阅读说明：本技术 一种基于rssi概率分布的加权定位方法 (RSSI probability distribution-based weighted positioning method ) 是由 聂大惟 朱海 吴飞 于 2021-10-29 设计创作，主要内容包括：本发明涉及一种基于RSSI概率分布的加权定位方法,该方法包括以下步骤：将定位空间简化为多个离散空间,获取待定位终端采集的RSSI数据；基于离线获取的距离与RSSI概率分布之间的映射关系以及定位空间中RSSI概率分布,采用贝叶斯公式估计各个离散空间的权重,以权重最大的离散空间的位置作为定位结果。与现有技术相比,本发明具有定位精度高、鲁棒性佳等优点。(The invention relates to a weighted positioning method based on RSSI probability distribution, which comprises the following steps: simplifying a positioning space into a plurality of discrete spaces, and acquiring RSSI data acquired by a terminal to be positioned; based on the mapping relation between the distance and the RSSI probability distribution obtained off-line and the RSSI probability distribution in the positioning space, the weight of each discrete space is estimated by adopting a Bayesian formula, and the position of the discrete space with the maximum weight is used as the positioning result. Compared with the prior art, the method has the advantages of high positioning precision, good robustness and the like.)

1. A weighted positioning method based on RSSI probability distribution is characterized by comprising the following steps:

simplifying a positioning space into a plurality of discrete spaces, and acquiring RSSI data acquired by a terminal to be positioned;

based on the mapping relation between the distance and the RSSI probability distribution obtained off-line and the RSSI probability distribution in the positioning space, the weight of each discrete space is estimated by adopting a Bayesian formula, and the position of the discrete space with the maximum weight is used as the positioning result.

2. The RSSI probability distribution-based weighted positioning method of claim 1 wherein the positioning space is reduced to a plurality of discrete spaces having a side length equal to Δ d.

3. The RSSI probability distribution-based weighted positioning method of claim 1, wherein the offline acquisition process of the mapping relationship between the distance and the RSSI probability distribution comprises:

arranging a plurality of APs in a positioning space;

and calculating the distance L from each discrete space to each AP, collecting the RSSI number by taking each discrete space as a data acquisition point, and establishing a mapping relation L-P (RSSI | L) between the distance and the RSSI probability distribution, wherein P (RSSI | L) represents the RSSI probability distribution when the distance L from the AP is obtained by Gaussian fitting.

4. The RSSI probability distribution based weighted positioning method of claim 3, wherein the same number of RSSIs is collected at each of the data acquisition points.

5. The RSSI probability distribution-based weighted positioning method of claim 3 wherein statistical analysis is performed on all collected RSSI quantities to obtain the RSSI probability distribution in the positioning space.

6. The RSSI probability distribution-based weighted positioning method of claim 3, wherein the distance L from each discrete space to each AP is represented as

Wherein x and y are horizontal and vertical coordinates of a central point of the discrete space, and xc and yc represent horizontal and vertical coordinates of the AP.

7. The RSSI probability distribution-based weighted positioning method of claim 1, wherein the estimating the weight of each discrete space using a Bayesian formula is specifically represented as:

wherein, W (x, y) is the final weight of the discrete space, N is the number of AP of RSSI received in the discrete space, and P (L)_i∣RSSI_i) The representation is based on Bayesian formulaAnd solving the obtained probability, wherein P (L | RSSI) represents the probability that the distance between the positioning terminal and the AP is L when the RSSI is known, and P (RSSI | L) represents the probability distribution of the RSSI when the distance between the positioning terminal and the AP is LP (l) is a distance distribution, and p (RSSI) represents the probability distribution of RSSI in the positioning space.

8. The RSSI probability distribution-based weighted positioning method of claim 7, wherein the distance distribution is an equal distribution.

9. The RSSI probability distribution-based weighted positioning method of claim 1, wherein a particle swarm algorithm is used to solve the discrete space position with the largest weight.

10. The RSSI probability distribution-based weighted positioning method of claim 1, wherein when obtaining the mapping relationship between distance and RSSI probability distribution, a cubic spline interpolation is used to obtain the mapping that cannot be directly obtained.

Technical Field

The invention relates to the technical field of location services, in particular to a weighted positioning method based on RSSI probability distribution.

Background

With the advent of the big data age, various data play more and more important roles in production and life, and the demand of Location Based Services (LBS) in various fields is increasing. A Global Navigation Satellite System (GNSS) can provide accurate positioning and navigation services outdoors, and is widely applied to various fields such as industry, agriculture, military industry, and commerce. However, in an indoor environment, GNSS signals are shielded by buildings, so that the positioning device cannot perform positioning due to poor quality of the received signals or the inability to receive the signals, and thus the indoor positioning technology gradually becomes a research hotspot in academic circles and industrial circles. Currently, indoor positioning technologies based on wireless signals can be divided into a UWB positioning technology, a bluetooth positioning technology, a ZigBee positioning technology, a RFID positioning technology, a Wi-Fi positioning technology, and the like. The Wi-Fi positioning technology has the advantages of convenience in deployment, low cost, high positioning precision and the like, and meanwhile, the mobile terminal is internally provided with the Wi-Fi receiving chip, so that the technology is convenient to popularize. Common indoor positioning methods include those based on time difference of arrival (TDOA), time of arrival (TOA), direction of arrival (DOA), and RSSI (Received Signal Strength Indication), among others. The indoor positioning technology of Wi-Fi signals based on RSSI does not require special hardware equipment, has good applicability and higher positioning accuracy, and is widely concerned by researchers. The technology can be mainly generalized to Wi-Fi positioning based on RSSI position fingerprints and Wi-Fi positioning technology based on RSSI ranging.

As shown in fig. 1, a fingerprint-based positioning (fingerprint-based localization) method relies on differences of RSSI at different positions to establish a unique mapping relationship from the RSSI to a spatial position, and performs positioning through a matching algorithm. When the fingerprint database is established, a large amount of RSSI data needs to be collected, and the fingerprint database needs to be updated continuously in the later period, so that the workload is huge. Because the fingerprint database is very closely combined with the environment, the information acquired by the acquisition work is difficult to migrate to the similar environment, and a large amount of repeated operation is caused.

The positioning method based on RSSI ranging measures the distance from a terminal to be positioned to a wireless Access Point (AP) according to the characteristic that RSSI attenuates along with the distance, and has the advantages of strong practicability, strong scalability, simple deployment and maintenance and the like. The current RSSI ranging and positioning technology is mainly based on a path loss model. With the AP as the signal source, the strength of the Wi-Fi signal emitted by the AP decreases as the reception distance increases. According to the attenuation characteristic, the distance between the terminal to be positioned and the AP can be calculated from the RSSI. Currently, most Wi-Fi positioning methods based on ranging are established on the basis of a logarithmic path loss model, and positioning is realized according to the logarithmic relationship between RSSI and the distance from a terminal to an AP. However, Wi-Fi signals are susceptible to interference from external uncertain factors such as noise and multipath effects, so that the signal strength indication received by the mobile terminal has volatility, thereby affecting the accuracy of the existing RSSI ranging and positioning technology based on a path loss model.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provide a weighted positioning method based on RSSI probability distribution, which has high positioning accuracy and good robustness.

Although the Wi-Fi based indoor positioning method has a problem that the positioning accuracy is affected by the RSSI noise, the inventors found that the RSSI at one point in space as a whole follows a gaussian distribution and is temporally stationary. Therefore, the purpose of the invention can be realized by the following technical scheme:

a weighted positioning method based on RSSI probability distribution comprises the following steps:

simplifying a positioning space into a plurality of discrete spaces, and acquiring RSSI data acquired by a terminal to be positioned;

based on the mapping relation between the distance and the RSSI probability distribution obtained off-line and the RSSI probability distribution in the positioning space, the weight of each discrete space is estimated by adopting a Bayesian formula, and the position of the discrete space with the maximum weight is used as the positioning result.

