Anti-interference indoor positioning method and system

Abstract

The invention discloses an anti-interference indoor positioning method and system, and relates to the technical field of indoor positioning, and the method comprises the steps: building a space model, selecting a signal domain adjacent fingerprint reference point in a test point sub-region through a multi-resolution method, and obtaining a signal domain adjacent reference point set; k neighbor reference points are selected from the signal domain neighbor reference point set, and a physical domain neighbor reference point set is obtained through screening again from the physical domain by adopting a line clustering degree method; and based on a neighbor reference point set under double screening of a signal domain and a physical domain, obtaining position estimation of a test point in a conjugate space by adopting a WKNN algorithm. According to the method, multi-level constraint of signal similarity and distribution structure robustness of neighbor points is realized by introducing the conjugate space, complementary information is allowed to be obtained from two dimensions of energy and physics, neighbor point positioning information effective for both the signal and the physical distance is provided, physical and signal distance performance characteristics are taken into consideration, and the positioning accuracy of the neighbor points is improved. And the balance between the signal similarity and the distribution structure robustness is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of indoor positioning, and particularly to an anti-interference indoor positioning method and system. Background Art

[0002] With the intelligent development of current society, the demand for high-precision positioning has become increasingly prominent. Although the Global Navigation Satellite System (GNSS) provides real-time and accurate navigation services outdoors, due to the shielding effect of buildings, the ability to receive satellite signals indoors is limited. Therefore, there is an urgent need for an efficient and accurate indoor positioning solution.

[0003] In a diverse indoor environment, there are multiple access points (APs), and indoor positioning can be achieved using information such as time of arrival (TOA), time difference of arrival (TDOA), angle of arrival (AOA), channel state information (CSI), and received signal strength (RSS). In particular, due to its universality and the characteristic of not relying on dedicated hardware modules, Bluetooth or Wi-Fi signals based on RSS have become the most commonly used signal sources in indoor positioning applications. And the fingerprint matching method using RSS as data basis has also become one of the commonly used methods by researchers.

[0004] The fingerprint matching process is divided into two stages: offline and online. The general fingerprint positioning method is to perform similarity matching between the reference points (RP) in the fingerprint database and the test points (TP) to be located, but this brings a large amount of computation. In order to reduce the complexity and time consumption during online positioning, clustering and simplification are performed in the offline stage to reduce the online calculation amount and improve the positioning accuracy. Commonly used clustering methods such as K-means, DBSCAN, and DPC divide the target area into sub-spaces and describe each space in the form of classes. Signals, as fingerprint information, are different from pure digital quantities because they are mixed with noise interference, making it such that even if signals are close to each other in the energy domain, they are quite different in the physical domain.

[0005] In addition to offline clustering, online matching methods will also affect the positioning results. WKNN uses the matching degree between TP and RP in the fingerprint library to select K RPs with the highest matching degree for weighted position calculation. However, it lacks consideration of scenarios such as multipath effects, AP aging, and mobile receiving devices; the selection of fixed neighboring RP points will inevitably introduce redundant information or lose important information. Therefore, it is particularly important to establish a fingerprint positioning system with strong reliability and high precision.

[0006] To address the problem of spatial ambiguity, the WKNN method based on signal similarity and spatial position matches neighboring points with the nearest neighbors in the signal domain, but this type of solution ignores the consideration of AP volatility; to solve the test point misjudgment caused by unstable APs, stability and discrimination factors are introduced to assist AP selection, but because only the ubiquitous signal space is considered, the joint consideration of the physical space is ignored, resulting in large deviations in positioning results; considering the physical trajectory characteristics of signal propagation, the region is divided into circular boundaries to constrain spatial ambiguity, and the relative relationship of RSS between APs is used to reduce the cost of fingerprint library construction, which is more adaptable to the transmission power differences caused by AP heterogeneity, but its positioning accuracy is still easily affected by RSS fluctuations and environmental interference, which limits the generalization ability of the positioning model. Summary of the invention

[0007] In view of the above-mentioned problems, the present invention is proposed.

[0008] Therefore, the problem to be solved by the present invention is that the positioning accuracy of the prior art is still easily affected by RSS fluctuations and environmental interference, which limits the generalization ability of the positioning model.

