An adaptive WKNN indoor positioning method, system, device, and medium

Abstract

This invention discloses an adaptive WKNN indoor positioning method, system, device, and medium, belonging to the field of wireless positioning technology. The method includes: collecting RSSI data of all beacons at each reference point to construct a fingerprint database and obtaining the confidence level of each beacon at each reference point; collecting RSSI data of each beacon at the point to be positioned in real time; filtering reference points based on the difference in beacon visibility between the point to be positioned and the reference points, and obtaining the comprehensive similarity between each filtered reference point and the point to be positioned; adaptively selecting a set number of target reference points, weighted fusing the positions of all target reference points, and obtaining the position of the point to be positioned. This invention can overcome device differences, perform differentiated processing on RSSI data, reduce the interference of overall bias / uniform gain on matching, and suppress the amplification effect of unstable signals, thereby improving the overall positioning accuracy and robustness in complex indoor environments.

Description

Technical Field

[0001] This invention belongs to the field of wireless positioning technology, specifically relating to an adaptive WKNN indoor positioning method, system, device, and medium. Background Technology

[0002] Indoor positioning, as a key foundational capability for smart cities, smart factories, commercial navigation, and security patrols, has evolved into a multi-technology parallel development pattern. Among them, fingerprint positioning based on Received Signal Strength Indicator (RSSI) is widely used due to its advantages such as low deployment cost, minimal modification to existing networks, and compatibility with multiple terminal brands. A typical process includes: collecting RSSIs of multiple beacons (such as Wi-FiAP and BLE Beacon) at reference points offline to build a fingerprint database; and obtaining the RSSI vector of the terminal to be located online, performing similarity matching with the fingerprint database, and outputting a location estimate.

[0003] Among various fingerprint algorithms, Weighted K-Nearest Neighbors (WKNN) is widely used due to its simplicity and superior robustness compared to KNN. Its basic idea is to calculate the similarity between the fingerprint to be located and the fingerprints of each reference point (usually based on magnitude measures such as Euclidean distance or Manhattan distance), select the K reference points with the highest similarity, and sum them according to their similarity weights to obtain the location estimate. However, with the increase in application scale and environmental complexity, several inherent assumptions and engineering simplifications of traditional WKNN have gradually revealed limitations, mainly in the following aspects:

[0004] 1. Amplitude metrics are sensitive to overall bias / uniform gain;

[0005] Different terminal models, transmit power control, temperature and battery status, and front-end gain differences can cause an overall shift or uniform scaling of RSSI. When measured solely based on amplitude differences, this type of global offset systematically amplifies the distance between originally similar fingerprints, causing the reference point order to be reversed, thus leading to position estimation errors.

[0006] 2. The beacon stability exhibits significant heteroscedasticity and time-varying characteristics;

[0007] Due to factors such as obstruction, pedestrian traffic, multipath propagation, and antenna orientation, the RSSI fluctuation amplitude of different beacons varies significantly and changes over time. If a uniform weight or a metric that does not distinguish between stability differences is used, unstable beacons can easily have a disproportionate adverse effect on distance and the weighting process, reducing positioning accuracy and robustness.

[0008] 3. The visibility pattern is not pre-constrained;

[0009] In real-world environments, the set of visible beacons between the target point and the reference point often varies due to differences in floor level, room, or wall structure. Directly calculating distances in the amplitude space can easily introduce false neighbors across floors or walls. The lack of consistent gating based on visibility patterns reduces the reliability of candidate neighborhoods from the outset.

[0010] 4. The selection of the K value often relies on fixed experience or heuristic thresholds;

[0011] A fixed K is difficult to take into account different distribution patterns such as dense or sparse fingerprints, sharp or flat similarity; a heuristic dynamic K based on the mean or threshold is also difficult to characterize the similarity distribution coverage or risk level, resulting in alternating overfitting (K too small) or underfitting (K too large).

[0012] 5. A single image channel lacks utilization of order invariance;

[0013] Traditional similarity methods only compare RSSI values ​​themselves, failing to utilize the structural information of strength order, which is inherently insensitive to overall translation / scaling. When device bias or uniform gain variations exist, the discriminative power of a single image channel significantly decreases.

