CN107801147A - One kind is based on the adaptive indoor orientation method of the improved multizone of RSSI rangings - Google Patents

Abstract

The invention discloses an improved multi-area self-adaptive indoor positioning method based on RSSI ranging. First, the target area is divided into multiple sub-area environments according to the indoor structure characteristics, and an extended Shadowing model based on an environmental parameter library is constructed, and the extended Shadowing model is designed and Match the hardware system structure, and then use the Shadowing extended model and hardware system structure to update the environmental parameter library. In the process, use the Kalman filter algorithm to filter the RSSI value received at the reference distance and regularly update the environmental parameter library. Updating, and finally, position estimation by maximum likelihood estimation, the present invention solves the problem of poor indoor positioning accuracy caused by excessive dependence on the external environment in the indoor positioning method existing in the prior art.

Description

An improved multi-region adaptive indoor positioning method based on RSSI ranging

technical field

The invention belongs to the technical field of indoor positioning, and in particular relates to an improved multi-area adaptive indoor positioning method based on RSSI ranging.

Background technique

Indoor positioning refers to the realization of position positioning in the indoor environment. It mainly adopts wireless communication, base station positioning, inertial navigation positioning and other technologies to form an indoor position positioning system, so as to realize the position monitoring of people and objects in the indoor space. The purpose is to solve the problem that the satellite signal is weak when it reaches the ground and cannot penetrate buildings. With the increasing development of mobile Internet technology, indoor positioning has become more and more practical and necessary in some specific occasions (such as shopping malls, museums, exhibition halls, airports, etc.), and its application prospects are broad. At present, the main technologies for realizing indoor positioning include: Zigbee, Wi-Fi, Bluetooth, RFID, etc. Among them, RFID technology has a problem of high investment cost, and Wi-Fi has a certain radiation problem. Based on this, Bluetooth devices with low cost, low energy consumption and minimal harm to human body are favored by indoor positioning researchers. However, because the indoor positioning accuracy is too dependent on the external environment, the problem of low positioning accuracy has not been effectively resolved so far. Therefore, exploring the indoor positioning method with adaptive environment and high precision has become a realistic problem that needs to be solved urgently.

Through literature review, it is found that the auxiliary hardware devices often used in indoor positioning are mainly wireless signal transmitters (such as Wi-Fi, Bluetooth, etc.), and the propagation forms of wireless signals of these devices are direct, diffraction and scattering (reflection), and The received signal strength (RSSI) at any point in the signal coverage area is not only a random variable, but also a vector sum of various propagation paths, which changes with the dynamic changes of the propagation environment, which seriously interferes with the accuracy of indoor positioning algorithms Spend. Fundamentally speaking, the root causes of this indoor positioning error problem are: first, different buildings have different indoor layouts, material structures, building scales, etc., resulting in different signal path losses; The internal structure of the signal will cause the reflection, diffraction, refraction and scattering of the signal, and superimpose each other to form a multipath phenomenon, which will cause errors in the amplitude, phase and arrival time of the received signal, and cause signal loss, which seriously affects indoor positioning. the accuracy.

In addition, because the RSSI value is easily disturbed by external environmental noise, it is easy to be in a fluctuating state. Taking 100 RSSI values collected continuously at a distance of 1m from the transmitter as an example, the resulting RSSI values are shown in Figure 3. Since the positioning method of the present invention is implemented based on RSSI ranging, the accuracy and stability of RSSI is an important factor affecting subsequent ranging and positioning accuracy.

Contents of the invention

The purpose of the present invention is to provide an improved multi-region adaptive indoor positioning method based on RSSI ranging, which solves the problem that the existing indoor positioning method in the prior art relies too much on the external environment to cause poor indoor positioning accuracy.