[0009] In order to solve the above technical problems, the present invention provides the following technical solutions: an anti-interference indoor positioning method, comprising: determining ground wireless access points and aerial wireless access points in a positioning area, deploying ground fingerprint reference points and aerial fingerprint reference points, and determining a unique physical address MAC for each wireless access point; extracting wireless access points AP coexisting in ground wireless access points and aerial wireless access points through MAC, obtaining S MAC addresses by intersection to determine a ground plane AP library, and obtaining a ground plane original fingerprint library according to the joint coordinates of the AP library; collecting the signal strength of a test point TP, roughly locating a sub-area of ​​the TP according to the Euclidean distance of the signal strength between the reference point and the TP, and obtaining a sub-area neighbor reference point set; selecting a signal domain neighbor fingerprint reference point in the sub-area by a multi-resolution method, and obtaining a signal domain neighbor reference point set; selecting K neighbor reference points from the signal domain neighbor reference point set, and using a line aggregation method to screen again from the physical domain to obtain a physical domain neighbor reference point set; based on the neighbor reference point set under dual screening of the signal domain and the physical domain, using a WKNN algorithm to obtain a position estimate of the test point TP in a conjugate space.

[0010] As a preferred solution of the anti-interference indoor positioning method described in the present invention, wherein: the determination method of the ground wireless access points includes detecting N ground wireless access points AP on the ground plane during the offline stage within the positioning area Ω n ; the deployment method of the ground fingerprint reference points includes deploying M fingerprint reference points RP at equal intervals on the ground plane m ; the deployment method of the air fingerprint reference points includes deploying reference points in the same manner as on the ground plane on the air plane with the same two-dimensional coordinates on the ground plane but a height of Z The determination method of the air wireless access points includes detecting J wireless access points AP on the air plane j .

[0011] As a preferred solution of the anti-interference indoor positioning method described in the present invention, wherein: the selection of the nearest neighbor fingerprint reference points in the signal domain within the sub-region by using the multi-resolution method includes calculating the correlation coefficient ξ between all the nearest neighbor reference points RP u in the sub-region Ω k and the test point TP based on the Pearson coefficient, which is expressed as k

[0012]

[0013] wherein, E TP represents the RSS measurement value of the ground wireless access points included in the test point TP, RSS represents the received signal strength, G k represents the RSS measurement value of the ground wireless access points included in the nearest neighbor reference point, Cov(G k , E TP ) represents the signal covariance between the nearest neighbor reference point RP k and the test point TP, represents the signal variance of AP k collected at the ground plane RP n , represents the signal variance of AP n collected at the test point TP; based on the correlation coefficient ξ k , set the range of layer-by-layer division of the multi-resolution model, and the formula is expressed as

[0014]

[0015] wherein, ε k represents the number of layers; and respectively represent the lower limit and the upper limit of each layer, g represents the rising gradient and g>0; define the resolution of the ε k th layer as

[0016]

[0017] The resolution is thus obtained With the number of layers ε k The mapping relationship is expressed as:

[0018]

[0019] Here, ε is represented as a very small non-negative number to prevent the denominator from being 0. Expressed as a floor function, according to the resolution Computation level ε k ; Calculate the Euclidean distance between the signal of the nearest reference point of this layer and the test point TP Select The smallest σ t reference points RP as trusted neighbors, and the final candidate reference point RP set in the signal space is obtained.

[0020] As a preferred solution of the anti-interference indoor positioning method described in the present invention, wherein: the selecting of K neighbor reference points from the signal domain neighbor reference point set includes: for the signal domain reference point set According to the signal Euclidean scale, the reference points RP are sorted in ascending order. Take the reference point RP with the smallest Euclidean distance as the initial reference point RP initial ; From RP initial The grouping is started based on the Euclidean distance, where each group selects three reference points with the closest distance as a set of reference points {RP k ,RP k+1 ,RP k+2}, each reference point RP participates in the selection only once, thus obtaining multiple sets of reference points.

[0021] As a preferred solution of the anti-interference indoor positioning method described in the present invention, wherein: the use of the linear aggregation method to screen again from the physical domain to obtain a physical domain neighbor reference point set also includes, for the reference point set {RP k ,RP k+1 ,RP k+2}, calculate RP respectively k With RP k+1 RP k With RP k+2 RP k+1 With RP k+2 Distance The stability of each angle is determined by the triangular cosine theorem. The formula is:

[0022]

[0023] Among them, χ k , χ k+1 , χ k+2 represents the cosine value of the included angle of the triangle formed by {RP k , RP k+1 , RP k+2}. Let the line concentration of the current group reference point set {RP k , RP k+1 , RP k+2} be

[0024]

[0025] The calculated line concentration is the line concentration of the current group reference point set, that is, each group reference point set shares a unified line concentration, denoted as

[0026] Traverse all the reference points RP in. If the number of RPs in the remaining reference point combinations cannot form a triangle, that is, when the number of remaining RPs is 2 or the number of remaining RPs is 1, set the line concentration to 0, that is or

[0027] Calculate the mean value L of all reference point sets th It is expressed as

[0028]

[0029] If then discard the current reference point set {RP k}, and obtain the final physical domain reference point set

[0030] As a preferred solution of the anti-interference indoor positioning method described in the present invention, wherein: the position estimation of the test point TP in the conjugate space obtained by using the WKNN algorithm includes calculating the conjugate space value based on the near neighbor point set under signal and physical domain screening, expressed as The conjugate space value is calculated as

[0031] C sk = ξ k + iχ k

[0032] Among them, C sk represents the proximity degree of the reference point RP k to the test point TP in the signal domain and the rational degree of the physical structure. i represents the imaginary unit; for the near neighbor reference point RP k set, conjugate weights are respectively configured, expressed as

[0033]

[0034] Here, ε is a very small non-negative number to prevent the denominator from being zero.