And simplifying the positioning space into a plurality of discrete spaces with side lengths equal to delta d.

Further, the offline acquisition process of the mapping relationship between the distance and the RSSI probability distribution includes:

arranging a plurality of APs in a positioning space;

and calculating the distance L from each discrete space to each AP, collecting the RSSI number by taking each discrete space as a data acquisition point, and establishing a mapping relation L-P (RSSI | L) between the distance and the RSSI probability distribution, wherein P (RSSI | L) represents the RSSI probability distribution when the distance L from the AP is obtained by Gaussian fitting.

Further, the same number of RSSIs are collected at each of the data acquisition points.

Further, statistical analysis is performed on all the collected RSSI quantities to obtain RSSI probability distribution in the positioning space.

Further, the distance L from each discrete space to each AP is expressed as

Wherein x and y are horizontal and vertical coordinates of a central point of the discrete space, and xc and yc represent horizontal and vertical coordinates of the AP.

Further, the estimating the weight of each discrete space by using the bayesian formula is specifically expressed as:

wherein, W (x, y) is the final weight of the discrete space, N is the number of AP of RSSI received in the discrete space, and P (L)_i∣RSSI_i) The representation is based on Bayesian formulaAnd solving the obtained probability, wherein P (L | RSSI) represents the probability that the distance between the positioning terminal and the AP is L when the RSSI is known, P (RSSI | L) represents the probability distribution of the RSSI when the distance between the positioning terminal and the AP is L, P (RSSI | L) represents the distance distribution, and P (RSSI) represents the probability distribution of the RSSI in the positioning space.

Further, the distance distribution is equal.

Further, the position of the discrete space with the maximum weight is obtained by solving through a particle swarm algorithm.

Further, when the mapping relation between the distance and the RSSI probability distribution is obtained, the mapping which cannot be directly obtained is obtained by adopting a cubic spline interpolation method.

Compared with the prior art, the invention has the following beneficial effects:

1. on the basis of researching the stability and distribution characteristics of the RSSI, the invention introduces the prior RSSI probability distribution into the calculation of the weight, gives a lower weight to an abnormal value, reduces the influence of environmental noise and external uncertain factors on the positioning precision, and effectively improves the positioning precision.

2. The RSSI probability distribution-based weighted positioning method obtains RSSI distribution at different positions away from an AP by utilizing probability statistics, further approximates the RSSI distribution to be Gaussian distribution, calculates corresponding weight by using the Gaussian distribution characteristics of each discrete space, obtains the position with the maximum weight by a particle swarm algorithm and uses the position as a positioning result. The positioning effect is better than that of the traditional method, and meanwhile, the robustness is better.

Drawings

FIG. 1 is a flow chart of a prior art fingerprint location;

FIG. 2 is a flow chart of the method of the present invention;

FIG. 3 is a diagram illustrating a mapping relationship between RSSI and spatial location in the present invention;

FIG. 4 is a diagram illustrating weight distribution according to the present invention.

Detailed Description

The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.

Referring to fig. 2, the present embodiment provides a weighted positioning method based on RSSI probability distribution, including the following steps: simplifying a positioning space into a plurality of discrete spaces, and acquiring RSSI data acquired by a terminal to be positioned; based on the mapping relation between the distance and the RSSI probability distribution obtained off-line and the RSSI probability distribution in the positioning space, the weight of each discrete space is estimated by adopting a Bayesian formula, and the position of the discrete space with the maximum weight is used as the positioning result. According to the method, the influence of environmental noise and external uncertain factors on the positioning precision is reduced by introducing the prior RSSI probability distribution into the calculation of the weight and giving the weight with a lower abnormal value.