The technical scheme adopted in the present invention is, a kind of multi-area self-adaptive indoor positioning method based on RSSI ranging improvement, specifically implements according to the following steps:

Step 1: Divide the target area into multiple sub-regional environments according to the characteristics of the indoor structure, and build an extended Shadowing model based on the environmental parameter library;

Step 2: Design and build a hardware system structure that matches the extended Shadowing model obtained in Step 1;

Step 3: With the help of the Shadowing extended model built in step 1 and the hardware system structure obtained in step 2, the environmental parameter library is regularly updated. In the process, the Kalman filter algorithm is used to analyze the parameters in the model. Value to filter;

Step 4: Maximum Likelihood Estimation for position estimation.

The present invention is also characterized in that,

Step 1 is specifically implemented according to the following steps:

Step (1.1), divide the positioning coverage area according to the structure of the target area to form n sub-areas, and use A to represent the positioning coverage area, then A={a ₁ , a ₂ , a ₃ ,..., a _n };

Step (1.2), assuming that each sub-area a _i is in the same environment, where i=1, 2,..., n, then

According to the Shadowing model, each a _i corresponds to the path loss index n _i of other areas expressed as:

n _i = {n _i1 , n _i2 , . . . , n _in }

Similarly, the RSSI value at the reference distance is Then the path loss index N and the received signal power Pr(D ₀ ) corresponding to the reference distance d ₀ point are respectively expressed as follows:

Among them, n _ij is any path loss index in N, indicating the path loss index between the area a _i where the signal transmitter AP belongs to and the area a _j where the point p to be measured is located; Pr(d _{0ij )} is The RSSI value at any reference distance indicates the received signal power value corresponding to the area a _i where the AP belongs to and the area a _j where the point p to be measured is located at the reference distance d ₀ , and n _ij =n _ji , Pr(d _0ij )=Pr (d _0ji ), build environment parameter library K=(Pr(D ₀ ),N),

So the Shadowing extension model is:

The specific expression of the Shadowing model is:

Pr(d)=Pr(d ₀ )-10nlg(d/d ₀ )+X _δ

Among them, Pr(d) represents the received RSSI value of the receiving end; Pr(d ₀ ) is the received RSSI value corresponding to the reference distance d ₀ ; n is the path loss index, and n is related to the environment; d represents the receiving end and The distance between transmitters; d ₀ is the reference distance; X _δ represents a Gaussian random variable, and the average value of X _δ is 0, which mainly reflects the change of received signal power when the distance is constant.

The environmental parameter library K is updated through the following steps:

Step a, obtain the RSSI at the reference distance in real time and perform Kalman filtering;

Step b, calculate the path loss index n corresponding to each AP through the Shadowing extended model, and average the path loss index n in the same area or corresponding area;

Step c, updating the environment parameter library K.

Step 2 Design and build the indoor positioning hardware system model compatible with the Shadowing extended model to meet the following conditions:

Build a LAN environment in the area to be located, and arrange APs and anchor nodes in each area that has been divided. At least one anchor node and four AP devices are arranged in each area, and one AP device must have one AP as a reference point AP. , the network environment ensures that the anchor node device can upload all scanned AP information to the server through the Internet in time to prepare for the update of the environmental parameter database. Library, complete the position determination at the point to be positioned,

From the indoor positioning hardware system model, it can be obtained that the hardware equipment in an area is represented by H, and there is a relationship: H={AP _ij , S _ij }; AP _ij represents any single AP, and S _ij represents any anchor node S, and the layout area and location are known, where i represents the area to which the device belongs, j represents the device number, i,j∈(1,2,3,...,n).

Step 3 is specifically:

Combined with the indoor positioning hardware system model obtained in step 2, set the signal transmitter AP around the coverage area, set the anchor node at the center of the area, and put them in the "on" state, then the anchor nodes will cycle within the equal time interval t Scan nearby AP information, and send the information to the server through the LAN at the same time. The server judges the area and location coordinates of each according to the received anchor node ID and the MAC of the signal transmitter AP. At the same time, calculate and update N and Received signal power Pr(D ₀ ).