[0035] As a preferred solution of the anti-interference indoor positioning method described in the present invention, wherein: the use of the WKNN algorithm to obtain the position estimation of the test point TP in the conjugate space also includes introducing a negative exponential function Using smoothing characteristics to limit the neighboring reference points RP k The conjugate weight fluctuation range of the signal domain and the physical domain double-domain screening of the neighbor set As the coordinates of the test point TP (x t ,y t ) is an ideal reference set for position estimation, by indexing the neighboring reference points RP k The positioning weight of the conjugate space is obtained based on WKNN and is expressed as:

[0036]

[0037] in, Expressed as the weight of the reference point RPk, (x k ,y k ) is represented as the physical coordinates of the reference point RPk.

[0038] Another object of the present invention is to provide an anti-interference indoor positioning system, which can accurately locate the indoor position to be measured.

[0039] To solve the above technical problems, the present invention provides the following technical solutions: a system for an anti-interference indoor positioning method, comprising: a space model building module, a signal domain neighbor point screening module, a physical domain neighbor point screening module and a conjugate space positioning module; the space model building module is used to detect wireless access points in a positioning area, deploy reference points, establish a ground plane original fingerprint library, and determine a sub-area of ​​a test point TP; the signal domain neighbor point screening module uses a distance measurement method to screen signal domain neighbor fingerprint reference points in the area to which the test point TP belongs, and obtains a signal domain neighbor reference point set; the physical domain neighbor point screening module uses the rationality of the physical geometric structure of the reference point as a basis for judging whether the signal domain neighbor reference point is credible, thereby obtaining a highly credible neighbor reference point set in the physical domain; the conjugate space positioning module selects K neighbor reference points from the neighbor reference point set, and uses the WKNN method to estimate the position of the point to be positioned TP.

[0040] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the anti-interference indoor positioning method described above are implemented.

[0041] A computer-readable storage medium stores a computer program thereon. When the computer program is executed by a processor, the steps of the anti-interference indoor positioning method as described above are implemented.

[0042] The beneficial effects of the present invention are as follows: The present invention adopts a multi-resolution method to enable the reference point RP k in the process of searching in the signal space, which is dominated by multiple sets of recursive resolutions to regulate the problem that the timeliness and accuracy cannot be balanced during the search of the reference point RP k and achieve the accurate analysis of the correlation degree between the reference point RP k and the test point TP in the signal domain, thereby simplifying complex retrieval into direct filtering.

[0043] The present invention adopts a dual-scale measurement method to provide a more comprehensive screening criterion for the reference point RP k and improve the positioning accuracy. Through the adaptive characteristic of the line concentration threshold L th , the problem of lack of flexibility of the fixed threshold is solved by dynamically adapting to environmental changes, so that the number of neighboring points can be adaptively selected.

[0044] The present invention realizes multi-level constraints on the signal similarity and distribution structure robustness of neighboring points through the introduction of a conjugate space. This strategy allows complementary information to be obtained from both the energy and physical dimensions, provides effective neighboring point positioning information for both signal and physical distances, takes into account the performance characteristics of physical and signal distances, and achieves a balance between signal similarity and distribution structure robustness. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0046] Figure 1 FIG. is a flowchart of an anti-interference indoor positioning method in Embodiment 1.

[0047] Figure 2 FIG. is a cumulative probability distribution diagram of the positioning errors of five algorithms of an anti-interference indoor positioning method in Embodiment 2.

[0048] Figure 3 FIG. is a box plot of the positioning errors of five algorithms of an anti-interference indoor positioning method in Embodiment 2.

[0049] Figure 4 FIG. is a module structure diagram of an anti-interference indoor positioning system in Embodiment 3. DETAILED DESCRIPTION OF THE INVENTION

[0050] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0051] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0052] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and which provides an anti-interference indoor positioning method including: Figure 1 As shown:

[0053] Step 1: In the offline phase, determine the ground wireless access points and the aerial wireless access points in the positioning area, deploy the ground fingerprint reference points and the aerial fingerprint reference points, and determine a unique physical address MAC for each wireless access point. Wireless access can be in the form of Bluetooth or Wi-Fi.

