CN111918228A - Wi-Fi indoor positioning method based on evidence synthesis rule optimization - Google Patents

Wi-Fi indoor positioning method based on evidence synthesis rule optimization Download PDF

Info

Publication number
CN111918228A
CN111918228A CN202010807850.8A CN202010807850A CN111918228A CN 111918228 A CN111918228 A CN 111918228A CN 202010807850 A CN202010807850 A CN 202010807850A CN 111918228 A CN111918228 A CN 111918228A
Authority
CN
China
Prior art keywords
rss
signal propagation
function
trust
evidence
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010807850.8A
Other languages
Chinese (zh)
Other versions
CN111918228B (en
Inventor
周牧
李欣玥
王勇
谢良波
聂伟
杨小龙
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University of Post and Telecommunications
Original Assignee
Chongqing University of Post and Telecommunications
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chongqing University of Post and Telecommunications filed Critical Chongqing University of Post and Telecommunications
Priority to CN202010807850.8A priority Critical patent/CN111918228B/en
Publication of CN111918228A publication Critical patent/CN111918228A/en
Application granted granted Critical
Publication of CN111918228B publication Critical patent/CN111918228B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/33—Services specially adapted for particular environments, situations or purposes for indoor environments, e.g. buildings
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)

Abstract

The invention relates to a Wi-Fi indoor positioning method based on evidence synthesis rule optimization, and belongs to the technical field of indoor positioning. Firstly, resume the relation state set of each reference point and the target position, and correct the mean value of the RSS sample by using a boundary error detection method; secondly, taking the normalized signal propagation distance distribution estimation as the basic probability assignment of a D-S evidence theory, and establishing the initial trust of the relation state of each reference point and the target position; based on a D-S evidence synthesis rule, multi-source RSS information is fused to obtain comprehensive trust estimation of each reference point, and meanwhile, a trust function based on a D-S evidence theory is used for selecting an ideal reference point; and (4) screening an ideal reference point with high trust degree according to a decision rule of the trust function as an ideal matching reference point, and estimating the target position by combining a centroid algorithm. The invention optimizes the precision of the positioning system on one hand and increases the stability and reliability of the positioning result on the other hand.