Calculate and update N and received signal power Pr(D ₀ ) according to the Shadowing extended model as follows:

Assuming that there are m signal transmitter APs in a sub-area or compound area, the calculation process of the path loss index n in this area is as follows:

In the above formula, i represents the i-th signal transmitter AP in the area. According to the calculation process of the path loss index n, the anchor node needs to obtain the UUID and RSSI information of the surrounding APs in real time, perform area identification and parameter calculation, and update the current The environmental parameter library K, so as to achieve the self-adaptation to the environment, and achieve the purpose of improving the accuracy of indoor positioning;

Assuming that the extended Shadowing model is a discrete control system, and the system can be described by linear stochastic differential equations, the process equation and observation equation of Kalman filter are as follows:

x _k ＝Ax _k-1 +Bu _k-1 +q _k-1

y _k =Hx _k +r _k

In the above formula, x _k is the RSSI value to be optimized at time k, u _k-1 is the control quantity of the system at time k-1, A and B are system parameters, y _k is the measured value of RSSI at time k, and H is the measurement parameter , r _k and q _k-1 represent noise respectively, and both mean values are 0, therefore, the covariance of the two is expressed as:

In the above formula, Q _k and R _k represent the covariance matrix of system noise and measurement noise respectively,

Therefore, the Kalman filter is expressed as the following two processes, namely:

(1) Time update:

(2) Status update:

In the above formula, Indicates the predicted system state value at time k; Denotes the new error of the process noise Q prediction; K _k denotes the Kalman gain; Indicates the optimal state value of the system at time k.

Step 4 is specifically:

Suppose the target position is p(x ₀ ,y ₀ ), then the position where n AP information is received at point p is expressed as AP( _xi ,y _i ), i∈1,2,…,n, then the target and The distance d _i between each AP, i∈1,2,...,n, the calculation process is as follows:

In the above formula, the difference between the last item and other items is used to calculate the difference, and the following equation is obtained:

Let x=(x ₀ ,y ₀ ) ^T , then the above formula is expressed in matrix form Ax=b as:

Solve the matrix by the method of least squares to get:

x＝(A ^T A) ^-1 A ^T b

By x=(x ₀ ,y ₀ ) ^T =(A ^T A) ^-1 A ^T b can determine the position coordinate information of the point to be positioned p(x ₀ ,y ₀ ), and p(x ₀ ,y ₀ ) point coordinates to realize position estimation.

The beneficial effect of the present invention is that a multi-area self-adaptive indoor positioning method based on RSSI ranging improvement firstly divides the target area into multiple sub-area environments according to the indoor structure characteristics, constructs an extended Shadowing model based on the environmental parameter library, and designs The hardware system structure that matches the extended model, and then use the Kalman filter algorithm to filter the RSSI value received at the reference distance and update the environmental parameter library regularly, and finally use the maximum likelihood estimation to locate the target, so as to realize the environmental Adaptive and improve the accuracy of indoor positioning.

Description of drawings

Fig. 1 is a kind of system hardware architecture diagram of the multi-area self-adaptive indoor positioning method based on RSSI ranging improvement of the present invention;

Fig. 2 is a kind of improved multi-area adaptive indoor positioning method flow diagram based on RSSI ranging of the present invention;

Fig. 3 is a graph of the RSSI value at the unit length in a kind of improved multi-area adaptive indoor positioning method based on RSSI ranging of the present invention;

Fig. 4 is a kind of multi-area self-adaptive indoor positioning method experimental scene figure based on RSSI ranging improvement of the present invention;

Fig. 5 is a kind of ranging error comparison result figure in the multi-area self-adaptive indoor positioning method based on RSSI ranging improvement of the present invention;

Fig. 6 is a diagram of comparison results of positioning errors of an improved multi-region adaptive indoor positioning method based on RSSI ranging in the present invention.