[0054] The method for determining the ground wireless access point includes: detecting N ground wireless access points AP at the ground level in the offline stage within the positioning area Ω n .

[0055] The deployment method of ground fingerprint reference points includes deploying M fingerprint reference points RP at equal intervals on the ground plane. m .

[0056] The deployment method of the aerial fingerprint reference point includes deploying the reference point in the same way as the ground plane on an empty plane with the same two-dimensional coordinates as the ground plane but with a height of Z.

[0057] The method for determining the wireless access point in the air includes detecting J wireless access points AP on the air plane. j .

[0058] Step 2: Extract the wireless access points APs that coexist between ground wireless access points and aerial wireless access points through MAC, obtain S MAC addresses through intersection to determine the ground plane AP library, and obtain the ground plane original fingerprint library based on the joint coordinates of the AP library.

[0059] Step 3: Collect the signal strength of the test point TP, roughly locate the sub-area of ​​TP based on the Euclidean distance of the signal strength between the reference point and TP, and obtain the set of neighboring reference points in the sub-area.

[0060] Step 4: Use the multi - resolution method to select the neighboring fingerprint reference points in the sub - region and obtain the set of neighboring reference points in the signal domain.

[0061] Based on the Pearson coefficient, calculate the correlation coefficient ξ u between all neighboring reference points RP k in the sub - region Ω k and the test point TP, which is expressed as

[0062]

[0063] where E TP represents the RSS measurement value of the ground - plane wireless access point included in the test point TP, RSS represents the received signal strength, and G k represents the RSS measurement value of the ground - plane wireless access point included in the neighboring reference point. Cov(G k , E TP ) represents the covariance of the signal strengths between the neighboring reference point RP k and the test point TP. represents the variance of the AP k signal collected at the ground - plane RP n , and represents the variance of the AP n signal collected at the test point TP.

[0064] Based on the correlation coefficient ξ k , set the range of layer - by - layer division of the multi - resolution model, which is expressed by the formula

[0065]

[0066] where ε k represents the number of layers; and represent the lower and upper limits of each layer respectively, and g represents the rising gradient with g > 0.

[0067] Define the resolution of the ε k th layer as

[0068]

[0069] Thus, the mapping relationship between the resolution and the number of layers ε k is expressed as

[0070]

[0071] where ε represents a non - negative minimum number to prevent the denominator from being 0, represents the floor function, and calculate the layer number ε according to the resolution . k。

[0072] Calculate the Euclidean distance of the signal between the nearest neighbor reference points of this layer and the test point TP Select The smallest σ t reference points RP as the confidence neighbors, and obtain the final candidate reference point RP set in the signal space

[0073] Thus, the present invention enables the reference point RP k During the process of searching in the signal space, it is dominated by the resolution of multiple sets of recurrence formulas to regulate the problem that the timeliness and accuracy cannot be balanced during the search of the reference point RP k and achieve the accurate analysis of the correlation degree between the reference point RP k and the test point TP in the signal domain, thereby simplifying the complex retrieval to direct filtering; the correlation coefficient ξ k preliminarily screens the high-correlation nearest neighbor reference points RP k , and the Euclidean criterion deeply quantifies the differences between signal features. The dual-scale measurement method provides a more comprehensive reference point RP k screening criterion and improves the positioning accuracy.

[0074] Step 5: Select K nearest neighbor reference points from the set of nearest neighbor reference points in the signal domain, and use the line concentration method to further screen from the physical domain to obtain the set of nearest neighbor reference points in the physical domain.

[0075] For the set of reference points in the signal domain Sort the reference points RP in ascending order according to the signal Euclidean scale to obtain Select the reference point RP with the smallest Euclidean distance as the initial reference point RP initial 。

[0076] Start from RP initial and divide into groups according to the Euclidean distance. Among them, three reference points with the closest distances are selected in each group as a set of reference points {RP k , RP k+1 , RP k+2}, and each reference point RP only participates in the selection once, thereby obtaining multiple sets of reference points.

[0077] For the set of reference points {RP k , RP k+1 , RP k+2}, calculate the distances between RP k and RP k+1 , between RP k and RP k+2 , and between RP k+1 and RP k+2 respectively Judge the stability degree of each angle through the cosine theorem of a triangle, and the formula is expressed as

[0078]

[0079] where χ k 、χ k+1 、χ k+2 represent the cosine values of the included angles of the triangle formed by {RP k , RP k+1 , RP k+2}. Let the line concentration of the current group reference point set {RP k , RP k+1 , RP k+2} be

[0080]

[0081] The calculated line concentration is the line concentration of the current group reference point set, that is, each group reference point set shares a unified line concentration, denoted as