[0014] 6. Over-reliance on a single strong beacon and lack of robustness assessment;

[0015] In some scenarios, individual high-power / near-range beacons may dominate similarity assessment; once the beacon is temporarily abnormal or occluded, the matching is easily driven by single-point distortion, and the traditional process lacks a robust correction mechanism that is sensitive to beacons.

[0016] In summary, the existing WKNN system still has structural shortcomings in terms of accuracy and stability when facing device differences, beacon heteroscedasticity, NLOS / occlusion, multimodal and time-varying environments. Summary of the Invention

[0017] This invention addresses the shortcomings of existing technologies by providing an adaptive WKNN indoor positioning method, system, device, and medium. It can overcome device differences, perform differentiated processing on RSSI data, reduce the interference of overall bias / uniform gain on matching, and suppress the amplification effect of unstable signals, thereby improving the overall positioning accuracy and robustness in complex indoor environments.

[0018] This invention provides the following technical solution:

[0019] Firstly, an adaptive WKNN indoor positioning method is provided, wherein several beacons and reference points are deployed indoors; including the following steps:

[0020] S1: Collect RSSI data of all beacons at each reference point to build a fingerprint database, and obtain the confidence level of each beacon at each reference point based on the stability of each beacon at the reference point;

[0021] S2: The RSSI data of each beacon is collected in real time at the location to be located;

[0022] S3: Based on the difference in beacon visibility between the target point and the reference point, filter reference points, and for each filtered reference point, obtain its comprehensive similarity with the target point. Specifically, based on the confidence of the beacon at the reference point and the RSSI data of all beacons received by the reference point and the target point, obtain the similarity between the reference point and the target point from the amplitude and order dual channels respectively, and then fuse them to obtain the comprehensive similarity between the reference point and the target point.

[0023] S4: Adaptively select a set number of target reference points, and based on the comprehensive similarity between the target reference points and the point to be located, weightedly fuse the positions of all target reference points to obtain the position of the point to be located.

[0024] Optionally, the fingerprint database includes: fingerprint data samples of all reference points; wherein, the fingerprint data samples of the reference points include: the position coordinates of the reference points and the RSSI vector of the reference points;

[0025] The RSSI vector of the reference point is constructed by repeatedly collecting RSSI data from all beacons at the reference point, and then performing moving average and noise reduction filtering to obtain the RSSI vector of the reference point.

[0026] Optionally, in step S1, obtaining the confidence level of the beacon at each reference point includes obtaining the beacon... b At the reference point r The confidence level, specifically, is based on the reference point. r Multiple beacon collections b variance of RSSI data The confidence level is obtained according to the following formula. Then normalization is performed;

[0027] ;

[0028] in, To prevent constants with a denominator of zero.

[0029] Optionally, in step S3, the step of filtering reference points based on the difference in beacon visibility between the point to be located and the reference point specifically involves:

[0030] S31: Define the point to be located x visibility vector and reference points r visibility vector ;

[0031] ;

[0032] ;

[0033] in, , M The total number of beacons. and Reference points r and the point to be located x RSSI vector with respect to beacon b RSSI value;

[0034] S32: Obtain the point to be located by calculating the Hamming distance between two visibility vectors. x and reference points r Visibility difference If visibility difference If the value exceeds the set threshold, the reference point is removed. r Otherwise, the reference point is retained. r Complete the screening;

[0035] ;

[0036] in, The Hamming distance is the visibility vector between the point to be located and the reference point.

[0037] Optionally, in step S3, obtaining the comprehensive similarity between each of the filtered reference points and the point to be located includes obtaining the point to be located. x With reference point r The overall similarity is as follows:

[0038] S33: Construct the point to be located x With reference point r Commonly visible beacon set ,in, and The points to be located are respectively x With reference point r Visibility vector with respect to common visibility beacons p The amount;

[0039] S34: Based on the RSSI data of all beacons received from the reference point and the point to be located, and the confidence level of the beacons at the reference point, calculate the reference point. r With the point to be located x Weighted Euclidean distance The amplitude channel similarity is obtained by mapping it to the interval [0,1]. ;

[0040] S35: Place the points to be located respectively x and reference points r Common Visible Beacon Set P Sort the corresponding RSSI values ​​in descending order to obtain the location of the point to be located. x and reference points r The two corresponding orders; using the Kendall rank correlation coefficient The consistency between two orders is measured and mapped onto the interval [0,1] to obtain the order channel similarity. ;

[0041] S36: Based on preset fusion weights The point to be located is obtained by weighting. x With reference point r Overall similarity ;

[0042] .