Detailed ways

The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

The present invention is a multi-area self-adaptive indoor positioning method based on RSSI ranging improvement, the flow chart is as shown in Figure 2, specifically implemented according to the following steps:

Step 1: Divide the target area into multiple sub-regional environments according to the characteristics of the indoor structure, and build an extended Shadowing model based on the environmental parameter library. Specifically, follow the steps below:

Step (1.1), divide the positioning coverage area according to the structure of the target area to form n sub-areas, and use A to represent the positioning coverage area, then A={a ₁ , a ₂ , a ₃ ,..., a _n };

Step (1.2), assuming that each sub-area a _i is in the same environment, where i=1, 2, ..., n, then according to the Shadowing model, each a _i corresponds to the path loss index n _i of other areas expressed as:

n _i = {n _i1 , n _i2 ,..., n _in }

Similarly, the RSSI value at the reference distance is Then the path loss index N and the received signal power Pr(D ₀ ) corresponding to the reference distance d ₀ point are respectively expressed as follows:

In the above formula, i=1,2,3,...,n; j=1,2,3,...,n;

Among them, n _ij is any path loss index in N, indicating the path loss index between the area a _i where the signal transmitter AP belongs to and the area a _j where the point p to be measured is located; Pr(d _{0ij )} is The RSSI value at any reference distance indicates the received signal power value corresponding to the area a _i where the AP belongs to and the area a _j where the point p to be measured is located at the reference distance d ₀ , and n _ij =n _ji , Pr(d _0ij )=Pr (d _0ji ), build environment parameter library K=(Pr(D ₀ ),N),

Then the Shadowing extension model is:

The specific expression of the Shadowing model is:

Pr(d)=Pr(d ₀ )-10nlg(d/d ₀ )+X _δ

Among them, Pr(d) represents the received RSSI value of the receiving end; Pr(d ₀ ) is the received RSSI value corresponding to the reference distance d ₀ point; n is the path loss index, and n is related to the environment; d represents the receiving end and The distance between the transmitters; d ₀ is the reference distance, usually d ₀ is usually 1m; X _δ represents a Gaussian random variable, and the average value of X _δ is 0, which mainly reflects the change of received signal power when the distance is constant;

The environmental parameter library K is updated through the following steps:

Step a, obtain the RSSI at the reference distance in real time and perform Kalman filtering;

Step b, calculate the path loss index n corresponding to each AP through the Shadowing extended model, and average the path loss index n in the same area or corresponding area;

Step c, updating the environment parameter database K;

Step 2: For the Shadowing extended model, when the environment is different, the corresponding environmental parameters are also different, and such different environments and parameters are likely to cause the path loss index N to be in an active state, and it is difficult to accurately determine its value , to this end, based on the extended Shadowing model, design a hardware system structure that matches the extended Shadowing model obtained in step 1, as shown in Figure 1, build a LAN environment in the area to be located, and arrange APs and anchors in each divided area Node, at least one anchor node and four AP devices are arranged in each area, and one AP device must have one AP as a reference point AP. Upload to the server to prepare for the update of the environmental parameter library. When the point to be positioned obtains the surrounding AP information, it will obtain the real-time environment parameter library from the Internet to complete the location determination of the point to be positioned.

From the indoor positioning hardware system model, it can be obtained that the hardware equipment in an area is represented by H, and there is a relationship: H={AP _ij , S _ij }; AP _ij represents any single AP, and S _ij represents any anchor node S, and the layout area and location are known, where i indicates the area to which the equipment belongs, j indicates the equipment number, i,j∈(1,2,3,...,n);

Step 3: With the help of the Shadowing extended model constructed in the step 1 and the hardware system structure obtained in the step 2, the update of the environmental parameter library is realized, and the RSSI value received at the reference distance is analyzed by the Kalman filter algorithm To filter, specifically:

Combined with the indoor positioning hardware system model obtained in step 2, set the signal transmitter AP (at least 0.3m away from the wall) around the coverage area, and set the anchor node at the center of the area, and put them in the "on" state, then the anchor The node scans the nearby AP information cyclically within the equal time interval t, and sends the information to the server through the LAN at the same time. The server judges the respective area and location coordinates according to the received anchor node ID and the MAC of the signal transmitter AP. The Shadowing extended model calculates and updates N and received signal power Pr(D ₀ );

Calculate and update N and received signal power Pr(D ₀ ) according to the Shadowing extended model as follows:

Assuming that there are m signal transmitter APs in a sub-area or compound area, the calculation process of the path loss index n in this area is as follows:

In the above formula, i represents the i-th signal transmitter AP in the area. According to the calculation process of the path loss index n, the anchor node needs to obtain the UUID, RSSI, MAC and other information of the surrounding APs in real time, perform area identification and parameter calculation, and real-time Update the current environmental parameter library K, so as to achieve self-adaptation to the environment and improve the accuracy of indoor positioning;

Since the RSSI value is easily disturbed by external environmental noise, it is easy to be in a fluctuating state. This shows to a certain extent that the RSSI value at the reference distance is not only a parameter, but also an index that affects the environmental parameter library K, or an important factor that affects subsequent ranging and positioning accuracy. As an optimal filtering algorithm in the Gaussian process, the Kalman filter has good optimization performance under the premise that the object model is accurate enough and the system state and parameters do not change suddenly. Therefore, the update of the environmental parameter library K During the process, the Kalman filter algorithm is used to preprocess the obtained RSSI value of the reference point AP to reduce the influence of the external environment on the Shadowing extended model, so as to achieve the purpose of improving the quality of the environmental parameter library and the accuracy of indoor positioning;

With the help of the extended Shadowing model obtained in step 2, assuming that the extended Shadowing model is a discrete control system, and the system can be described by a linear stochastic differential equation, the process equation and observation equation of the Kalman filter are as follows:

x _k ＝Ax _k-1 +Bu _k-1 +q _k-1

y _k =Hx _k +r _k

In the above formula, x _k is the RSSI value to be optimized at time k, u _k-1 is the control quantity of the system at time k-1, A and B are system parameters, y _k is the measured value of RSSI at time k, and H is the measurement parameter , represented by a matrix in a multi-measurement system, q _k-1 and r _k represent noise respectively, according to the theory of probability and statistics, q _k-1 and r _k are generally considered to be Gaussian white noise, and both have an important mean value of 0 characteristics, for which the covariance of the two is expressed as:

In the above formula, Q _k and R _k represent the covariance matrix of system noise and measurement noise, respectively.

For satisfying the above two conditions, the Kalman filter has the optimal information processor. Therefore, in the discrete control system in the previous step, the Kalman filter is expressed as the following two processes, namely:

(1) Time update:

(2) Status update:

In the above formula, Indicates the predicted system state value at time k; Denotes the new error of the process noise Q prediction; K _k denotes the Kalman gain; Indicates the optimal state value of the system at time k;

Step 4: Maximum likelihood estimation for position estimation, specifically:

Filter the RSSI value obtained at the reference distance through the Kalman filter algorithm in step 3 and calculate the corresponding path loss index and update the environmental parameter library. Assuming that the target position is p(x ₀ ,y ₀ ), then n is received at point p Among them, n>=4, the location of the AP information is expressed as AP( _xi , y _i ), i∈1,2,...,n, combined with the Shadowing extended model, the distance between the target and each AP is calculated d _i , i∈1,2,…,n, the specific calculation process is as follows:

In the above formula, use the last item to calculate the difference with other items respectively, and get the following equation:

Let x=(x ₀ ,y ₀ ) ^T , then the above formula is expressed in matrix form Ax=b as:

Then solve the matrix by least squares method, get:

x＝(A ^T A) ^-1 A ^T b

Thus, target location is performed using maximum likelihood estimation.