[0082] Traverse All reference points RP in. If the number of RPs in the remaining reference point combination cannot form a triangle, that is, when the remaining number of RPs is 2 or the remaining number of RPs is 1, set the line concentration to 0, that is Or

[0083] Calculate the mean value L of all reference point sets th It is expressed as

[0084]

[0085] If Then discard the current reference point set {RP k}, and obtain the final physical domain reference point set

[0086] In this way, through the adaptive characteristic of the line concentration threshold L th , the problem of lack of flexibility of the fixed threshold is solved by dynamically adapting to environmental changes, so that the number of neighboring points can be adaptively selected. And the physical domain reference point set under the line concentration Considers the geometric structure of the signal domain reference point set . By verifying the accuracy of the neighboring information in the energy domain through the robustness of the physical domain structure, the problem of over-relying on the signal space is avoided, and the positioning accuracy is improved by the neighboring points with high confidence under the double constraints.

[0087] Step 6: Based on the neighboring reference point set under the double screening of the signal domain and the physical domain, use the WKNN algorithm to obtain the position estimate of the test point TP in the conjugate space.

[0088] Neighbor point set based on signal and physical domain screening The calculation of the conjugate space value is expressed as,

[0089] C sk =ξ k +iχ k

[0090] Among them, C sk Denoted as reference point RP k The proximity degree to the signal domain of the test point TP and the rationality of the physical structure, i is expressed as an imaginary unit.

[0091] For the neighboring reference point RP k Set, respectively configure the conjugate weights expressed as,

[0092]

[0093] Here, ε is a very small non-negative number to prevent the denominator from being zero.

[0094] Introducing negative exponential function Using smoothing characteristics to limit the neighboring reference points RP k The conjugate weight fluctuation range.

[0095] Neighbor set based on dual-domain screening of signal and physical domains As the coordinates of the test point TP (x t ,y t ) is an ideal reference set for position estimation, by indexing the neighboring reference points RP k The positioning weight of the conjugate space is obtained based on WKNN and is expressed as:

[0096]

[0097] in, Expressed as the weight of the reference point RPk, (x k ,y k ) is represented as the physical coordinates of the reference point RPk.

[0098] In this way, multi-level constraints on neighboring points in signal similarity and distribution structure robustness are achieved through the introduction of conjugate space. This strategy allows complementary information to be obtained from both energy and physical dimensions, provides neighboring point positioning information that is valid for both signal and physical distances, takes into account both physical and signal distance performance characteristics, and achieves a balance between signal similarity and distribution structure robustness.

[0099] Example 2, reference Figure 2 and Figure 3, which is the second embodiment of the present invention. The difference from the first embodiment is that an anti-interference indoor positioning method further includes: to verify and illustrate the technical effects adopted in this method, in this embodiment, a traditional technical solution is compared with the method of the present invention through a comparative test, and the experimental results are compared by means of scientific demonstration to verify the real effects of this method.

[0100] Select the T-shaped connected corridor on a certain floor of a certain college in a certain university as the actual test site. There is frequent personnel movement and overlapping signal interference in this site. Selecting this scenario can test the impact of environmental variability and signal dynamic characteristics on positioning, ensuring that the test results have high actual representativeness. To fully achieve the effect of fingerprint database aging, one year after the collection of the ground plane offline database, the collection of the empty plane offline database and the ground plane online database is carried out in this area again.

[0101] The RPs are in a uniform grid form, with an adjacent reference point interval of τ = 1 m, and a total of M = 160 RPs are deployed. At the ground plane with a global positioning area height Z 地 = 0.7 m and the empty plane with a height Z 空 = 2 m, N = 313 and J = 362 AP signals are respectively detected. After screening and processing, finally S = 161 APs are left for the experiment. To avoid the dimensional difference during operation, without loss of generality, the AP signal strength not detected at the reference point is filled with -100 dBm. At each reference point, Q = 50 samples are collected, and the sampling time interval is t = 2.3 s. TP is collected in the middle of the T-shaped corridor, with an adjacent test point interval of d = 0.5 m, and a total of 160 test points are sampled. To exclude the potential impact of equipment differences on the evaluation of algorithm performance, the same mobile terminal is used as the data collection tool during both the offline and online processes. All data processing operations are carried out on the Matlab 2022b platform to ensure consistency and comparability.

[0102] To verify the overall positioning performance of an anti-interference indoor positioning method (An Anti-Jamming Indoor Positioning Method, AJIP) proposed in the present invention, the paper selects four positioning algorithms, namely AAS, WKNN, EWKNN, and VWKNN, as comparative algorithms, uses error descriptive statistics as the measurement index, and analyzes the effectiveness of the AJIP method in this embodiment.

