Indoor positioning method based on a hybrid of Wi-Fi and ultra-wideband technologies

Abstract

This invention discloses an indoor positioning method based on a hybrid Wi-Fi and ultra-wideband (UWB) technology, comprising: Step 1: Deploying a UWB positioning base station and multiple Wi-Fi signal access nodes indoors, recording device locations, and compensating for UWB ranging errors; Step 2: Collecting signal strength values ​​of each access point (AP) at pre-defined fingerprint points, and recording the processed signal strength values ​​and the corresponding location coordinates of the fingerprint points into a fingerprint database; Step 3: Determining multiple fingerprint points and valid APs within a corresponding range based on the distance between the target user and the base station measured by UWB, in order to reduce the solution space and eliminate positioning errors introduced by unreliable APs; Step 4: Using the fingerprint information obtained in Step 3 and the current measured data, calculating the distance to the UWB base station, and selecting the target location estimate with the smallest difference from the UWB distance measurement value as the final positioning result. This invention has the advantages of strong penetration, good anti-interference effect, and high positioning accuracy.

Description

Technical Field

[0001] This invention belongs to the field of indoor positioning system (IPS) technology, and particularly relates to an indoor positioning method based on a hybrid of Wi-Fi and ultra-wideband (UWB) technology. Background Technology

[0002] Indoor Positioning Systems (IPS) are a critical infrastructure for enabling location-based services (LBS) such as indoor navigation and geofencing. Achieving a balance between equipment deployment costs and positioning accuracy is one of the main challenges in current indoor positioning research.

[0003] Wi-Fi and ultra-wideband (UWB) positioning technologies are important means of realizing indoor positioning systems. As an emerging wireless communication technology in recent years, UWB boasts advantages such as strong penetration, good anti-interference performance, and high positioning accuracy; however, its high hardware deployment cost limits its application in indoor positioning. Wi-Fi positioning technology, as a widely used method, offers advantages such as low cost and wide coverage, but its positioning accuracy still cannot fully meet people's needs.

[0004] There is an urgent need for an indoor positioning system that can combine Wi-Fi and ultra-wideband technologies to balance the dual requirements of device deployment cost and positioning accuracy. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention provides an indoor positioning method based on a hybrid of Wi-Fi and ultra-wideband technology. By utilizing the ranging capability of ultra-wideband technology, it further improves indoor positioning accuracy while taking into account the equipment deployment cost.

[0006] The technical solution adopted by this invention to solve its technical problem is:

[0007] An indoor positioning method based on a hybrid of Wi-Fi and ultra-wideband technology, characterized in that the method includes the following steps:

[0008] Step 1: Infrastructure Deployment;

[0009] Deploy a UWB positioning base station and multiple Wi-Fi signal access nodes (APs) indoors, and record the device location to compensate for UWB ranging errors;

[0010] Step 2: Establish an offline fingerprint database;

[0011] The signal strength values ​​of each AP are collected at the pre-set fingerprint points, and the processed signal strength values ​​and the location coordinates corresponding to the fingerprint points are entered into the fingerprint database.

[0012] Step 3: Fingerprint information screening;

[0013] Based on the distance between the target user and the base station obtained by UWB measurement, multiple fingerprint points and valid APs within the corresponding range are determined to reduce the solution space and eliminate the positioning error introduced by unreliable APs.

[0014] Step 4: Target location determined;

[0015] Using the fingerprint information obtained in step 3 and the current measured data, the current target location is estimated through various nearest neighbor algorithms, and the distance between the target and the UWB base station is calculated. The target location estimate with the smallest difference from the UWB distance measurement value is selected as the final positioning result.

[0016] Furthermore, in step 1, the UWB ranging error compensation steps are as follows:

[0017] Step 1-1: Based on the relationship between the actual distance and the UWB measurement error, a univariate regression model is established, yielding the following relationship:

[0018] y=β0+β1x+ε (1)

[0019] In the formula, y represents the dependent variable, x represents the independent variable, β0 and β1 represent the regression coefficients, and ε represents the error term that follows a normal distribution;

[0020] Steps 1-2: Obtain multiple sets of actual distance and UWB measurement error data, and calculate regression parameters:

[0021]

[0022] In the formula, Let represent the estimated values ​​of β1 and β0, respectively. Let x and y represent the average values, respectively.

