A positioning switching method based on accuracy estimation

Abstract

The present invention provides a positioning switching method based on accuracy estimation, which belongs to the field of navigation positioning technology. The positioning switching method includes three parts: soft and hard switching positioning, selection of decision threshold and switching decision. Compared with hard switching positioning and soft switching positioning, the present invention has the advantage that the positioning mode can be switched according to different situations under each positioning, and the better result of the soft / hard switching positioning result can be output. It not only inherits the seamlessness of soft and hard switching positioning, but also improves the positioning accuracy.

Description

Technical Field

[0001] The invention belongs to the technical field of navigation and positioning, and relates to hard switching positioning and soft switching positioning, and in particular to a positioning mode switching and accuracy estimation method. Background Art

[0002] In heterogeneous positioning networks, multiple positioning networks can be integrated by using hard handover positioning and soft handover positioning to expand coverage and improve positioning accuracy. However, in each positioning, the integrated positioning method with better positioning results is not fixed. Sometimes soft handover positioning accuracy is better, and sometimes hard handover positioning has better positioning performance. During the positioning process, the device cannot determine whether the hard handover positioning or soft handover positioning result is better in this positioning, so it is necessary to switch the positioning fusion method to a positioning method with better performance.

[0003] After searching the existing literature, it was found that in the paper titled "Research on Indoor and Outdoor Seamless Positioning Technology of Heterogeneous Wireless Systems" (hereinafter referred to as Document 1), the concepts of hard switching positioning and soft switching positioning were analyzed. Soft / hard switching positioning is applicable to various types of heterogeneous positioning systems, which can improve the accuracy of positioning, meet the needs of equipment for seamless positioning, and have better universality. However, it only discusses the principles of hard switching positioning and soft switching positioning respectively, without considering the switching problem between soft switching positioning and hard switching positioning.

[0004] The paper "Fusion Localization Based on Accuracy Estimation" published at the International Conference on Advanced Electronic Materials, Computers and Software Engineering (AEMCSE) proposed an accuracy estimation algorithm for positioning, which takes advantage of the monotonic characteristics of the "set of arrival time differences between the terminal and the base station" and the "positioning accuracy" within a certain range. After further research, it was found that the accuracy estimation algorithm proposed in the paper can provide a basis for positioning switching.

[0005] Based on the above, the present invention further studies the positioning under heterogeneous systems, selects soft / hard switching as the fusion mode of this positioning according to the accuracy estimation algorithm mentioned above, and thus proposes a switching method for the fusion of soft switching and hard switching. Summary of the invention

[0006] The technical problem to be solved by the present invention is that in a heterogeneous system, the device cannot determine whether to select hard switching positioning or soft switching positioning, and may select the worse positioning result of the two, resulting in a large positioning error. The purpose of the present invention is to enable the device to select the better positioning result of soft / hard switching positioning in each positioning process, thereby improving the accuracy of the positioning result.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is:

[0008] A positioning switching method based on accuracy estimation includes three parts: soft and hard switching positioning, selection of decision threshold and switching decision. The specific steps are as follows:

[0009] The first step is to use the estimated accuracy value to perform soft / hard switching positioning

[0010] Step 1.1: Assume that there are two heterogeneous positioning systems A and B in the positioning environment, and the test point TP obtains the positioning result P under system A at the same time A And the accuracy estimation result E A , the positioning result P under system B B And the accuracy estimation result E B .

[0011] Step 1.2: The two accuracy estimates E obtained at the test point TP A and E B Compare and select the positioning result of the system with the smaller value as the hard switching positioning result P H .

[0012] Step 1.3: Take the precision estimate E A , E B Substitute the weight calculation formula in Reference 1 Get the weight w of system A and the weight 1-w of system B, and bring them into the fusion positioning calculation formula P S =wP A +(1-w)P B Get the soft handover positioning result P S .