Description

Wi-Fi indoor positioning method based on evidence synthesis rule optimization
Technical Field
The invention belongs to the technical field of indoor positioning, and relates to a Wi-Fi indoor positioning method based on evidence synthesis rule optimization.
Background
In recent years, with the rapid increase of the demand of people for Location-based services (LBS), indoor positioning systems play an important role in a variety of application scenarios, such as pedestrian positioning and navigation in large-scale storage warehouses, underground shopping malls, garage navigation, and other scenarios. Because it is difficult to continuously and undulant collect satellite signals such as Global Positioning System (GPS) and big dipper in the indoor environment, the satellite Positioning System usually can not meet the Positioning performance requirement of the indoor location service. In addition, the common Infrared (IR), bluetooth, ZigBee [4], Ultra Wide Band (UWB) indoor positioning system, etc. usually require many additional infrastructure to be deployed, resulting in a limited application range. In contrast, Wi-Fi indoor positioning systems have received increasing attention due to their wide communication range, ease of deployment, and general lack of expensive hardware.
Wi-Fi indoor positioning methods include location fingerprinting and geometry measurement. The location fingerprinting method generally comprises an off-line stage and an on-line stage, wherein the off-line stage measures Received Signal Strength (RSS) from different Access points (Access points, APs) at a plurality of marked Reference points (Reference points, RPs) and establishes a location fingerprint database based on the RSS; and in the online stage, the newly measured received signal strength at the target is matched with the position fingerprint database to obtain the estimated position of the target. Geometric measurement methods mainly use signal characteristics such as angle of Arrival (AOA), Time of Arrival (TOA), Time Difference of Arrival (TDOA), RSS, etc. to estimate the geometric relative position between the AP and the target.
However, since the indoor scene is usually very complex, the indoor signal usually has diversity, which affects the effective estimation of the signal propagation distance by the positioning system, and thus the performance of the positioning system deteriorates and the robustness is poor. Aiming at the problem, the invention provides a Wi-Fi indoor positioning method based on multi-source RSS fusion of a D-S evidence theory, which solves the geometrical relationship between the signal propagation distances and RSSs of different APs according to a heuristic distribution model and completes the estimation of the distribution of the signal propagation distances by using a Gaussian kernel density estimation method; meanwhile, the normalized signal propagation distance distribution estimation is used as basic probability assignment of a DS evidence theory, and multi-source RSS information is fused through a D-S evidence synthesis rule; and finally, selecting an ideal matching reference point based on a decision rule of a trust function, and estimating the target position through a centroid algorithm. The method not only improves the precision of the positioning system, but also enhances the positioning robustness.
Disclosure of Invention
In view of the above, the present invention provides a Wi-Fi indoor positioning method based on evidence synthesis rule optimization. The fusion of multi-source RSS information is completed by using a D-S evidence synthesis rule, and an ideal matching reference point is screened out based on the fusion, so that the target position estimation is completed.
In order to achieve the purpose, the invention provides the following technical scheme:
the Wi-Fi indoor positioning method based on evidence synthesis rule optimization comprises the following steps:
step one, in an indoor scene, establishing signal propagation models d (v), d for the mean value, the maximum value and the minimum value of the signal propagation distanceu(v) And dl(v)；
Step two, order d1,…,dlRepresenting the i signal propagation distances randomly selected when the RSS is v, constructing a corresponding gaussian kernel density estimation function about the distribution of the signal propagation distances:
Figure BDA0002629823420000021
wherein K (·) represents a kernel function, and h represents a bandwidth;
step three, calculating the optimal bandwidth h of the Gaussian kernel density estimation functionMISE；
Step four, the formula in the step two
Figure BDA0002629823420000022
And (3) carrying out normalization processing to obtain normalized signal propagation distance density distribution estimation when the RSS is v:
Figure BDA0002629823420000023
step five, obtaining a trust function Bel of the target about the jth APj(I) And likelihood function Plj(I)；
Step six, deleting RSS singular samples by a boundary error detection method to correct the mean value of the RSS samples;
step seven, based on the formula in step four
Figure BDA0002629823420000024
The normalized signal propagation distance density distribution estimate shown, establishes piInitial trust m with target location about relationship status of jth APj(I),mj(N),mj(Θ)}：
Figure BDA0002629823420000025
Wherein the content of the first and second substances,
Figure BDA0002629823420000031
representing an RSS of
Figure BDA0002629823420000032
A time distance of dijNormalized density of dijDenotes the distance of the ith reference point from the jth AP, ajRepresenting a probability of a target position uncertainty state with respect to the jth AP;