[0043] Optionally, in step S34, the reference point r With the point to be located x Weighted Euclidean distance for: ,in, and The points to be located are respectively x and reference points r RSSI vectors with respect to common visible beacons p RSSI value, beacon p At the reference point r Normalized confidence level; amplitude channel similarity for: .

[0044] Optionally, step S4 specifically includes:

[0045] S41: Sort the comprehensive similarity between the point to be located and each filtered reference point in descending order to obtain a comprehensive similarity descending sequence;

[0046] S42: Let To calculate the cumulative sum of the first K elements of the descending sequence based on overall similarity, To find the smallest integer K such that the sum of all elements in a descending sequence of comprehensive similarity satisfies... ,in, To set the coverage weight value;

[0047] S43: The boundary of the preset K value. If the obtained K is within the boundary, the number of target reference points is determined to be K. If the obtained K is outside the boundary, the number of target reference points is determined to be boundary points. If the total number of elements in the descending sequence of comprehensive similarity is less than the boundary minimum value of K, the number of target reference points is the total number of elements in the descending sequence of comprehensive similarity.

[0048] S44: Weight the positions of all target reference points to obtain the position of the point to be located. ;

[0049] ;

[0050] in, The normalized target reference point k With the point to be located x The overall similarity value, Target reference point k The location coordinates.

[0051] Secondly, an adaptive WKNN indoor positioning system is provided, including:

[0052] Several beacons are deployed in the environment where positioning is required, according to a set spatial interval;

[0053] Several reference points are evenly distributed in the positioning environment according to the path nodes or grid density of the target area;

[0054] Fingerprint database construction module: Collect RSSI data of all beacons at each reference point to construct a fingerprint database, and obtain the confidence level of each beacon at each reference point based on the stability of each beacon at the reference point;

[0055] Real-time acquisition module: The RSSI data of each beacon is acquired in real time at the point to be located;

[0056] The comprehensive similarity acquisition module filters reference points based on the difference in beacon visibility between the target point and the reference point. For each filtered reference point, it acquires the comprehensive similarity between the target point and the reference point. Specifically, based on the confidence of the beacon at the reference point and the RSSI data of all beacons received by the reference point and the target point, it acquires the similarity between the reference point and the target point from both amplitude and order channels, and then fuses them to obtain the comprehensive similarity between the reference point and the target point.

[0057] The localization module adaptively selects a set number of target reference points and, based on the comprehensive similarity between the target reference points and the point to be located, weightedly fuses the positions of all target reference points to obtain the position of the point to be located.

[0058] Thirdly, a computer device is provided, including a processor and a memory; wherein the processor executes a computer program stored in the memory to implement the steps of the adaptive WKNN indoor positioning method according to any one of the first aspects.

[0059] Fourthly, a computer-readable storage medium is provided for storing a computer program; when the computer program is executed by a processor, it implements the steps of the adaptive WKNN indoor positioning method as described in any one of the first aspects.

[0060] Compared with the prior art, the beneficial effects of the present invention are:

[0061] This application, on the one hand, utilizes the invariance of the strength order in the similarity construction to reduce the interference of the overall bias / uniform gain on matching; on the other hand, in the candidate neighborhood screening and weighting process, it combines beacon stability confidence and visibility pattern gating, and suppresses the amplification effect of unstable signals through mechanisms such as adaptive K-value and robustness correction, thereby improving the overall positioning accuracy and robustness in complex indoor environments. Therefore, this application can overcome device differences, perform differentiated processing on RSSI data, reduce the interference of the overall bias / uniform gain on matching, and improve indoor positioning accuracy and stability, making it particularly suitable for complex and changeable indoor environments. Attached Figure Description

[0062] Figure 1 This is a flowchart of the adaptive WKNN indoor positioning method of the present invention;

[0063] Figure 2 This is a structural block diagram of the adaptive WKNN indoor positioning system of the present invention. Detailed Implementation

[0064] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be used to limit the scope of protection of the present invention. It should be noted that the term "comprising" and any variations thereof in the specification, claims and the above-mentioned drawings of the present invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or devices.