[0103] The AAS method can refer to Tao Y, ZHAO L. Fingerprint Localization With Adaptive Area Search [J]. IEEE Communications Letters, 2020, 24(7): 1446 - 1450.

[0104] The WKNN method can be referred to Siyang LIU, de Lacerda R, Fiorina J. WKNN indoor Wi-Fi localization method using k-means clustering based radio mapping[C] / / 2021 IEEE 93rd Vehicular Technology Conference (VTC2021-Spring). IEEE, 2021: 1-5.

[0105] The EWKNN method can be referred to Wang Peizhong, Zheng Nanshan, Zhang Yanzhe. Indoor positioning algorithm based on dynamic K value and AP MAC address screening[J]. Computer Science, 2016, 43(01): 163-165.

[0106] The VWKNN method can be referred to Xu Tian, He Jingsha, Zhu Naifei, etc. Improved VWKNN location fingerprint positioning algorithm based on coefficient of dispersion[J]. Journal of Beijing University of Aeronautics and Astronautics, 2022, 48(7): 1242-1251.

[0107] Moreover, to ensure the comparability of the experiments, the parameters of the comparison algorithms have been adjusted to the optimal configuration in this experimental scenario.

[0108] Figure 2 The cumulative probability distribution of the positioning errors of the 5 algorithms is given. Generally speaking, the upward trends of the 5 algorithms are relatively consistent. However, after the positioning error is greater than 0.8m, the AJIP algorithm proposed in this embodiment begins to show an advantage, and the cumulative probability of the error reaches 30.63%. Moreover, the cumulative probability of the error of the AJIP algorithm proposed in this embodiment within 2m reaches 73.13%, and the cumulative probability of the error within 3m reaches 85.63%, which is better than the other 4 comparison algorithms by the best optimization of 10%. Although the AJIP algorithm proposed in this embodiment is not always the best between 0.4m and 0.8m, it can be seen that there is not a large gap between the AJIP algorithm proposed in this embodiment and the comparison algorithms at this time. The maximum error of the AJIP algorithm proposed in this embodiment within this range is only slightly worse than the optimal algorithm VWKNN at this time by 3.13%.

[0109] The VWKNN algorithm improves the positioning accuracy by pre-eliminating APs with strong volatility. However, some APs with strong volatility but high single-point discrimination are also eliminated. The AJIP algorithm proposed in this embodiment retains APs with high volatility but strong discrimination through double-plane screening, with only a slight loss in accuracy at some test points TP, and it is in an advantageous state in most regions. The EWKNN algorithm finds neighboring points through a threshold, but it is overly dependent on the number of neighboring points. When the number of neighboring points is small, the threshold lacks representativeness, resulting in poor positioning performance. The WKNN algorithm is a classic positioning algorithm that can adapt to the current environment, but fixing the K value will include reference points RP k that are far away in the calculation. Therefore, it has a high degree of fitting with the VWKNN algorithm and the EWKNN algorithm; The AAS algorithm sorts the reference points RP k by the difference in AP signal strength to find better neighboring points. However, when the AP signal of the test point TP fluctuates violently, the positioning stability will be reduced. Therefore, the cumulative error distribution is lower than that of the AJIP algorithm proposed in this embodiment in the long term. Experiments show that the AJIP algorithm proposed in this embodiment has strong dynamic adaptability and can effectively screen out outliers and find the best neighboring points to improve the positioning accuracy in a complex environment.

[0110] The box plot of the positioning error distribution of the AJIP algorithm proposed in this embodiment and the other 4 comparison algorithms is as Figure 3 shown. Compared with the other 4 algorithms, the positioning errors of the AJIP proposed in this embodiment are more concentrated in the low-error region less than 2.147m, far less than the other four algorithms. The middle 50% data interval of the positioning error of the AJIP proposed in this embodiment is the narrowest, which is 1.420m, and the median error is the smallest, which is 1.286m. Although the number of abnormal test points whose positioning errors exceed 1.5 times the middle 50% data interval is more than that of WKNN and VWKNN, the number of abnormal points is controlled between 1 and 2, and the error values of the abnormal points are less than those of the other 2 algorithms. It can be seen that the AJIP algorithm proposed in this embodiment has strong anti-interference ability because it can eliminate non-neighboring reference points RP to a large extent through multi-resolution th hierarchical screening and the constraint of the line concentration threshold L k , improving the positioning accuracy.