step eight, fusing multi-source RSS information by using basic probability assignment values of m APs according to a D-S evidence synthesis rule to obtain comprehensive trust estimation of each RP;
the nine steps,Based on step eight, obtaining the target about piTwo deterministic state trust estimates m (i) and m (n), and one indeterminate state trust estimate m (Θ); at this time, when m (I) > m (N) and m (I) > m (Θ) are satisfied, p is definediIdeally matching the RP;
and step ten, acquiring an ideal matching RP set psi' and realizing estimation of the target position by combining a centroid algorithm.
Optionally, the step one specifically includes the following steps:
step one, based on three heuristic distribution models F1:
Figure BDA0002629823420000033
F2：d(v)＝b0+b1v+b2v2and F3:
Figure BDA0002629823420000034
fitting the mean value, the maximum value and the minimum value of the signal propagation distance changing along with the RSS by a least square method;
step one (two), based on the step one (one), establishing a mathematical relation between RSS and signal propagation distance from different APs by selecting an optimal heuristic distribution model with the minimum fitting error, and establishing a signal propagation model about the mean value, the maximum value and the minimum value of the signal propagation distance based on the mathematical relation
Figure BDA0002629823420000035
And
Figure BDA0002629823420000036
optionally, the step three specifically includes the following steps:
step three (one), constructing an average integral square error MISE function:
MISE(h)＝E{∫R[ph(x)-f(x)]2dx}
wherein the symbol "E" denotes the desired operation, ph(x) Representing a kernel density estimation function of the sample, f (x) representing a probability density function of the sample, R representing a spatial domain of the sample；
And step three (two), based on the step three (one), under the weak hypothesis condition, calculating to obtain:
Figure BDA0002629823420000037
step three, obtaining the optimal bandwidth h when the MISE (h) is taken as the minimum valueMISEMise (h) is given a partial derivative for h and made equal to 0, i.e.:
Figure BDA0002629823420000041
further obtaining:
Figure BDA0002629823420000042
under the condition of selecting the Epanechnikov kernel function, h isMISE≈2.345σl-0.2Where σ represents the standard deviation of the signal propagation distance.
Optionally, the step five specifically includes the following steps:
and step five (one), constructing a relation state set H of each RP and the target position, wherein p is { I, N, theta }, andi(I is 1, …, N), N is the number of RP, I and N respectively indicate that the target is located at p and not located at piAt least one of (1) and (b); theta is I or N and represents the uncertain state of the target position;
step five (two), based on the step five (one), establishing corresponding basic probability assignment m related to the jth APj：
Figure BDA0002629823420000043
Wherein φ represents an empty set; j is 1, …, mjIs H to [0,1]Mapping of (2);
step five (three), based on step five (two), defining the target about pjBelief function Bel ofj(I)And likelihood function Plj(I) Comprises the following steps:
Figure BDA0002629823420000044
optionally, the step six specifically includes the following steps:
step six (one), the ratio of the number of singular samples in the RSS data to the total number of samples is represented as a, and the probabilities of the target position uncertain state and the determined state are p (Θ) ═ a and p (i) + p (n) ═ 1-a, respectively;
step six (two), acquiring RSS sample set X from jth AP collected at targetj＝{xj,1,…,xj,sWhere s represents the total number of samples);
step six (step three), obtaining a signal propagation distance mean value fitting function d related to the jth APj(v)；
Step six (four), obtaining a fitting function d of the maximum value of the signal propagation distance of the jth APj,u(v)
Step six (five), obtaining a fitting function d related to the signal propagation distance minimum value of the jth APj,l(v)
Step six, setting the initial value of the RSS singular sample number as (0)
Step six (seven), calculating XjMean of medium RSS samples
Figure BDA0002629823420000051
Step six (eight), calculating
Figure BDA0002629823420000052
And
Figure BDA0002629823420000053
step six (nine), let k equal to 1, if
Figure BDA0002629823420000054
Or
Figure BDA0002629823420000055
Entering a sixth step (a tenth), and otherwise, entering a sixth step (a eleventh);
step six (ten), making ═ 1, wherein (═ 0) represents the initial value of the RSS singular sample number;
step six (eleven), mixing xj,kFrom XjRemoving to obtain a new RSS sample set X'j＝Xj-xj,k；
Step six (twelve), let k be k +1, judge whether k is greater than s, s is RSS sample total number, if yes, enter step six (twelve), if no, enter step six (nine);
step six (thirteen), calculating a ═ s, and further calculating X 'based thereon'jMean of medium RSS samples
Figure BDA0002629823420000056
And this is taken as a correction value for the mean of the RSS samples.
Optionally, the step eight specifically includes the following steps:
step eight (one), calculating mj(j＝1,…,m)；
Step eight (two), let m (i) ═ m1(I)、m(N)＝m1(N) and m (Θ) m1(Θ)；
Step eight (three), changing j to 2;
step eight (four), calculating
Figure BDA0002629823420000057
Step eight (five), calculate
Figure BDA0002629823420000058
Step eight (six), calculating
Figure BDA0002629823420000059
And step eight (seven), making j equal to j +1, and repeating the step eight (four), the step eight (five) and the step eight (six) until j is larger than m, wherein the { m (I), m (N), m (theta) } is the RP comprehensive trust estimation.
Optionally, the step ten specifically includes the following steps:
step ten (one), based on step nine, let Ψ ═ p1,…,peRepresents a set of e ideal RPs, whose corresponding trust functions are defined as
Figure BDA0002629823420000061
And eliminating the ideal RP from the psi until the rest ideal RP set psi ═ p1,…,pe′Trust function of } (e' ≦ e)
Figure BDA0002629823420000062
No longer satisfy
Figure BDA0002629823420000063
Wherein, a preset threshold value is shown;
and step ten (two), based on the step ten (one), defining psi' as an ideal matching RP set, namely the ideal RP set with high confidence, and taking the center of mass of all ideal matching RPs in the set as the estimation of the target position.
The invention has the beneficial effects that:
the method utilizes the advantages of the D-S evidence theory in the aspects of processing incomplete and inaccurate RSS information, and estimates the target position by constructing a trust function based on multi-source RSS information fusion. Firstly, resume the relation state set of each reference point and the target position, and correct the mean value of the RSS sample by using a boundary error detection method; secondly, taking the normalized signal propagation distance distribution estimation as the basic probability assignment of a D-S evidence theory, and establishing the initial trust of the relation state of each reference point and the target position; then, fusing multi-source RSS information to obtain comprehensive trust estimation of each reference point based on a D-S evidence synthesis rule, and meanwhile, selecting an ideal reference point by using a trust function based on a D-S evidence theory; and finally, screening an ideal reference point with high trust degree as an ideal matching reference point according to a decision rule of a trust function, and estimating the target position by combining a centroid algorithm. The invention optimizes the precision of the positioning system on one hand and increases the stability and reliability of the positioning result on the other hand.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
For the purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:
FIG. 1 is a flow chart of the present invention;
fig. 2 is a diagram of the average positioning error difference bitmap of the method of the present invention and other positioning methods when different AP numbers are selected.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.
Wherein the showings are for the purpose of illustrating the invention only and not for the purpose of limiting the same, and in which there is shown by way of illustration only and not in the drawings in which there is no intention to limit the invention thereto; to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The same or similar reference numerals in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if there is an orientation or positional relationship indicated by terms such as "upper", "lower", "left", "right", "front", "rear", etc., based on the orientation or positional relationship shown in the drawings, it is only for convenience of description and simplification of description, but it is not an indication or suggestion that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes, and are not to be construed as limiting the present invention, and the specific meaning of the terms may be understood by those skilled in the art according to specific situations.
Referring to fig. 1 to 2, in the first step, in the indoor scenario, signal propagation models d (v), d (m) are established with respect to the mean, maximum and minimum of the signal propagation distancesu(v) And dl(v) (ii) a The method specifically comprises the following steps:
step one, based on three heuristic distribution models F1:
Figure BDA0002629823420000071
F2：d(v)＝b0+b1v+b2v2and F3:
Figure BDA0002629823420000072
fitting the mean value, the maximum value and the minimum value of the signal propagation distance changing along with the RSS by a least square method;
step one (two), based on the step one (one), establishing a mathematical relation between RSS and signal propagation distance from different APs by selecting an optimal heuristic distribution model with the minimum fitting error, and establishing a signal propagation model about the mean value, the maximum value and the minimum value of the signal propagation distance based on the mathematical relation
Figure BDA0002629823420000073
And
Figure BDA0002629823420000074
step two, order d1,…,dlWhen the RSS is represented as vAnd (3) constructing a corresponding Gaussian kernel density estimation function about the signal propagation distance distribution by using the selected l signal propagation distances:
Figure BDA0002629823420000075
where K (·) represents a kernel function and h represents a bandwidth.
Step three, calculating the optimal bandwidth h of the Gaussian kernel density estimation functionMISE(ii) a The method specifically comprises the following steps:
step three (one), constructing an average Integrated Squared Error (MISE) function:
MISE(h)＝E{∫R[ph(x)-f(x)]2dx}
wherein the symbol "E" denotes the desired operation, ph(x) A kernel density estimation function representing the samples, f (x) a probability density function representing the samples, R representing the sample spatial domain;
step three (two), based on step three (one), under weak hypothesis condition, can calculate and obtain:
Figure BDA0002629823420000081
step three, obtaining the optimal bandwidth h when MISE (h) takes the minimum valueMISEMise (h) is given a partial derivative for h and made equal to 0, i.e.:
Figure BDA0002629823420000082
further obtaining:
Figure BDA0002629823420000083
under the condition of selecting the Epanechnikov kernel function, h isMISE≈2.345σl-0.2Where σ represents the standard deviation of the signal propagation distance。