[0065] Example 1

[0066] like Figure 1 As shown, an adaptive WKNN indoor positioning method is provided, which includes the following steps:

[0067] S1: Collect RSSI data of all beacons at each reference point to build a fingerprint database, and obtain the confidence level of each beacon at each reference point based on the stability of each beacon at the reference point.

[0068] In indoor or semi-outdoor environments requiring positioning, several Bluetooth beacon nodes are deployed at certain spatial intervals. The number and location of the beacons can be flexibly set according to the site area, the distribution of obstructions, and the required positioning accuracy. All beacons are set to broadcast periodically, with consistent operating frequency and transmission power to ensure the comparability and reproducibility of RSSI values.

[0069] Multiple reference points are manually set within the target area. Each reference point has a clear two-dimensional location coordinate. The reference points can be evenly distributed based on grid density, path nodes, or regions of interest.

[0070] The fingerprint database includes fingerprint data samples of all reference points; wherein, the fingerprint data samples of reference points include: the location coordinates of the reference points and the RSSI vector of the reference points.

[0071] S11: Construct the RSSI vector of the reference point, specifically:

[0072] At each reference point, RSSI data was collected using a positioning terminal device (such as a smartphone). To enhance the stability and representativeness of the data, multiple consecutive RSSI samples were collected at each point (sampling once per second for a total of 30 seconds), and the timestamp, beacon number, and received signal strength value corresponding to each RSSI vector were recorded.

[0073] The reference point repeatedly collects RSSI data from all beacons, and after performing moving average and noise reduction filtering, obtains the RSSI vector for the reference point. Missing values ​​in the RSSI vector can be filled with -100dBm.

[0074] S12: Obtain the confidence level of the beacon at each reference point, including obtaining the beacon's confidence level. b At the reference point r The confidence level is as follows:

[0075] Based on reference point r Multiple beacon collections b variance of RSSI data The confidence level is obtained according to the following formula. Then normalization is performed;

[0076] ;

[0077] in, To prevent the denominator from being zero, a constant is optionally set to 0.1 dB. 2 .

[0078] In the offline phase, to improve the ability to characterize the stability of each beacon at the reference point, statistical RSSI variance is used. A higher confidence score indicates a more stable beacon. b At the reference point r The more reliable the signal, the better.

[0079] For ease of calculation, the confidence level is... Perform a normalization transformation to obtain the normalized confidence weights. :

[0080] ;

[0081] in, This represents the total number of beacons.

[0082] S2: The RSSI data of each beacon is collected in real time at the location to be located.

[0083] In the actual positioning process, the mobile terminal first uses its Bluetooth module to broadcast and listen to multiple beacons deployed in the environment, collecting the Received Signal Strength Indication (RSSI) value of each beacon at the current moment. Assuming there are M Bluetooth beacons deployed in the environment, any point to be located can form an RSSI vector of dimension M, denoted as:

[0084] ;

[0085] in This indicates that the point to be located has received the first... b The RSSI value of each beacon, in dBm.

[0086] Considering that Bluetooth signals are susceptible to interference from multipath fading and obstruction in indoor environments, this invention employs a time window sampling mechanism for preprocessing to improve the stability and usability of RSSI vectors. Specifically, n sets of RSSI data are continuously collected within a short period (e.g., 1-2 seconds); noise is denoised from the n RSSI samples for each beacon, and the average is taken; the final input vector is then constructed. It is used for similarity calculation and location inference with fingerprint database samples; the default RSSI value of uncollected beacons is -100dBm.