[0111] Table 1 shows the performance comparison results of five algorithms. The proposed positioning algorithm AJIP in this embodiment is significantly better than the best cases of the other four algorithms in terms of average error and error variance, reaching 14.8% and 23.8% respectively. Moreover, the maximum error of AJIP proposed in this embodiment is also 3.3% better than the best of the other four algorithms. Although its minimum error is not the best, it is only 0.0034 m less than the best WKNN algorithm. The average positioning error of AJIP proposed in this embodiment is 1.6840 m, with a precision improvement of 14.8% to 19.5% compared to the comparative algorithms. The experimental results show that AJIP proposed in this embodiment has great advantages in solving the problems of aging and movement of access point devices and signal interference in the environment. This is because the fingerprint database construction strategy of the double plane can screen out unstable and low - contribution access points AP to a large extent, and the use of the conjugate space CS during online matching can maximize the similar features between the test point TP and the reference point RP k and evaluate the rationality of the physical layout of the reference point, increasing the rationality of the selection of the reference point RP k .

[0112] Table 1 Performance Comparison of 5 Positioning Algorithms

[0113]

[0114] Example 3, referring to Figure 4 , is the third embodiment of the present invention. The difference from the previous two embodiments is that: a system for an anti - interference indoor positioning method, characterized in that it includes a spatial model establishment module 100, a signal domain near - neighbor point screening module 200, a physical domain near - neighbor point screening module 300, and a conjugate space positioning module 400; the spatial model establishment module 100 is used to detect wireless access points in the positioning area, deploy reference points, establish a ground - plane original fingerprint database, and determine the sub - area of the test point TP; the signal domain near - neighbor point screening module 200 uses a distance measurement method to screen signal domain near - neighbor fingerprint reference points in the area where the test point TP is located to obtain a signal domain near - neighbor reference point set; the physical domain near - neighbor point screening module 300 uses the rationality of the physical geometric structure of the reference point as the basis for judging whether the signal domain near - neighbor reference point is credible, so as to obtain a physically highly credible near - neighbor reference point set; the conjugate space positioning module 400 selects K near - neighbor reference points from the near - neighbor reference point set and uses the WKNN method to estimate the location of the point TP to be located.

[0115] When the above-mentioned functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0116] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in combination with an instruction execution system, apparatus, or device.

[0117] More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection parts with one or more wirings (electronic devices), portable computer disk cartridges (magnetic devices), random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memories), optical fiber devices, and portable compact disc read-only memories (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, then editing, interpreting, or otherwise processing it as necessary, and then storing it in a computer memory.

[0118] It should be understood that each part of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0119] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An anti-interference indoor positioning method, characterized in that: include, Determine ground wireless access points and aerial wireless access points in the positioning area, deploy ground fingerprint reference points and aerial fingerprint reference points, and determine a unique physical address MAC for each wireless access point; The wireless access points APs coexisting on the ground and in the air are extracted through MAC, and S MAC addresses are obtained by intersection to determine the ground plane AP library. The ground plane original fingerprint library is obtained based on the joint coordinates of the AP library. The signal strength of the test point TP is collected, and the sub-area of ​​the TP is roughly located according to the Euclidean distance of the signal strength between the reference point and the TP, and a set of neighboring reference points of the sub-area is obtained; Selecting neighboring fingerprint reference points in the signal domain in the sub-region by using a multi-resolution method to obtain a signal domain neighboring reference point set; Select K neighbor reference points from the signal domain neighbor reference point set, and use the line aggregation method to screen again from the physical domain to obtain the physical domain neighbor reference point set; Based on the set of neighboring reference points under dual screening of signal domain and physical domain, the WKNN algorithm is used to obtain the position estimation of the test point TP in the conjugate space.

2. The anti-interference indoor positioning method according to claim 1, characterized in that: The method for determining the ground wireless access point includes detecting N ground wireless access points AP at the ground level within the positioning area Ω in the offline stage. n ; The deployment method of the ground fingerprint reference points includes deploying M fingerprint reference points RP at equal intervals on the ground plane. m ; The deployment method of the aerial fingerprint reference point includes deploying the reference point in the same manner as the ground plane on an empty plane with the same two-dimensional coordinates as the ground plane but with a height Z. The method for determining the aerial wireless access point includes detecting J wireless access points AP on the aerial plane. j .

3. An anti-interference indoor positioning method as claimed in claim 2, characterized in that: The method of selecting the neighboring fingerprint reference points in the signal domain in the sub-region by using the multi-resolution method includes calculating the sub-region Ω based on the Pearson coefficient. u All neighboring reference points RP k Correlation coefficient ξ with the test point TP k It is expressed as, Among them, E TP It is represented by the RSS measurement value of the ground plane wireless access point contained in the test point TP. RSS represents the received signal strength, G k It is represented by the RSS measurement value of the ground plane wireless access point included in the neighboring reference points, Cov(G k ,E TP ) represents the nearest reference point RP k Covariance with the signal strength of the test point TP, Represented as ground plane RP k AP collected at n Signal variance, Represents the AP collected at the test point TP n Signal variance; Based on the correlation coefficient ξ k , set the multi-resolution model to divide the range layer by layer, the formula is expressed as, Among them, ε k Expressed as number of layers; and They are represented as the lower and upper limits of each layer, respectively, g is represented as the ascending gradient and g>0; Define ε k The layer resolution is expressed as, The resolution is thus obtained With the number of layers ε k The mapping relationship is expressed as: Here, ε is represented as a very small non-negative number to prevent the denominator from being 0. Expressed as a floor function, according to the resolution Computation level ε k ; Calculate the signal Euclidean distance between the nearest reference point of this layer and the test point TP Select The smallest σ t reference points RP as trusted neighbors, and the final candidate reference point RP set in the signal space is obtained.