Step four, the formula in the step two
Figure BDA0002629823420000084
And (3) carrying out normalization processing to obtain a normalized signal propagation distance density distribution estimation when the RSS is v:
Figure BDA0002629823420000085
step five, obtaining a trust function Bel of the target about the jth APj(I) And likelihood function Plj(I) (ii) a The method specifically comprises the following steps:
step five (one), construct each RP (i.e., p)i(I-1, …, N), where N is the number of RPs) and a set of relationship states for the target position H-I, N, Θ, where I and N represent the target position and non-position at p, respectivelyiWhere Θ (═ I or N) represents the uncertainty state of the target location;
step five (two), based on step five (one), establishing corresponding basic probability assignments m for j (j ═ 1, …, m) th APsj(i.e., H to [0,1 ]]Mapping of (d):
Figure BDA0002629823420000091
wherein φ represents an empty set;
step five (three), based on step five (two), defining the target about pjBelief function Bel ofj(I) And likelihood function Plj(I) Comprises the following steps:
Figure BDA0002629823420000092
step six, deleting RSS singular samples by a boundary error detection method to correct the mean value of the RSS samples; the method specifically comprises the following steps:
step six (one), the ratio of the number of singular samples in the RSS data to the total number of samples is represented as a, and the probabilities of the target position uncertain state and the determined state are p (Θ) ═ a and p (i) + p (n) ═ 1-a, respectively;
step six (two), acquiring RSS sample set X from jth AP collected at targetj＝{xj,1,…,xj,sWhere s represents the total number of samples);
step six (step three), obtaining a signal propagation distance mean value fitting function d related to the jth APj(v)；
Step six (four), obtaining a fitting function d of the maximum value of the signal propagation distance of the jth APj,u(v)
Step six (five), obtaining a fitting function d related to the signal propagation distance minimum value of the jth APj,l(v)
Step six, setting the initial value of the RSS singular sample number as (0)
Step six (seven), calculating XjMean of medium RSS samples
Figure BDA0002629823420000093
Step six (eight), calculating
Figure BDA0002629823420000094
And
Figure BDA0002629823420000095
step six (nine), let k equal to 1, if
Figure BDA0002629823420000096
Or
Figure BDA0002629823420000097
Entering a sixth step (a tenth), and otherwise, entering a sixth step (a eleventh);
step six (ten), making ═ 1, wherein (═ 0) represents the initial value of the RSS singular sample number;
step six (eleven), mixing xj,kFrom XjRemoving to obtain a new RSS sample set X'j＝Xj-xj,k；
Step six (twelve), let k be k +1, judge whether k is greater than s (RSS sample total number), if yes, enter step six (twelve), if no, enter step six (nine);
step six (thirteen), calculating a ═ s, and further calculating X 'based thereon'jMean of medium RSS samples
Figure BDA0002629823420000101
And this is taken as a correction value for the mean of the RSS samples.
Step seven, based on the formula in step four
Figure BDA0002629823420000102
The normalized signal propagation distance density distribution estimate shown, establishes piInitial trust m with target location about relationship status of jth APj(I),mj(N),mj(Θ)}：
Figure BDA0002629823420000103
Wherein the content of the first and second substances,
Figure BDA0002629823420000104
representing an RSS of
Figure BDA0002629823420000105
A time distance of dijNormalized density of dijDenotes the distance of the ith reference point from the jth AP, ajIndicating the probability of the target position uncertainty state with respect to the jth AP.
Step eight, fusing multi-source RSS information by using basic probability assignment values of m APs according to a D-S evidence synthesis rule to obtain comprehensive trust estimation of each RP; the method specifically comprises the following steps:
step eight (one), calculating mj(j＝1,…,m)；
Step eight (two), let m (i) ═ m1(I)、m(N)＝m1(N) and m (Θ) m1(Θ)；
Step eight (three), changing j to 2;
step eight (four), calculating
Figure BDA0002629823420000106
Step eight (five), calculate
Figure BDA0002629823420000107
Step eight (six), calculating
Figure BDA0002629823420000108
Step eight (seven), making j equal to j +1, and repeating the step eight (four), the step eight (five) and the step eight (six) until j is larger than m, wherein the { m (I), m (N), m (theta) } is the RP comprehensive trust estimation;
step nine, based on step eight, the target can be obtained about piTwo deterministic state trust estimates m (i) and m (n), and one indeterminate state trust estimate m (Θ); at this time, when m (I) > m (N) and m (I) > m (Θ) are satisfied, p is definediIdeally matching the RP;
step ten, acquiring an ideal matching RP set psi' and realizing estimation of the target position by combining a centroid algorithm; the method specifically comprises the following steps:
step ten (one), based on step nine, let Ψ ═ p1,…,peRepresents a set of e ideal RPs, whose corresponding trust functions are defined as
Figure BDA0002629823420000111
Rejecting as many ideal RPs as possible in Ψ until the remaining set of ideal RPs Ψ' ═ p1,…,pe′Trust function of } (e' ≦ e)
Figure BDA0002629823420000112
No longer satisfy
Figure BDA0002629823420000113
Wherein, a preset threshold value is shown;
and step ten (two), based on the step ten (one), defining psi' as an ideal matching RP set (namely, an ideal RP set with high confidence), and taking the center of mass of all ideal matching RPs in the set as the estimation of the target position.
Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.