The probability density function of the RSSI at a fixed point can be approximately in Gaussian distribution and is stable in time, but the positioning algorithm aims to calculate the position of the terminal according to the RSSI measured by the terminal to be positioned, so that the method combines the Gaussian distribution of the RSSI with the position information. Fig. 3 depicts the probability density P (RSSI | L) for a known distance L of the terminal from the AP. In the positioning practice, the position of the terminal is unknown, and the RSSI measured by the terminal to be positioned has different probability density values at different distances under the condition that the distance between the terminal and the AP cannot be obtained. In order to combine the known RSSI gaussian distribution with the position information, L is used as an independent variable, a bayesian formula is adopted to solve P (L | RSSI), and the problem is converted into the known RSSI, and the probability distribution of the terminal on the distance L is solved, as shown in formula (1):

wherein P (L | RSSI) represents the probability that the distance between the terminal and the AP is L meters when the RSSI is known; p (RSSI) represents the distribution of RSSI values in space; l can be assumed to be equally distributed throughout the scene. P (RSSI | L) follows Gaussian distribution and can be obtained by acquiring RSSI data at a position L meters away from a signal source and performing Gaussian fitting on the frequency distribution of the data.

In this embodiment, the P (RSSI | L) reference fingerprint location method is obtained, and mapping is established by establishing equal-interval acquisition points and acquiring RSSI in advance at each acquisition point:

L～P(RSSI∣L) (2)

to ensure the accuracy of p (RSSI), the number of RSSI collected at each data acquisition point should be the same, and then statistical analysis is performed on all RSSI data to obtain p (RSSI).

In indoor positioning, in order to reduce the complexity of the problem, attention is usually paid to the two-dimensional position of a positioning target, and the height of a terminal to be positioned is not paid to. In a preferred embodiment, the two-dimensional space is reduced to discrete spaces with side lengths equal to Δ d, each of which serves as a data acquisition point. The distance L from each discrete space to the AP is calculated in advance, and a gaussian distribution parameter describing the RSSI characteristic is given to each discrete space according to equation (2).

In two dimensions, L is represented as

Wherein x and y are horizontal and vertical coordinates of a central point of the discrete space, and xc and yc represent horizontal and vertical marks of the AP. Weight of discrete space with coordinates (x, y):

w(x,y)＝P(L∣RSSI) (4)

in the practice of indoor positioning, a large number of APs are deployed to improve the accuracy and stability of the positioning result, and the receiving device often receives the RSSI of multiple APs. Suppose that the terminal to be positioned receives the RSSI of N APs, where 3 is equal to or less than N. Final weight of discrete space with coordinates (x, y)

Wherein L is_i、RSSI_iRespectively representing the distance of the discrete space from the ith AP and the RSSI received from the ith AP.

As shown in fig. 4, after the weights of the discrete spaces are obtained, the position of the discrete space with the largest weight is used as the positioning result, i.e., the estimated coordinates (x) of the positioning terminal^*,y^*) At the maximum value of W (x, y), it can be obtained by solving the following optimization function

To take advantage of the multi-core performance of modern computers, this embodiment uses a particle swarm algorithm to solve equation (6). To reduce the shrinkage time, the initial value of the particle position is near the last positioning result.

The weighted positioning method based on the RSSI probability distribution can be divided into an offline training stage and an online positioning stage, as shown in fig. 2, which can be specifically described as

An off-line training stage:

step 1: selecting an experimental scene, selecting an area and arranging APs according to positioning requirements;

step 2: collecting the same amount of data at each data acquisition point;

and step 3: calculating frequency distribution on each group of collected data, performing Gaussian fitting, establishing mappings L-P (RSSI |) and obtaining the mapping which cannot be directly obtained by adopting a cubic spline interpolation method;

and 4, step 4: summarizing all collected data to calculate RSSI prior probability P (RSSI);

and (3) in an online positioning stage:

and 5: the positioning terminal collects RSSI data in a selected area;

step 6: establishing an expression (6) according to the mapping relations L-P (RSSI | L) and P (RSSI);

and 7: solving the formula (6) by using a particle swarm algorithm to obtain the estimated position of the terminal to be positioned.

The above method, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

There is also provided in another embodiment an electronic device comprising one or more processors, memory, and one or more programs stored in the memory, the one or more programs including instructions for performing the weighted positioning method as described above.

The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the prior art according to the concept of the present invention should be within the scope of protection defined by the claims.