[0013] The second step is to select the appropriate decision threshold:

[0014] Step 2.1: Bring the experimental scene into the actual environment and adjust the position (x TP ,y TP ) The known test point TP' is positioned by hard handover and soft handover through steps 1.1 to 1.3. In the actual environment, there are two heterogeneous positioning systems, and the accuracy estimate E can be obtained. A 、E B , and the hard switching positioning result (x H ,y H) and soft handover positioning results (x S ,y S ), and then calculate the positioning error, that is, the distance between the actual position of the test point and the positioning result, and get the hard switching positioning error as The error of soft switching positioning is

[0015] Step 2.2: In the same environment, repeat step 2.1 multiple times with the test point TP' at different locations. Each repetition can obtain a set of E A 、E B , R H , R S Therefore, multiple groups of samples can be obtained, and different positioning accuracy estimation values ​​E can be counted based on these samples. A 、E B and positioning error R H , R S The corresponding relationship.

[0016] Step 2.3: Estimate the precision E in each group A 、E B Take the smaller value between the two and compare it with the larger value to get the accuracy estimation ratio E AB .

[0017] Step 2.4: In multiple groups of samples, each sample point has a corresponding accuracy estimation ratio E AB , hard switching positioning error R H and soft switching positioning error R S , statistics of each sample point in the hard switching positioning error R H The corresponding accuracy estimation ratio E is small. AB , and the soft switching positioning error R S The corresponding accuracy estimation ratio E is small. AB , analyze the two cases E AB From the histogram trends in the two cases, we can get the curve equation F with the horizontal axis being the precision estimation ratio and the vertical axis being the number of histogram samples. H (e) and F S (e) Select the value corresponding to the horizontal axis e when the two curves intersect as the switching decision threshold D T .

[0018] The third step is to perform switching positioning based on the decision threshold:

[0019] Step 3.1: In the real environment corresponding to the second step, perform hard handover positioning and soft handover positioning on the test point TP' through steps 1.1 to 1.3 to obtain the accuracy estimate E A 、E B , precision estimation ratio EAB , and the hard switching positioning result (x H ,y H ) and soft handover positioning results (x S ,y S ).

[0020] Step 3.2: Get E from this positioning AB and decision threshold D T For comparison, when E AB Less than D T Then the hard switching positioning result (x H ,y H ) as the positioning result, otherwise the soft switching positioning result (x S ,y S ) as the positioning result.

[0021] The effects and benefits of the present invention are as follows: compared with hard switching positioning and soft switching positioning, the present invention has the advantage of being able to switch positioning modes according to different situations in each positioning, and outputting the better result of soft / hard switching positioning results. It not only inherits the seamlessness of soft and hard switching positioning, but also improves the positioning accuracy. DETAILED DESCRIPTION

[0022] The first step is to use the estimated accuracy value to perform soft / hard switching positioning

[0023] Step 1.1: Using two heterogeneous positioning systems A and B in the environment, test point TP simultaneously obtains the positioning result P under system A A And the accuracy estimation result E A , the positioning result P under system B B And the accuracy estimation result E B .

[0024] Step 1.2: The two accuracy estimates E obtained at the test point TP A and E B Compare and select the positioning result of the system with the smaller value as the hard switching positioning result P H .

[0025] Step 1.3: Estimate E from the precision A , E B Get the weight w of system A and the weight 1-w of system B, and bring them into the fusion positioning calculation formula P S =wP A +(1-w)P B Get the soft handover positioning result P S .

[0026] The second step is to select the appropriate decision threshold:

[0027] Step 2.1: In a real scene with a length and width of 100 meters, there are two heterogeneous positioning systems A and B, corresponding to 5G OTDOA positioning and WLAN fingerprint positioning respectively. TP ,y TP ) The known test point TP' is positioned by hard handover and soft handover through steps 1.1 to 1.3 to obtain the accuracy estimate E A 、E B , hard switching positioning result (x H ,y H ) and soft handover positioning results (x S ,y S ), the hard switching positioning error is calculated as The error of soft switching positioning is

[0028] Step 2.2: In the same environment, repeat step 2.1 100,000 times with the test point TP' at different locations. Each repetition will produce a set of A 、E B , R H , R S Therefore, multiple groups of samples can be obtained, and different positioning accuracy estimation values ​​E can be counted based on these samples. A 、E B and positioning error R H , R S The corresponding relationship.

[0029] Step 2.3: Estimate the precision E in each group A 、E B Take the smaller value between the two and compare it with the larger value to get the accuracy estimation ratio E AB .