[0087] S3: Based on the difference in beacon visibility between the target point and the reference point, filter reference points, and for each filtered reference point, obtain its comprehensive similarity with the target point. Specifically, based on the confidence of the beacon at the reference point and the RSSI data of all beacons received by the reference point and the target point, obtain the similarity between the reference point and the target point from the amplitude and order dual channels respectively, and then fuse them to obtain the comprehensive similarity between the reference point and the target point.

[0088] Step S3 specifically includes the following sub-steps:

[0089] S31: Define the point to be located x visibility vector and reference points r visibility vector ;

[0090] ;

[0091] ;

[0092] in, , M The total number of beacons. and Reference points r and the point to be located x RSSI vector with respect to beacon b The RSSI value; since uncollected RSSI values ​​will be assigned a value of -100dBm, therefore, and When missing, beacon b The value is -100 dBm; of course, other values ​​can also be assigned according to the actual situation.

[0093] S32: Obtain the point to be located by calculating the Hamming distance between two visibility vectors. x and reference points r Visibility difference If visibility difference Greater than the set threshold Then remove the reference point. r Otherwise, the reference point is retained. r Complete the screening;

[0094] ;

[0095] in, Let be the Hamming distance between the visibility vectors of the point to be located and the reference point, i.e. The number of distinct elements between the two vectors.

[0096] Alternatively, threshold The value is typically 0.3; if the visibility difference is... Greater than the set threshold This proves that the point to be located is not connected to the reference point, and the reference point is discarded. r Otherwise, it proves that the point to be located is connected to the reference point, and the reference point is retained.

[0097] By defining a visibility vector, the visibility consistency between the point to be located and the reference point is checked, and only spatially connected reference points are retained, thereby suppressing the "false nearest neighbor" problem of cross-regional mismatch.

[0098] S33: Construct the point to be located x With reference point r Commonly visible beacon set ,in, and The points to be located are respectively x With reference point r Visibility vector with respect to common visibility beacons p The amount.

[0099] S34: Based on the RSSI data of all beacons received from the reference point and the point to be located, and the confidence level of the beacons at the reference point, calculate the reference point. r With the point to be located x Weighted Euclidean distance The amplitude channel similarity is obtained by mapping it to the interval [0,1]. .

[0100] Specifically, reference point r With the point to be located x Weighted Euclidean distance for:

[0101] ;

[0102] in, and The points to be located are respectively x and reference points r RSSI vectors with respect to common visible beacons p RSSI value, beacon p At the reference point r Normalized confidence level;

[0103] Amplitude channel similarity for:

[0104] .

[0105] By calculating the similarity of amplitude channels, it is possible to suppress noise interference from high-fluctuation beacons and improve the stability of signal values.

[0106] S35: Place the points to be located respectively x and reference points r Common Visible Beacon Set P Sort the corresponding RSSI values ​​in descending order to obtain the location of the point to be located. xand reference points r The two corresponding orders; using the Kendall rank correlation coefficient The consistency between two orders is measured and mapped onto the interval [0,1] to obtain the order channel similarity. .

[0107] Calculate the point to be located x and reference points r Kendall rank correlation coefficients of the two corresponding orders The approach can refer to existing technologies, specifically:

[0108] Step 1: Generating Order: and Commonly visible beacon set P Sort the RSSI values ​​in descending order to obtain the order. , Among them, if the interpolation between two beacons is less than a threshold... If the value is 2dB, it is considered as parallel and the average rank is used for calculation.

[0109] Step 2: Calculate the Kendall rank correlation coefficient For two ordered sequences The relative order of each element is determined to obtain the number of identical pairs and inversion pairs, and then the Kendall rank correlation coefficient is calculated to measure the consistency between the two sequences.

[0110] ;

[0111] in, Represented as a pair of identically ordered objects, the two beacons are in... The relative order is consistent. Represented as an inversion pair, the two beacons are in The relative order is inconsistent; Represented as the total number of beacon pairs, calculated as follows: ,in This represents the number of common visible beacons for the point to be located and the reference point.

[0112] Step 3: Calculate the Kendall rank correlation coefficient. The similarity range [-1,1] is mapped to [0,1] to obtain the order channel similarity. ;

[0113] .