Claims (7)

1. Wi-Fi indoor positioning method based on evidence synthesis rule optimization is characterized in that: the method comprises the following steps:
step one, in an indoor scene, establishing signal propagation models d (v), d for the mean value, the maximum value and the minimum value of the signal propagation distanceu(v) And dl(v)；
Step two, order d1,…,dlRepresenting the i signal propagation distances randomly selected when the RSS is v, constructing a corresponding gaussian kernel density estimation function about the distribution of the signal propagation distances:
Figure FDA0002629823410000011
wherein K (·) represents a kernel function, and h represents a bandwidth;
step three, calculating the optimal bandwidth h of the Gaussian kernel density estimation functionMISE；
Step four, the formula in the step two
Figure FDA0002629823410000012
And (3) carrying out normalization processing to obtain normalized signal propagation distance density distribution estimation when the RSS is v:
Figure FDA0002629823410000013
step five, obtaining a trust function Bel of the target about the jth APj(I) And likelihood function Plj(I)；
Step six, deleting RSS singular samples by a boundary error detection method to correct the mean value of the RSS samples;
step seven, based on the formula in step four
Figure FDA0002629823410000014
The normalized signal propagation distance density distribution estimate shown, establishes piInitial trust m with target location about relationship status of jth APj(I),mj(N),mj(Θ)}：
Figure FDA0002629823410000015
Wherein the content of the first and second substances,
Figure FDA0002629823410000016
representing an RSS of
Figure FDA0002629823410000017
A time distance of dijNormalized density of dijDenotes the distance of the ith reference point from the jth AP, ajRepresenting a probability of a target position uncertainty state with respect to the jth AP;
step eight, fusing multi-source RSS information by using basic probability assignment values of m APs according to a D-S evidence synthesis rule to obtain comprehensive trust estimation of each RP;
step nine, based on step eight, obtaining the target related to piTwo deterministic state trust estimates m (i) and m (n), and one indeterminate state trust estimate m (Θ); at this time, when m (I) > m (N) and m (I) > m (Θ) are satisfied, p is definediIdeally matching the RP;
and step ten, acquiring an ideal matching RP set psi' and realizing estimation of the target position by combining a centroid algorithm.
2. The Wi-Fi indoor positioning method based on evidence-based synthesis rule optimization of claim 1, wherein: the first step specifically comprises the following steps:
step one, based on three heuristic distribution models F1:
Figure FDA0002629823410000021
F2：d(v)＝b0+b1v+b2v2and F3:
Figure FDA0002629823410000022
fitting the mean value, the maximum value and the minimum value of the signal propagation distance changing along with the RSS by a least square method;
step one (two), based on the step one (one), establishing a mathematical relation between RSS and signal propagation distance from different APs by selecting an optimal heuristic distribution model with the minimum fitting error, and establishing a signal propagation model about the mean value, the maximum value and the minimum value of the signal propagation distance based on the mathematical relation
Figure FDA0002629823410000023
And
Figure FDA0002629823410000024
3. the Wi-Fi indoor positioning method based on evidence-based synthesis rule optimization of claim 1, wherein: the third step specifically comprises the following steps:
step three (one), constructing an average integral square error MISE function:
MISE(h)＝E{∫R[ph(x)-f(x)]2dx}
wherein the symbol "E" denotes the desired operation, ph(x) A kernel density estimation function representing the samples, f (x) a probability density function representing the samples, R representing the sample spatial domain;
and step three (two), based on the step three (one), under the weak hypothesis condition, calculating to obtain:
Figure FDA0002629823410000025
step three, obtaining the optimal bandwidth h when the MISE (h) is taken as the minimum valueMISEMise (h) is given a partial derivative for h and made equal to 0, i.e.:
Figure FDA0002629823410000026
further obtaining:
Figure FDA0002629823410000031
under the condition of selecting the Epanechnikov kernel function, h isMISE≈2.345σl-0.2Where σ represents the standard deviation of the signal propagation distance.
4. The Wi-Fi indoor positioning method based on evidence-based synthesis rule optimization of claim 1, wherein: the fifth step specifically comprises the following steps:
step five (one), constructing a relation state set H of each RP and the target position, wherein the relation state set H is { I, N, theta }; wherein RP is piI is 1, …, N is the number of RP, I and N respectively indicate that the target is located at p and not located at piAt least one of (1) and (b); theta is I or N and represents the uncertain state of the target position;
step five (two), based on the step five (one), establishing corresponding basic probability assignment m related to the jth APj：
Figure FDA0002629823410000032
Wherein φ represents an empty set; j is 1, …, mjIs H to [0,1]Mapping of (2);
step five (three), based on step five (two), defining the target about pjBelief function Bel ofj(I) And likelihood function Plj(I) Comprises the following steps:
Figure FDA0002629823410000033
5. the Wi-Fi indoor positioning method based on evidence-based synthesis rule optimization of claim 1, wherein: the sixth step specifically comprises the following steps:
step six (one), the ratio of the number of singular samples in the RSS data to the total number of samples is represented as a, and the probabilities of the target position uncertain state and the determined state are p (Θ) ═ a and p (i) + p (n) ═ 1-a, respectively;
step six (two), acquiring RSS sample set X from jth AP collected at targetj＝{xj,1,…,xj,sWhere s represents the total number of samples);
step six (step three), obtaining a signal propagation distance mean value fitting function d related to the jth APj(v)；
Step six (four), obtaining a fitting function d of the maximum value of the signal propagation distance of the jth APj,u(v)
Step six (five), obtaining a fitting function d related to the signal propagation distance minimum value of the jth APj,l(v)
Step six, setting the initial value of the RSS singular sample number as (0)
Step six (seven), calculating XjMean of medium RSS samples
Figure FDA0002629823410000041
Step six (eight), calculating
Figure FDA0002629823410000042
And
Figure FDA0002629823410000043
step six (nine), let k equal to 1, if
Figure FDA0002629823410000044
Or
Figure FDA0002629823410000045
Entering a sixth step (a tenth), and otherwise, entering a sixth step (a eleventh);
step six (ten), making ═ 1, wherein (═ 0) represents the initial value of the RSS singular sample number;
step six (eleven), mixing xj,kFrom XjRemoving to obtain new RSS sample set Xj′＝Xj-xj,k；
Step six (twelve), let k be k +1, judge whether k is greater than s, s is RSS sample total number, if yes, enter step six (twelve), if no, enter step six (nine);
step six (thirteen), calculating a ═ s, and further calculating X 'based thereon'jMean of medium RSS samples
Figure FDA0002629823410000046
And this is taken as a correction value for the mean of the RSS samples.
6. The Wi-Fi indoor positioning method based on evidence-based synthesis rule optimization of claim 1, wherein: the eighth step specifically comprises the following steps:
step eight (one), calculating mj(j＝1,…,m)；
Step eight (two), let m (i) ═ m1(I)、m(N)＝m1(N) and m (Θ) m1(Θ)；
Step eight (three), changing j to 2;
step eight (four), calculating
Figure FDA0002629823410000047
Step eight (five), calculate
Figure FDA0002629823410000048
Step eight (six), calculating
Figure FDA0002629823410000049
And step eight (seven), making j equal to j +1, and repeating the step eight (four), the step eight (five) and the step eight (six) until j is larger than m, wherein the { m (I), m (N), m (theta) } is the RP comprehensive trust estimation.
7. The Wi-Fi indoor positioning method based on evidence-based synthesis rule optimization of claim 1, wherein: the step ten specifically comprises the following steps:
step ten (one), based on step nine, let Ψ ═ p1,…,peRepresents a set of e ideal RPs, whose corresponding trust functions are defined as
Figure FDA0002629823410000051
And eliminating the ideal RP from the psi until the rest ideal RP set psi ═ p1,…,pe′Trust function of } (e' ≦ e)
Figure FDA0002629823410000052
No longer satisfy
Figure FDA0002629823410000053
Wherein, a preset threshold value is shown;
and step ten (two), based on the step ten (one), defining psi' as an ideal matching RP set, namely the ideal RP set with high confidence, and taking the center of mass of all ideal matching RPs in the set as the estimation of the target position.
CN202010807850.8A 2020-08-12 2020-08-12 Wi-Fi indoor positioning method based on evidence synthesis rule optimization Active CN111918228B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010807850.8A CN111918228B (en) 2020-08-12 2020-08-12 Wi-Fi indoor positioning method based on evidence synthesis rule optimization