[0023] Furthermore, in step 2, the steps for establishing the fingerprint database are as follows:

[0024] Step 2-1: Set fingerprint points at certain intervals according to actual needs;

[0025] Step 2-2: Measure the signal strength of each access point (AP) at its current location multiple times at predefined fingerprint points. Then, use Gaussian filtering to remove values ​​with low probability of occurrence. Finally, take the average value as the fingerprint information and enter it into the fingerprint database. The Gaussian probability density function used is:

[0026]

[0027] In the formula, Let μ represent the signal strength value of the j-th AP collected at fingerprint point i. Each fingerprint point is measured n times. μ represents the mean of the signal strength values, and σ is the standard deviation of the signal strength values. The calculation formula is as follows:

[0028]

[0029]

[0030] Furthermore, in step 3, the fingerprint information screening steps are as follows:

[0031] Step 3-1: Use the UWB module to measure the distance d between the target user's current location and the UWB base station;

[0032] Step 3-2, select the option that satisfies |dd i Fingerprint points with |<Δ are considered as candidates, where d i This represents the distance from fingerprint point i to the base station, where Δ is the range threshold.

[0033] Step 3-3: Filter to satisfy α×d+s j ≤L j The AP is used to eliminate positioning errors introduced by unreliable APs, where the parameter α∈(0,1), s j L represents the effective radius of APj. j This represents the distance from AP j to the base station.

[0034] Furthermore, in step 4, the steps for determining the target location are as follows:

[0035] Step 4-1: Measure the signal strength of each AP selected in Step 3 at the target user's current location multiple times, and average the values ​​to reduce positioning errors caused by signal strength fluctuations, thus obtaining the fingerprint vector T = (RSS) for the current location. 1 RSS 2 ,...,RSS N ),in n represents the number of measurements, and RSS is the RSS value. j The signal strength value of AP j;

[0036] Step 4-2: Calculate the fingerprint points selected in Step 3. The Euclidean distance D between T and i :

[0037]

[0038] In the formula, RSS represents the signal strength received at the j-th AP at the i-th fingerprint point. j The signal strength value of APj;

[0039] Calculate the coordinates (X) corresponding to the minimum Euclidean distance value. m ,Y m The mean of the coordinates corresponding to the k smallest Euclidean distances (X) k ,Y k The weighted average of the coordinates corresponding to the k smallest Euclidean distances (X) w ,Y w The formula is as follows:

[0040]

[0041]

[0042]

[0043] In the formula ω i ε represents the weight corresponding to the i-th fingerprint point. ε is a very small number, usually fixed at three orders of magnitude smaller than the test distance, to prevent the denominator from being zero.

[0044] Steps 4-3, calculate (X) respectively. m ,Y m ), (X k ,Y k ), (X w ,Y w The distance between the target location and the UWB base station is determined, and the location with the smallest difference from the UWB module's measured value d is selected as the current target location.

[0045] Compared with the prior art, the beneficial effects of the present invention are mainly reflected in the following aspects: by introducing a single UWB base station for distance measurement in Wi-Fi fingerprint positioning, and by taking advantage of the advantages of UWB such as strong penetration, good anti-interference effect and high positioning accuracy, fingerprint information is screened to eliminate positioning errors introduced by unreliable APs, thereby improving the accuracy of the indoor positioning system while limiting the equipment deployment cost. Attached Figure Description

[0046] Figure 1 This is the overall flowchart of the present invention.

[0047] Figure 2 This is a performance comparison chart of the method in this patent compared to other methods. Detailed Implementation

[0048] The present invention will now be further described with reference to the accompanying drawings.

[0049] Reference Figure 1 and Figure 2The present invention discloses an indoor positioning method based on a hybrid of Wi-Fi and ultra-wideband technology, the method comprising the following steps:

[0050] Step 1: Infrastructure Deployment;

[0051] Deploy a UWB positioning base station and multiple Wi-Fi signal access nodes (APs) indoors, and record the device location to compensate for UWB ranging errors;

[0052] Step 2: Establish an offline fingerprint database;

[0053] The signal strength values ​​of each AP are collected at the pre-set fingerprint points, and the processed signal strength values ​​and the location coordinates corresponding to the fingerprint points are entered into the fingerprint database.

[0054] Step 3: Fingerprint information screening;

[0055] Based on the distance between the target user and the base station obtained by UWB measurement, multiple fingerprint points and valid APs within the corresponding range are determined to reduce the solution space and eliminate the positioning error introduced by unreliable APs.

[0056] Step 4: Target location determined;

[0057] Using the fingerprint information obtained in step 3 and the current measured data, the current target location is estimated through various nearest neighbor algorithms, and the distance between the target and the UWB base station is calculated. The target location estimate with the smallest difference from the UWB distance measurement value is selected as the final positioning result.

[0058] In one embodiment, the UWB ranging error compensation steps in step 1 are as follows:

[0059] Step 1-1: Based on the relationship between the actual distance and the UWB measurement error, a univariate regression model is established, yielding the following relationship:

[0060] y=β0+β1x+ε (1)

[0061] In the formula, y represents the dependent variable, x represents the independent variable, β0 and β1 represent the regression coefficients, and ε represents the error term that follows a normal distribution;

[0062] Steps 1-2: Obtain multiple sets of actual distance and UWB measurement error data, and calculate regression parameters:

[0063]

[0064] In the formula, Let represent the estimated values ​​of β1 and β0, respectively. Let x and y represent the average values, respectively.

[0065] In one embodiment, the fingerprint database establishment process in step 2 is as follows:

[0066] Step 2-1: Set fingerprint points at certain intervals according to actual needs;

[0067] Step 2-2: Measure the signal strength of each access point (AP) at its current location multiple times at predefined fingerprint points. Then, use Gaussian filtering to remove values ​​with low probability of occurrence. Finally, take the average value as the fingerprint information and enter it into the fingerprint database. The Gaussian probability density function used is:

[0068]

[0069] In the formula, Let μ represent the signal strength value of the j-th AP collected at fingerprint point i. Each fingerprint point is measured n times. μ represents the mean of the signal strength values, and σ is the standard deviation of the signal strength values. The calculation formula is as follows:

[0070]

[0071]

[0072] In one embodiment, the fingerprint information screening step in step 3 is as follows:

[0073] Step 3-1: Use the UWB module to measure the distance d between the target user's current location and the UWB base station;

[0074] Step 3-2, select the option that satisfies |dd i Fingerprint points with |<Δ are considered as candidates, where d i This represents the distance from fingerprint point i to the base station, where Δ is the range threshold.

[0075] Step 3-3: Filter to satisfy α×d+s j ≤L j The AP is used to eliminate positioning errors introduced by unreliable APs, where the parameter α∈(0,1), s j L represents the effective radius of APj. j This represents the distance from AP j to the base station.

[0076] In one embodiment, step 4, the step of determining the target location, is as follows:

[0077] Step 4-1: Measure the signal strength of each AP selected in Step 3 at the target user's current location multiple times, and average the values ​​to reduce positioning errors caused by signal strength fluctuations, thus obtaining the fingerprint vector T = (RSS) for the current location. 1 RSS 2 ,...,RSS N),in n represents the number of measurements, and RSS is the RSS value. j The signal strength value of AP j;

[0078] Step 4-2: Calculate the fingerprint points selected in Step 3. The Euclidean distance D between T and i :

[0079]

[0080] In the formula, RSS represents the signal strength received at the j-th AP at the i-th fingerprint point. j The signal strength value of APj;

[0081] Calculate the coordinates (X) corresponding to the minimum Euclidean distance value. m ,Y m The mean of the coordinates corresponding to the k smallest Euclidean distances (X) k ,Y k The weighted average of the coordinates corresponding to the k smallest Euclidean distances (X) w ,Y w The formula is as follows:

[0082]

[0083]

[0084]

[0085] In the formula ω i ε represents the weight corresponding to the i-th fingerprint point. ε is a very small number, usually fixed at three orders of magnitude smaller than the test distance, to prevent the denominator from being zero.

[0086] Steps 4-3, calculate (X) respectively. m ,Y m ), (X k ,Y k ), (X w ,Y w The distance between the target location and the UWB base station is determined, and the location with the smallest difference from the UWB module's measured value d is selected as the current target location.

[0087] Compared to other methods, the indoor positioning method based on a hybrid Wi-Fi and ultra-wideband (UWB) technology proposed in this invention introduces a single UWB base station for distance measurement within Wi-Fi fingerprint positioning. Leveraging the advantages of UWB—strong penetration, good anti-interference performance, and high positioning accuracy—finds fingerprint information to eliminate positioning errors introduced by unreliable access points (APs), thereby improving the accuracy of the indoor positioning system while limiting equipment deployment costs. Furthermore, the method proposed in this invention can be combined with other indoor positioning technologies to further enhance performance.

[0088] This invention was deployed and its positioning effect tested in a real office setting: the experimental area measured 17.552m × 13.196m × 2.768m. A horizontal and vertical coordinate axis was set up in this area, with a fingerprint point placed at 2.5m intervals in each direction, for a total of 24 fingerprint points. UWB ranging employed the ULM1 tag / base station integrated module based on the DWM1000. This module is based on the TW-TOF ranging principle and achieves a ranging accuracy of 5cm. The Wi-Fi signal strength data acquisition device was a VIVO X6SA mobile phone. Six existing Wi-Fi routers detectable in the experimental area were used as access points (APs). Data was collected and recorded every second at each fingerprint point, with 100 data points collected from each AP.