[0114] S36: Based on preset fusion weights The point to be located is obtained by weighting. x With reference point r Overall similarity ;

[0115] .

[0116] Fusion weights This indicates the proportion of the order channel in the overall similarity score, and its function is optional, incorporating weights. The value is set to 0.4. Of course, in some other embodiments, a global bias is used for weight fusion. Adaptive adjustment.

[0117] S4: Adaptively select a set number of target reference points, and based on the comprehensive similarity between the target reference points and the point to be located, weightedly fuse the positions of all target reference points to obtain the position of the point to be located.

[0118] Step S4 specifically includes the following sub-steps:

[0119] S41: Sort the comprehensive similarity between the point to be located and each filtered reference point in descending order to obtain a descending sequence of comprehensive similarity.

[0120] S42: Let To calculate the cumulative sum of the first K elements of the descending sequence based on overall similarity, To find the smallest integer K such that the sum of all elements in a descending sequence of comprehensive similarity satisfies... ,in, To set the coverage weight value.

[0121] That is, the cumulative similarity of the first K reference points must reach the total similarity. Times, under normal circumstances The value is 0.85. Substitute each value of K from 1 to N into the summation, and iterate. When the condition is met... When the condition is met, K is the smallest K that satisfies the condition.

[0122] ;

[0123] ;

[0124] S43: The boundary of the preset K value. If the obtained K is within the boundary, the number of target reference points is determined to be K; if the obtained K is outside the boundary, the number of target reference points is determined to be boundary points; if the total number of elements in the descending sequence of comprehensive similarity is less than the boundary minimum value of K, the number of target reference points is the total number of elements in the descending sequence of comprehensive similarity.

[0125] When the value of K is too small, that is, when When, K is adjusted to .

[0126] When the value of K is too large, that is, when When, K is adjusted to .

[0127] When there are insufficient candidate points (the total number of elements in the descending sequence of comprehensive similarity is less than the boundary minimum of K), K is taken as the number of all available points.

[0128] S44: Weight the positions of all target reference points to obtain the position of the point to be located. ;

[0129] ;

[0130] in, The normalized target reference point k With the point to be located x The overall similarity value, Target reference point k The location coordinates.

[0131] Compared to the traditional WKNN algorithm, which directly uses the reciprocal of the Euclidean distance between K reference points as weights, the weight fusion mechanism adopted in this invention introduces both RSSI confidence factor and signal difference metric, which is more adaptable to the uncertainty caused by channel quality differences in different environments and improves positioning robustness.

[0132] Example 2

[0133] like Figure 2 As shown, an adaptive WKNN indoor positioning system includes:

[0134] Several beacons are deployed in the environment where positioning is required, according to a set spatial interval;

[0135] Several reference points are evenly distributed in the positioning environment according to the path nodes or grid density of the target area;

[0136] Fingerprint database construction module: Collect RSSI data of all beacons at each reference point to construct a fingerprint database, and obtain the confidence level of each beacon at each reference point based on the stability of each beacon at the reference point;

[0137] Real-time acquisition module: The RSSI data of each beacon is acquired in real time at the point to be located;

[0138] The comprehensive similarity acquisition module filters reference points based on the difference in beacon visibility between the target point and the reference point. For each filtered reference point, it acquires the comprehensive similarity between the target point and the reference point. Specifically, based on the confidence of the beacon at the reference point and the RSSI data of all beacons received by the reference point and the target point, it acquires the similarity between the reference point and the target point from both amplitude and order channels, and then fuses them to obtain the comprehensive similarity between the reference point and the target point.

[0139] The localization module adaptively selects a set number of target reference points and, based on the comprehensive similarity between the target reference points and the point to be located, weightedly fuses the positions of all target reference points to obtain the position of the point to be located.

[0140] For more detailed information about the system described above, please refer to the relevant content disclosed in the foregoing embodiments, which will not be repeated here.

[0141] Example 3

[0142] The present invention provides a computer device, including a processor and a memory; wherein, when the processor executes the computer program stored in the memory, it implements the steps of the above-described adaptive WKNN indoor positioning method.