In order to verify the proposed improved multi-area adaptive indoor positioning method based on RSSI ranging, the experimental scene shown in Figure 4 is selected and laid out. At the same time, according to the characteristics of the environment space, the experimental scene is divided into 3 areas A, B and C (respectively 6.65m×3.6m, 8m×1.75m and 6.65m×3.6m), in which 4 APs are placed at fixed points in each area ( One of them can also be a reference point AP) and one anchor node. The AP comes from BrightBeacon, the iBeacon signal transmitting base station provided by Zhishi Technology, and the anchor node comes from the remote control terminal CloudBeacon of Zhishi Technology. A total of 21 points are selected as the target points to be measured, each point is collected for 3s, and the signal acquisition device at the target point is Xiaomi 2S (Android 5.0.2).

First, sample data is collected in the target area to obtain the mean values of the environmental parameters in the three areas (the reference distance corresponding to the RSSI value is 1m, and the format is (Pr(d ₀ ), n)), and the results are shown in Table 1 :

Table 1 Mean values of environmental parameters in each region

According to the data in Table 1, the Shadowing model mean method and the Shadowing extended model mean method (that is, the extended mean method) are selected, and the method proposed in this paper is tested in terms of ranging accuracy. Before the positioning test, initialize each parameter required for the experiment. The specific initial values of the parameters are shown in Table 2:

Table 2 Initial value setting

During the experiment, firstly, 10 points are randomly selected in the target area, each point is collected for 3 seconds, and the mean value processing is performed according to the AP information obtained at each point. Since the method proposed in this paper is based on RSSI distance measurement, the four AP points with the largest average value are selected, and the average value method and the extended average method are used for calculation respectively, and the performance of distance measurement is compared. The results are shown in Figure 5 Show.

It can be seen from Figure 5 that the multi-area environment adaptive indoor positioning method based on the Kalman filter algorithm to improve RSSI ranging has a maximum error value of 1.36m and an average error of 0.524m; the maximum error value of the extended mean method is 2.57m, The average error is 0.647m; the maximum error value of the average method is 2.29m, and the average error is 0.802m. This experimental result shows that the average method has the largest ranging error. Moreover, compared with the mean method and the extended mean method, the positioning method based on the Kalman filter algorithm to improve RSSI ranging reduces the average error by 34.66% and 19.01%, respectively. This result also shows that the method with a smaller ranging error is beneficial to Improve the accuracy of indoor positioning.

On this basis, in order to further verify and explore the mechanism of the problem of low accuracy, 21 points to be positioned are selected from the three areas A, B, and C for experimental design and testing. The indoor positioning test results are shown in Figure 6 shown.

It can be seen from Figure 6 that the average error of the improved RSSI ranging and positioning method based on the Kalman filter algorithm is 1.0005m, the average positioning error of the extended mean method is 1.1785m, and the average positioning error of the mean method is 1.2895m. In comparison, the errors are respectively A drop of 15.1% and 22.41%. Moreover, among the 21 points to be located, the error rate of 15 points based on the improved RSSI ranging and positioning method based on the Kalman filter algorithm is lower than that of the other two, and the reliability of this method is as high as 71.43%. Furthermore, at point 13, the three methods all reached the maximum value of their respective errors, and the mean positioning error was the largest, while at point 5, the error of the improved RSSI ranging and positioning method based on the Kalman filter algorithm was less than 0.5 meters. In addition, the extended mean positioning error is almost close to the error of the improved RSSI ranging positioning method based on the Kalman filter algorithm. This result means that when the local environment of the target area changes, the indoor positioning error will become larger, but the improved RSSI ranging and positioning method based on the Kalman filter algorithm regularly collects data in the area and updates the environmental parameter library K, effectively Weaken the impact of local environmental changes on indoor positioning accuracy, thus reflecting the characteristics of good environmental adaptation.