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010807850.8A CN111918228B (en) 2020-08-12 2020-08-12 Wi-Fi indoor positioning method based on evidence synthesis rule optimization

Publications (2)

Publication Number Publication Date
CN111918228A true CN111918228A (en) 2020-11-10
CN111918228B CN111918228B (en) 2022-04-29

Family

ID=73284356

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010807850.8A Active CN111918228B (en) 2020-08-12 2020-08-12 Wi-Fi indoor positioning method based on evidence synthesis rule optimization

Country Status (1)

Country Link
CN (1) CN111918228B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101778399A (en) * 2010-01-18 2010-07-14 哈尔滨工业大学 Method for optimizing WLAN (Wireless Local Area Network) indoor ANN (Artificial Neural Network) positioning based on FCM (fuzzy C-mean) and least-squares curve surface fitting methods
CN105872972A (en) * 2016-04-29 2016-08-17 武汉大学 Self-adaptation AP selection method based on multi-target optimization
CN108449708A (en) * 2018-03-15 2018-08-24 重庆邮电大学 A kind of Wi-Fi indoor orientation methods examined based on signal distributions

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101778399A (en) * 2010-01-18 2010-07-14 哈尔滨工业大学 Method for optimizing WLAN (Wireless Local Area Network) indoor ANN (Artificial Neural Network) positioning based on FCM (fuzzy C-mean) and least-squares curve surface fitting methods
CN105872972A (en) * 2016-04-29 2016-08-17 武汉大学 Self-adaptation AP selection method based on multi-target optimization
CN108449708A (en) * 2018-03-15 2018-08-24 重庆邮电大学 A kind of Wi-Fi indoor orientation methods examined based on signal distributions

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
张明华: "基于WLAN的室内定位技术研究", 《上海交通大学》 *
杨帅: "基于证据K近邻与网格聚类的WiFi定位算法设计与实现", 《华南理工大学》 *

Also Published As

Publication number Publication date
CN111918228B (en) 2022-04-29

Similar Documents

Publication Publication Date Title
CN105424030B (en) Fusion navigation device and method based on wireless fingerprint and MEMS sensor
CN101526601B (en) Self-adaptive localization method, equipment and system adopting TOA and RSS fusion mode
EP2270535B1 (en) Indoor/outdoor decision apparatus and indoor/outdoor decision method
US9763034B2 (en) Channel-type supervised node positioning method for a wireless network
US6456956B1 (en) Algorithm for selectively suppressing NLOS signals in location estimation
CN112533163B (en) Indoor positioning method based on NB-IoT (NB-IoT) improved fusion ultra-wideband and Bluetooth
CN109951798A (en) Merge the enhancing location fingerprint indoor orientation method of Wi-Fi and bluetooth
US20100309059A1 (en) Method and Apparatus of Positioning for a Wireless Communication System
JP2004215258A (en) Positioning method of radio communication device using selective fusion
CN110933599A (en) Self-adaptive positioning method fusing UWB and WIFI fingerprints
CN113031031B (en) GNSS signal accurate classification-based weighted positioning method in urban canyon
KR20110116565A (en) Method determining indoor location using bayesian algorithm
WO2012016355A1 (en) Method of and system for locating the position of user equipment
Wigren LTE fingerprinting localization with altitude
Mazan et al. A Study of Devising Neural Network Based Indoor Localization Using Beacons: First Results.
CN112533144B (en) Indoor positioning method, device, computing equipment and computer storage medium
Sinha et al. Range based improved localization scheme in densely populated wireless sensor network
CN111918228B (en) Wi-Fi indoor positioning method based on evidence synthesis rule optimization
Jung et al. Future Parking Applications: Wireless Sensor Network Positioning for Highly Automated in-House Parking.
Leitinger et al. Data fusion for radio frequency SLAM with robust sampling
Afousi et al. Wi-Fi RSS indoor positioning system using online layer clustering and weighted DCP-KNN
CN113810846B (en) Indoor positioning method based on WiFi and IMU fusion
CN101924986A (en) Positioning method for wireless communication system and related device thereof
Mikhalev et al. Passive emitter geolocation using agent-based data fusion of AOA, TDOA and FDOA measurements
Ramaneti et al. IoT based 2D indoor navigation system using BLE beacons and Dijkstra's algorithm

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant