated as MAD = median(|x1-MED|, ..., |x N -MED|) In the formula, x i is the amplitude of the i-th subcarrier, N is the total number of subcarriers, i=1, 2, ..., N. For the CSI data packet extracted by the CSI tool, N=30, γ is set to 1.2, and the threshold is set to 15. That is, when more than 15 of the 30 subcarriers are outside the interval [MED-γ·MAD, MED+γ·MAD], they are regarded as outliers. Median means taking the median; S22, performing phase extension on the filtering state information to obtain extended state information; S23, performing a phase linear transformation on the extended state information to obtain linear phase state information; S24, performing beam synthesis on the linear phase state information to obtain channel state information; S25, extracting features from the channel state information to obtain positioning feature parameter information, including: S251, filtering the channel state information to obtain smoothed channel state information; S252, enhancing the smoothed channel state information to obtain enhanced channel state information, including: S2521, processing the smoothed channel state information to obtain smoothed channel state amplitude information and smoothed channel state phase information; S2522, processing the smoothed channel state amplitude information and the smoothed channel state phase information to obtain reconstructed amplitude information and reconstructed phase information; Calculate the median and median absolute deviation of the smoothed channel state amplitude information in each packet; Calculate the number of outliers in each data packet, the variation, and the discrimination threshold. The discrimination threshold is set by experiment. If variation < threshold, the data packet is retained, otherwise the data packet is discarded. If the phase of subcarrier k in each data packet is k The phase of the previous subcarrier in the packet k-1 If the difference is greater than π, then the phase i →phase i -2π(i≥k); if the phase of subcarrier k in each data packet is k The phase of the previous subcarrier in the packet k-1 If the difference is less than -π, then the phase i →phase i +2π(i≥k); Calculate the amplitude information and phase information of the retained data packet as reconstructed amplitude information and reconstructed phase information; S2523, processing the reorganized amplitude information and the reorganized phase information to obtain enhanced channel state information; S253, extracting features from the enhanced channel state information to obtain positioning feature parameter information, including: S2531, preprocessing the enhanced channel state information to obtain preprocessed enhanced channel state information; S2532, processing the preprocessed enhanced channel state information to obtain fuzzy parameter information; The preprocessing enhanced channel state information is calculated as follows: Among them, x(t) is the pre-processed enhanced channel state information, A x (θ, τ) is the fuzzy parameter information, t is the time variable, τ is the displacement variable, θ is the frequency variable corresponding to τ, and * indicates taking the conjugate; S2533, performing Hadamard product calculation on the fuzzy parameter information and a preset kernel function to obtain optimized fuzzy parameter information; The Hadamard product calculation formula is: Among them, C x To optimize the fuzzy parameter information, the integral range is -∞~∞, g(θ,τ) is the preset kernel function, a=0.001,β=50,u,t is the time variable, τ is the time shift variable, Ω is the frequency variable corresponding to u, θ is the frequency variable corresponding to τ; S2534, processing the optimized fuzzy parameter information to obtain positioning feature parameter information; A model training module is used to train a preset positioning model using the round-trip time distance information and the positioning feature parameter information to obtain an optimized positioning model, including: S31, extracting features from the round-trip time distance information to obtain distance parameter information; S32, fusing the distance parameter information and the positioning feature parameter information to obtain positioning parameter information, including: S321, processing the distance parameter information to obtain a distance covariance matrix; S322, processing the positioning feature parameter information to obtain a positioning covariance matrix; S323, processing the distance parameter information and the positioning feature parameter information to obtain a first cross-covariance matrix and a second cross-covariance matrix; S324, processing the distance covariance matrix, the positioning covariance matrix, the first cross-covariance matrix, and the second cross-covariance matrix to obtain a feature fusion matrix; S325, processing the feature fusion matrix to obtain a first singular vector and a second singular vector; S326, processing the first singular vector and the second singular vector to obtain a distance projection vector and a positioning projection vector; S327, processing the distance projection vector and the positioning projection vector to obtain positioning parameter information; S33, using the positioning parameter information, training a preset positioning model to obtain an optimized positioning model; The preset positioning model is a deep belief network optimized by an extreme learning machine. The extreme learning machine is used instead of the BP network for the top-level supervised learning of the DBN. The positioning model training phase is divided into two processes, including the pre-training of the deep belief network and the fast training process of the extreme learning machine. The deep belief network with 4 hidden layers is used for data pre-training, and each hidden layer consists of a different number of neurons. The positioning module is used to process the channel state information and round-trip time distance information to be processed by using the optimized positioning model to obtain a positioning result.

3. A long-distance WiFi signal positioning device, characterized in that: The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the long-distance WiFi signal positioning method as described in claim 1.

4. A computer storable medium, characterized in that: The computer storable medium stores computer instructions, and when the computer instructions are called, they are used to execute the long-distance WiFi signal positioning method as claimed in claim 1.

Images