[0089] from Figure 2 As can be seen from the results, the method of this patent is a significant improvement over other methods.

[0090] The above embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Although the invention has been described in detail with reference to the embodiments, those skilled in the art should understand that various combinations, modifications, or equivalent substitutions of the technical solutions of the invention do not depart from the spirit and scope of the invention and should be covered within the scope of the claims of the invention.

Claims

1. An indoor positioning method based on a hybrid of Wi-Fi and ultra-wideband technology, characterized in that, The method includes the following steps: Step 1: Infrastructure Deployment; Deploy a UWB positioning base station and multiple Wi-Fi signal access nodes (APs) indoors, and record the device location to compensate for UWB ranging errors; Step 2: Establish an offline fingerprint database; The signal strength values ​​of each AP are collected at the pre-set fingerprint points, and the processed signal strength values ​​and the location coordinates corresponding to the fingerprint points are entered into the fingerprint database. Step 3: Fingerprint information screening; Based on the distance between the target user and the base station obtained by UWB measurement, multiple fingerprint points and valid APs within the corresponding range are determined to reduce the solution space and eliminate the positioning error introduced by unreliable APs. Step 4: Target location determined; Using the fingerprint information obtained in step 3 and the current measured data, the current target location is estimated through various nearest neighbor algorithms, and the distance between the target and the UWB base station is calculated. The target location estimate with the smallest difference from the UWB distance measurement value is selected as the final positioning result. In step 4, the steps for determining the target location are as follows: Step 4-1: Measure the signal strength of each AP selected in Step 3 at the target user's current location multiple times, and average the values ​​to reduce positioning errors caused by signal strength fluctuations, thus obtaining the fingerprint vector of the current location. ,in n is the number of measurements. The signal strength value of AP j; Step 4-2: Calculate the fingerprint points selected in Step 3. Euclidean distance between T and T : （6） In the formula, This represents the signal strength received at the j-th AP at the i-th fingerprint point. The signal strength value of APj; Calculate the coordinates corresponding to the minimum Euclidean distance value. The mean of the coordinates corresponding to the k smallest Euclidean distances The weighted average of the coordinates corresponding to the k smallest Euclidean distances. The formula is as follows: （7） （8） （9） In the formula This represents the weight corresponding to the i-th fingerprint point. It is a very small number, usually fixed at three orders of magnitude smaller than the test distance, to prevent the denominator from being zero; Steps 4-3, calculate respectively , , The distance to the UWB base station is selected based on the smallest difference between the distance and the measured value d from the UWB module, which is taken as the current target location.

2. The indoor positioning method based on a hybrid of Wi-Fi and ultra-wideband technology as described in claim 1, characterized in that, In step 1, the UWB ranging error compensation steps are as follows: Step 1-1: Based on the relationship between the actual distance and the UWB measurement error, a univariate regression model is established, yielding the following relationship: （1） In the formula, Indicates the dependent variable. Indicates the independent variable. , Represents the regression coefficient. This represents the error term that follows a normal distribution. Steps 1-2: Obtain multiple sets of actual distance and UWB measurement error data, and calculate regression parameters: （2） In the formula, , They represent respectively to , The estimated value, , Let x and y represent the average values, respectively.

3. The indoor positioning method based on a hybrid of Wi-Fi and ultra-wideband technology as described in claim 1, characterized in that, In step 2, the steps for establishing the fingerprint database are as follows: Step 2-1: Set fingerprint points at certain intervals according to actual needs; Step 2-2: Measure the signal strength of each access point (AP) at its current location multiple times at predefined fingerprint points. Then, use Gaussian filtering to remove values ​​with low probability of occurrence. Finally, take the average value as the fingerprint information and enter it into the fingerprint database. The Gaussian probability density function used is: （3） In the formula, This represents the signal strength value of the t-th measurement from the j-th AP collected at fingerprint point i. Each fingerprint point is measured n times. This represents the mean of the signal strength values. The standard deviation of the signal strength values ​​is calculated using the following formula: （4） （5）。 4. The indoor positioning method based on a hybrid of Wi-Fi and ultra-wideband technology as described in claim 1, characterized in that, In step 3, the fingerprint information screening process is as follows: Step 3-1: Use the UWB module to measure the distance d between the target user's current location and the UWB base station; Step 3-2, select those that meet the requirements. Fingerprint points were selected as candidates, among which This represents the distance from fingerprint point i to the base station. For range thresholds; Step 3-3, filter to meet the requirements The AP is used to eliminate positioning errors introduced by unreliable APs, where parameters , Denotes the effective radius of APj. This represents the distance from AP j to the base station.

Images