[0030] Step 2.4: Count multiple groups of samples, and each sample point has an accuracy estimation ratio E AB , soft switching positioning error R S and hard switching positioning error R H These three parameters are used to calculate the hard switching positioning error R of each sample point. H The corresponding accuracy estimation ratio E is small. AB , and the soft switching positioning error R S The corresponding accuracy estimation ratio E is small. AB , analyze the two cases E AB According to the statistical results, select E AB =0.155 as the decision threshold for switching T .

[0031] The third step is to perform switching positioning based on the decision threshold:

[0032] Step 3.1: Perform hard handover positioning and soft handover positioning on the test point TP' through steps 1.1 to 1.3 to obtain the accuracy estimate E A =13, E B =20, precision estimation ratio E AB =0.65, as well as the hard switching positioning result (60,80) and the soft switching positioning result (60.8,81).

[0033] Step 3.2: Compare the estimated accuracy ratio of 0.65 obtained from this positioning with the decision threshold of 0.155, E AB Greater than D T , the soft switching positioning result (60.8,81) is used as the positioning result.

[0034] The above-described embodiments merely express the implementation methods of the present invention, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A positioning switching method based on accuracy estimation, characterized in that: The positioning switching method includes three parts: soft and hard switching positioning, selection of decision threshold and switching decision. The specific steps are as follows: The first step is to use the estimated accuracy value to perform soft / hard switching positioning Step 1.1: Assume that there are two heterogeneous positioning systems A and B in the positioning environment, and the test point TP obtains the positioning result P under system A at the same time A And the accuracy estimation result E A , the positioning result P under system B B And the accuracy estimation result E B ; Step 1.2: The two accuracy estimates E obtained at the test point TP A and E B Compare and select the positioning result of the system with the smaller value as the hard switching positioning result P H ; Step 1.3: Take the precision estimate E A , E B Substitute into the weight calculation formula Get the weight w of system A and the weight 1-w of system B, and bring them into the fusion positioning calculation formula P S =wP A +(1-w)P B Get the soft handover positioning result P S ; The second step is to select the appropriate decision threshold: Step 2.1: Bring the experimental scene into the actual environment and adjust the position (x TP ,y TP ) The known test point TP' is positioned by hard handover and soft handover through steps 1.1 to 1.

3. In the actual environment, there are two heterogeneous positioning systems, and the accuracy estimate E can be obtained. A 、E B , and the hard switching positioning result (x H ,y H ) and soft handover positioning results (x S ,y S ), and then calculate the positioning error, that is, the distance between the actual position of the test point and the positioning result, and get the hard switching positioning error as The error of soft switching positioning is Step 2.2: In the same environment, repeat step 2.1 multiple times with the test point TP' at different locations. Each repetition can obtain a set of E A 、E B , R H , R S Therefore, multiple groups of samples can be obtained, and different positioning accuracy estimation values ​​E can be statistically calculated using multiple groups of samples. A 、E B and positioning error R H , R S The corresponding relationship; Step 2.3: Estimate the precision E in each group A 、E B Make a ratio and get the accuracy estimate ratio E AB ; Step 2.4: In multiple groups of samples, each sample point has a corresponding accuracy estimation ratio E AB , hard switching positioning error R H and soft switching positioning error R S , statistics of each sample point in the hard switching positioning error R H The corresponding accuracy estimation ratio E is small. AB , and the soft switching positioning error R S The corresponding accuracy estimation ratio E is small. AB , analyze the two cases E AB The distribution of the histogram; from the histogram trends in the two cases, the curve equation F with the horizontal axis as the precision estimation ratio and the vertical axis as the number of histogram samples is obtained. H (e) and F S (e); Select the value corresponding to the horizontal axis e when the two curves intersect as the switching decision threshold D T ; The third step is to perform switching positioning based on the decision threshold: Step 3.1: In the real environment corresponding to the second step, perform hard handover positioning and soft handover positioning on the test point TP' through steps 1.1 to 1.3 to obtain the accuracy estimate E A 、E B , precision estimation ratio E AB , and the hard switching positioning result (x H ,y H ) and soft handover positioning results (x S ,y S ); Step 3.2: Get E from this positioning AB and decision threshold D T For comparison, when E AB Less than D T Then the hard switching positioning result (x H ,y H ) as the positioning result, otherwise the soft switching positioning result (x S ,y S ) as the positioning result.

Images