[0143] For more detailed information on the above methods, please refer to the relevant content disclosed in the foregoing embodiments, which will not be repeated here.

[0144] Example 4

[0145] The present invention provides a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, it implements the steps of the above-described adaptive WKNN indoor positioning method.

[0146] For more detailed information on the above methods, please refer to the relevant content disclosed in the foregoing embodiments, which will not be repeated here.

[0147] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. The systems, devices, and storage media disclosed in the embodiments are described simply because they correspond to the methods disclosed in the embodiments; relevant details can be found in the method section.

[0148] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

[0149] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. An adaptive WKNN indoor positioning method, wherein several beacons and reference points are deployed indoors; characterized in that, Includes the following steps: S1: Collect RSSI data of all beacons at each reference point to build a fingerprint database, and obtain the confidence level of each beacon at each reference point based on the stability of each beacon at the reference point; S2: The RSSI data of each beacon is collected in real time at the location to be located; S3: Based on the difference in beacon visibility between the target point and the reference point, filter reference points, and for each filtered reference point, obtain its comprehensive similarity with the target point. Specifically, based on the confidence of the beacon at the reference point and the RSSI data of all beacons received by the reference point and the target point, obtain the similarity between the reference point and the target point from the amplitude and order dual channels respectively, and then fuse them to obtain the comprehensive similarity between the reference point and the target point. Step S3 specifically includes: S31: Define the point to be located x visibility vector and reference points r visibility vector ; ； ； in, , M The total number of beacons. and Reference points r and the point to be located x RSSI vector with respect to beacon b RSSI value; S32: Obtain the point to be located by calculating the Hamming distance between two visibility vectors. x and reference points r Visibility difference If visibility difference If the value exceeds the set threshold, the reference point is removed. r Otherwise, the reference point is retained. r Complete the screening; ； in, The Hamming distance between the visibility vectors of the point to be located and the reference point; S33: Construct the point to be located x With reference point r Commonly visible beacon set ,in, and The points to be located are respectively x With reference point r Visibility vector with respect to common visibility beacons p The amount; S34: Based on the RSSI data of all beacons received from the reference point and the point to be located, and the confidence level of the beacons at the reference point, calculate the reference point. r and the point to be located x Weighted Euclidean distance The amplitude channel similarity is obtained by mapping it to the interval [0,1]. The reference point r and the point to be located x Weighted Euclidean distance for: ,in, and The points to be located are respectively x and reference points r RSSI vectors with respect to common visible beacons p RSSI value, beacon p At the reference point r Normalized confidence level; amplitude channel similarity for: ; S35: Place the points to be located respectively x and reference points r Commonly Visible Beacon Set P Sort the corresponding RSSI values ​​in descending order to obtain the location of the point to be located. x and reference points r The two corresponding orders; using the Kendall rank correlation coefficient The consistency between two orders is measured and mapped onto the interval [0,1] to obtain the order channel similarity. ; S36: Based on preset fusion weights The point to be located is obtained by weighting. x With reference point r Overall similarity ; ； S4: Adaptively select a set number of target reference points, and based on the comprehensive similarity between the target reference points and the point to be located, weightedly fuse the positions of all target reference points to obtain the position of the point to be located.

2. The adaptive WKNN indoor positioning method according to claim 1, characterized in that, In step S1, the fingerprint database includes: fingerprint data samples of all reference points; wherein, the fingerprint data samples of reference points include: the position coordinates of the reference points and the RSSI vector of the reference points; The RSSI vector of the reference point is constructed by repeatedly collecting RSSI data from all beacons at the reference point, and then performing moving average and noise reduction filtering to obtain the RSSI vector of the reference point.

3. The adaptive WKNN indoor positioning method according to claim 2, characterized in that, In step S1, obtaining the confidence level of the beacon at each reference point includes obtaining the beacon... b At the reference point r The confidence level, specifically, is based on the reference point. r Multiple beacon collections b variance of RSSI data The confidence level is obtained according to the following formula. Then normalization is performed; ； in, To prevent constants with a denominator of zero.