The above is only a preferred embodiment of the present invention, it should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, some improvements and modifications can also be made, and these improvements and modifications should also be It is regarded as the protection scope of the present invention.

Claims (8)

1. one kind is based on the adaptive indoor orientation method of the improved multizone of RSSI rangings, it is characterised in that specifically according to following Step is implemented：

Step 1：Target area is divided into more sub-regions environment according to doors structure feature, built based on ambient parameter storehouse Shadowing extended models；

Step 2：Design and build the hardware system structure that the Shadowing extended models obtained with the step 1 match；

Step 3：The Shadowing extended models built by the step 1 and the hardware system knot obtained with the step 2 Structure, the renewal in ambient parameter storehouse is realized, in the process using Kalman filtering algorithm to the RSSI that is received at reference distance Value is filtered；

Step 4：Maximum-likelihood estimation carries out location estimation.

2. one kind according to claim 1 is based on the adaptive indoor orientation method of the improved multizone of RSSI rangings, it is special Sign is that the step 1 is specifically implemented according to following steps：

Step (1.1), will positioning overlay area split according to target area structure, formed n sub-regions, will positioning covering Region represents with A, then A={ a₁, a₂, a₃..., a_n}；

Step (1.2), set every sub-regions a_iUnder same environment, wherein i=1,2 ..., n, then according to Shadowing moulds Type, each a_iThe path loss index n in other corresponding regions_iIt is expressed as：

n_i={ n_i1, n_i2..., n_in}

Similarly, RSSI value is at reference distanceThen path loss index N and Reference distance d₀Received signal power Pr (D corresponding to point₀) represent as follows respectively：

Wherein, n_ijFor any one path loss index in N, signal projector AP affiliated areas a is represented_iWith tested point p places Region a_jBetween path loss index；Pr(d_0ij) it is Pr (D₀) in RSSI value at any one reference distance, represent belonging to AP Region a_iWith tested point p region a_jIn reference distance d₀Received signal power value corresponding to place, and n_ij=n_ji, Pr (d_0ij) =Pr (d_0ji), constructing environment parameter library K=(Pr (D₀), N),

So Shadowing extended models are：

3. one kind according to claim 2 is based on the adaptive indoor orientation method of the improved multizone of RSSI rangings, it is special Sign is that the Shadowing models expression is：

Pr (d)=Pr (d₀)-10nlg(d/d₀)+X_δ

Wherein, Pr (d) represents the RSSI value received of receiving terminal；Pr(d₀) it is reference distance d₀Received corresponding to place RSSI value；N is path loss index, n and environmental correclation；D represents the distance between receiving terminal and transmitting terminal；d₀For with reference to away from From；X_sRepresent Gaussian random variable, X_sAverage value is 0, and main reflection is when the timing of distance one, the change of received signal power.

4. one kind according to claim 2 is based on the adaptive indoor orientation method of the improved multizone of RSSI rangings, it is special Sign is that the ambient parameter storehouse K is updated by following steps：

Step a, RSSI at reference distance is obtained in real time and carries out Kalman filtering filtering；

Step b, path loss index n corresponding to each AP is calculated by Shadowing extended models, and to the same area or Meet each path loss index n equalizations processing in region；

Step c, ambient parameter storehouse K is updated.

5. one kind according to claim 1 is based on the adaptive indoor orientation method of the improved multizone of RSSI rangings, it is special Sign is that the indoor positioning hardware system model being adapted with Shadowing extended models that the step 2 designs meets following Condition：

LAN environment is built in area to be targeted, and AP and anchor node are arranged in the regional divided, each region is extremely Arrange that 1 anchor node and 4 AP equipment, wherein AP equipment there must be 1 AP to serve as reference point AP less.Meanwhile the network environment Ensure that anchor node device can be uploaded onto the server all AP information scanned in time by internet, in case ambient parameter Storehouse updates, and obtains real-time ambient parameter storehouse when point to be determined gets periphery AP information, and from internet end, completes undetermined Position determination at site,

By indoor positioning hardware system model and then obtain：If the hardware device in a region is represented with H, and relation be present：H= {AP_ij,S_ij}；AP_ijRepresent any one separate unit AP, S_ijThen represent that any one anchor node S, wherein i represent the affiliated area of equipment Domain, j represent device numbering, i, j ∈ (1,2,3 ..., n).