4. An anti-interference indoor positioning method as claimed in claim 3, characterized in that: The selecting K neighbor reference points from the signal domain neighbor reference point set includes: for the signal domain reference point set According to the signal Euclidean scale, the reference points RP are sorted in ascending order. Take the reference point RP with the smallest Euclidean distance as the initial reference point RP initial ; From RP initial The grouping is started based on the Euclidean distance, where each group selects three reference points with the closest distance as a reference point set {RP k ,RP k+1 ,RP k+2 }, each reference point RP participates in the selection only once, thus obtaining multiple sets of reference points.

5. An anti-interference indoor positioning method as claimed in claim 4, characterized in that: The method of using the line aggregation method to screen again from the physical domain to obtain a physical domain neighbor reference point set also includes, for the reference point set {RP k ,RP k+1 ,RP k+2 }, calculate RP respectively k With RP k+1 RP k With RP k+2 RP k+1 With RP k+2 Distance The stability of each angle is determined by the triangular cosine theorem. The formula is: Among them, χ k , χ k+1 , χ k+2 Expressed as {RP k ,RP k+1 ,RP k+2 }, let the current group reference point set {RP k ,RP k+1 ,RP k+2 The linear convergence of} is, The calculated linear convergence is the linear convergence of the current group of reference points, that is, each group of reference points shares a unified linear convergence, denoted as Traversal If the number of RPs in the remaining reference point combination cannot form a triangle, that is, when the number of remaining RPs is 2 or when the number of remaining RPs is 1, the line convergence is set to 0, that is, or Calculate the mean L of all reference points th It is expressed as, like Then discard the current reference point set {RP k }, get the final physical domain reference point set 6. An anti-interference indoor positioning method as claimed in claim 5, characterized in that: The WKNN algorithm is used to obtain the position estimation of the test point TP in the conjugate space, including: based on the neighbor point set screened by the signal and physical domains. The calculated conjugate space value is expressed as, C sk =ξ k +iχ k Among them, C sk Denoted as reference point RP k The proximity degree to the signal domain of the test point TP and the rationality of the physical structure, i is expressed as an imaginary unit; For the neighboring reference point RP k Set, respectively configure the conjugate weights expressed as, Here, ε is a very small non-negative number to prevent the denominator from being zero.

7. An anti-interference indoor positioning method as claimed in claim 6, characterized in that: The method of using the WKNN algorithm to obtain the position estimation of the test point TP in the conjugate space also includes introducing a negative exponential function Using smoothing characteristics to limit the neighboring reference points RP k The conjugate weight fluctuation range of ; Neighbor set based on dual-domain screening of signal and physical domains As the coordinates of the test point TP (x t ,y t ) is an ideal reference set for position estimation, by indexing the neighboring reference points RP k The positioning weight of the conjugate space is obtained based on WKNN and is expressed as: in, Denoted as reference point RP k The weight of (x k ,y k ) is represented as the reference point RP k The physical coordinates of .

8. A system using an anti-interference indoor positioning method as claimed in any one of claims 1 to 7, characterized in that: It includes a space model building module (100), a signal domain neighbor point screening module (200), a physical domain neighbor point screening module (300) and a conjugate space positioning module (400); The space model building module (100) is used to detect wireless access points in the positioning area, deploy reference points, build a ground plane original fingerprint library, and determine the sub-area of ​​the test point TP; The signal domain neighbor point screening module (200) uses a distance measurement method to screen signal domain neighbor fingerprint reference points in the area to which the test point TP belongs, and obtains a signal domain neighbor reference point set; The physical domain neighbor point screening module (300) uses the rationality of the physical geometric structure of the reference point as a basis for judging whether the signal domain neighbor reference point is credible, thereby obtaining a set of highly credible neighbor reference points in the physical domain; The conjugate space positioning module (400) selects K neighbor reference points from a set of neighbor reference points and uses the WKNN method to estimate the position of the point to be positioned TP.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of an anti-interference indoor positioning method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of an anti-interference indoor positioning method according to any one of claims 1 to 7 are implemented.

Images