4. The adaptive WKNN indoor positioning method according to claim 1, characterized in that, In step S4, specifically: S41: Sort the comprehensive similarity between the point to be located and each filtered reference point in descending order to obtain a comprehensive similarity descending sequence; S42: Let To calculate the cumulative sum of the first K elements of the descending sequence based on overall similarity, To find the smallest integer K such that the sum of all elements in a descending sequence of comprehensive similarity satisfies... ,in, To set the coverage weight value; S43: The boundary of the preset K value. If the obtained K is within the boundary, the number of target reference points is determined to be K. If the obtained K is outside the boundary, the number of target reference points is determined to be boundary points. If the total number of elements in the descending sequence of comprehensive similarity is less than the boundary minimum value of K, the number of target reference points is the total number of elements in the descending sequence of comprehensive similarity. S44: Weight the positions of all target reference points to obtain the position of the point to be located. ; ； in, The normalized target reference point k and the point to be located x The overall similarity value, Target reference point k The location coordinates.

5. An adaptive WKNN indoor positioning system, characterized in that, include: Several beacons are deployed in the environment where positioning is required, according to a set spatial interval; Several reference points are evenly distributed in the positioning environment according to the path nodes or grid density of the target area; Fingerprint database construction module: Collect RSSI data of all beacons at each reference point to construct a fingerprint database, and obtain the confidence level of each beacon at each reference point based on the stability of each beacon at the reference point; Real-time acquisition module: The RSSI data of each beacon is acquired in real time at the point to be located; The comprehensive similarity acquisition module filters reference points based on the difference in beacon visibility between the target point and the reference point. For each filtered reference point, it acquires the comprehensive similarity between the target point and the reference point. Specifically, based on the confidence of the beacon at the reference point and the RSSI data of all beacons received by the reference point and the target point, it acquires the similarity between the reference point and the target point from both amplitude and order channels, and then fuses them to obtain the comprehensive similarity between the reference point and the target point. The process of filtering reference points based on the difference in beacon visibility between the target point and reference points, and then obtaining the comprehensive similarity between each filtered reference point and the target point, specifically includes: Define the point to be located x visibility vector and reference points r visibility vector ; ； ； in, , M The total number of beacons. and Reference points r and the point to be located x RSSI vector with respect to beacon b RSSI value; The point to be located is obtained by calculating the Hamming distance between two visibility vectors. x and reference points r Visibility difference If visibility difference If the value exceeds the set threshold, the reference point is removed. r Otherwise, the reference point is retained. r Complete the screening; ； in, The Hamming distance between the visibility vectors of the point to be located and the reference point; Construct the point to be located x With reference point r Commonly visible beacon set ,in, and The points to be located are respectively x With reference point r Visibility vector with respect to common visibility beacons p The amount; Based on the RSSI data of all beacons received from the reference point and the point to be located, and the confidence level of the beacons at the reference point, the reference point is calculated. r and the point to be located x Weighted Euclidean distance The amplitude channel similarity is obtained by mapping it to the interval [0,1]. The reference point r and the point to be located x Weighted Euclidean distance for: ,in, and The points to be located are respectively x and reference points r RSSI vectors with respect to common visible beacons p RSSI value, beacon p At the reference point r Normalized confidence level; amplitude channel similarity for: ; The points to be located x and reference points r Commonly Visible Beacon Set P Sort the corresponding RSSI values ​​in descending order to obtain the location of the point to be located. x and reference points r The two corresponding orders; using the Kendall rank correlation coefficient The consistency between two orders is measured and mapped onto the interval [0,1] to obtain the order channel similarity. ; Based on preset fusion weights The point to be located is obtained by weighting. x With reference point r Overall similarity ; ； The localization module adaptively selects a set number of target reference points and, based on the comprehensive similarity between the target reference points and the point to be located, weightedly fuses the positions of all target reference points to obtain the position of the point to be located.

6. A computer device, characterized in that, It includes a processor and a memory; wherein, when the processor executes the computer program stored in the memory, it implements the steps of the adaptive WKNN indoor positioning method according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, Used to store computer programs; when executed by a processor, the computer programs implement the steps of the adaptive WKNN indoor positioning method according to any one of claims 1-4.

Images