6. one kind according to claim 1 is based on the adaptive indoor orientation method of the improved multizone of RSSI rangings, it is special Sign is that the step 3 is specially：

The indoor positioning hardware system model obtained with reference to the step 2, the setting signal transmitter at the surrounding of overlay area AP, heart position sets anchor node in the zone, and is placed in "On" state, then anchor node is circulated in constant duration t and swept Neighbouring AP information is retouched, while these information are sent to server by LAN, server is according to the anchor node received ID, signal projector AP MAC judge respective affiliated area and position coordinates, meanwhile, calculated according to Shadowing extended models With renewal N and received signal power Pr (D₀)。

7. one kind according to claim 6 is based on the adaptive indoor orientation method of the improved multizone of RSSI rangings, it is special Sign is, is calculated according to Shadowing extended models and updates N and received signal power Pr (D₀) be specially：

Assuming that there is m signal projector AP in a certain subregion or recombination region, then the zone routing loss index n calculating process It is as follows：

In above formula, i represents i-th of region signal projector AP, according to path loss index n calculating process, it is necessary to anchor section Point obtains the information such as surrounding AP UUID, RSSI in real time, carries out region recognition and parameter calculates, and the environment that real-time update is current Parameter library K, so as to reach to the adaptive of environment, realize the purpose that indoor positioning precision improves；

Assuming that Shadowing extended models are discrete control system, moreover, the system can be retouched with the linear random differential equation State, then the process equation of Kalman filtering and observational equation are as follows：

x_k=Ax_k-1+Bu_k-1+q_k-1

y_k=Hx_k+r_k

In above formula, x_kIt is k moment RSSI value to be optimized, u_k-1It is controlled quentity controlled variable of the k-1 moment to system, A and B are systematic parameters, y_kIt is k moment rssi measurement values, H is measurement parameter, is represented in more measuring systems with matrix, q_k-1And r_kNoise is represented respectively, According to Probability Statistics Theory, q_k-1And r_kWhite Gaussian noise is typically considered, it is 0 this key property that both, which have average, Therefore, both covariances are expressed as：

In above formula, Q_kAnd R_kThe covariance matrix of system noise and measurement noise is represented respectively,

Therefore, Kalman filter is expressed as following two processes, i.e.,：

(1) time updates：

(2) state updates：

In above formula,Represent the k moment system mode values of prediction；Represent the new error of process noise Q predictions；K_kRepresent karr Graceful gain；Represent k moment system optimal state values.

8. one kind according to claim 1 is based on the adaptive indoor orientation method of the improved multizone of RSSI rangings, it is special Sign is that the step 4 is specially：

Assuming that target location is p (x₀,y₀), then n are received in point p, wherein, n>Positional representation where=4, AP information is AP(x_i,y_i), i ∈ 1,2 ..., n, then the distance between target and each AP d_i, i ∈ 1,2 ..., n, calculating process is as follows：

In above formula, mathematic interpolation is done with other respectively using last, obtains following equation：

Make x=(x₀,y₀)^T, then above formula be expressed as with matrix form Ax=b：

Matrix is solved by least square method, obtained：

X=(A^TA)^-1A^Tb

By x=(x₀,y₀)^T=(A^TA)^-1A^TB can determine point to be determined p (x₀,y₀) location coordinate information, according to target area Coordinate diagram and p (x₀,y₀) point coordinates realizes location estimation.