CN113590587A - Offline position fingerprint database construction method based on self-adaptive simulated annealing-particle swarm-kriging interpolation algorithm - Google Patents
Offline position fingerprint database construction method based on self-adaptive simulated annealing-particle swarm-kriging interpolation algorithm Download PDFInfo
- Publication number
- CN113590587A CN113590587A CN202110871465.4A CN202110871465A CN113590587A CN 113590587 A CN113590587 A CN 113590587A CN 202110871465 A CN202110871465 A CN 202110871465A CN 113590587 A CN113590587 A CN 113590587A
- Authority
- CN
- China
- Prior art keywords
- value
- self
- particle
- simulated annealing
- fingerprint database
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Mathematical Analysis (AREA)
- Artificial Intelligence (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Algebra (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Pure & Applied Mathematics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Complex Calculations (AREA)
Abstract
The invention discloses an off-line position fingerprint database construction method based on self-adaptive simulated annealing-particle swarm-kriging interpolation, which comprises the following steps of: firstly, collecting the received signal intensity values of partial points, then carrying out parameter optimization on a theoretical spherical model variation function by using a self-adaptive simulated annealing particle swarm optimization algorithm, and establishing a self-adaptive simulated annealing-particle swarm-kriging interpolation model. And estimating the received signal intensity value of the predicted point through the interpolation model, and constructing a complete off-line position fingerprint database together with the numerical value of the acquisition point.
Description
Technical Field
The invention relates to the field of indoor positioning, in particular to an offline position fingerprint database construction method based on a self-adaptive simulated annealing-particle swarm-kriging interpolation algorithm.
Background
The personnel positioning system is an important component in operation and maintenance management of urban underground comprehensive pipe gallery operation, and from the aspect of safety, inspection personnel of the underground pipe gallery not only need to overcome complex working conditions caused by complex environments during working, but also face dangers to a certain degree. In case of an emergency, the system of the utility tunnel operation and maintenance center needs to calibrate the specific position of the staff in time in order to deploy the rescue. According to different applications of positioning technology, positioning systems in the construction of domestic comprehensive pipe galleries can be divided into passive personnel positioning based on online patrol, personnel positioning based on WiFi, personnel positioning based on Ultra Wideband (UWB) technology, personnel positioning based on Radio Frequency Identification (RFID) technology, personnel positioning based on Bluetooth technology and the like. Among them, the WiFi-based location fingerprint positioning method is most widely used due to its properties such as low cost, wide coverage, strong signal interference, etc.
However, the main structure of the urban underground comprehensive pipe gallery is mainly located underground, mobile phone signals and GPS satellite positioning signals cannot be received in the pipe gallery, 3-4 wireless access points are arranged in each area, and the distance between every two access points is about 50 m. However, such distribution is sparse, and workers in the pipe gallery cannot receive 3 or more signal intensity values at the same time. And when the traditional position fingerprint method is adopted for positioning, a large number of off-line data points need to be collected, and the method needs great labor cost.
In summary, an offline position fingerprint database construction method based on adaptive simulated annealing-particle swarm-kriging interpolation algorithm is provided.
Disclosure of Invention
The invention aims to provide an off-line position fingerprint database construction method based on a self-adaptive simulated annealing-particle swarm-kriging interpolation algorithm, and aims to reduce a large amount of labor cost required by a traditional position fingerprint method for positioning in an urban comprehensive pipe gallery environment and improve wireless positioning efficiency.
The invention is realized by the following technical method:
(1) in the acquisition stage, the WiFi receiving signal intensity values of all reference points in the area are acquired;
(2) carrying out mean value filtering processing on the acquired signal intensity values, and storing the signal intensity values into a position fingerprint database;
(3) in the interpolation stage, firstly, selecting a prediction point in a region to prepare for interpolation processing;
(4) the interpolation processing is carried out by adopting the self-adaptive simulated annealing-particle swarm-kriging interpolation algorithm;
(5) establishing an interpolation position fingerprint database through the interpolated received signal strength data, namely the data of the predicted point;
(6) and combining the collected and established position fingerprint database with the interpolated position fingerprint database to jointly form the off-line position fingerprint database provided by the invention.
The adaptive simulated annealing-particle swarm-kriging interpolation algorithm process comprises the following steps:
(1) measuring the received signal strength value of a node in a target area;
(2) calculating the physical distance of each node;
(3) calculating the actual variation function value of the kriging interpolation;
(4) selecting a spherical model as a theoretical variation function model;
(5) calculating a variation model parameter by using a self-adaptive simulated annealing particle swarm optimization algorithm;
(6) constructing a Kriging equation set, and calculating a weight coefficient lambda;
(7) calculating to obtain a received signal strength value of a node to be interpolated;
compared with the positioning effect of the traditional position fingerprint method in the environment of the urban underground comprehensive pipe gallery, the method has the following advantages:
1. compared with the traditional particle swarm optimization algorithm, the self-adaptive simulated annealing particle swarm optimization algorithm provided by the invention has stronger optimization capability and higher optimization precision;
2. the invention can reduce the workload of manual off-line acquisition by 50 percent;
3. the average positioning precision and the error accumulation probability of the invention are improved.
Drawings
Fig. 1 is a flow chart of the wireless location fingerprint database construction according to the present invention.
FIG. 2 is a flow chart of the adaptive simulated annealing-particle swarm-kriging interpolation algorithm of the present invention.
Fig. 3 is a schematic diagram of the distribution of wireless APs of the urban underground comprehensive pipe gallery.
Detailed Description
The invention is described in detail below with reference to the drawings and specific embodiments.
Fig. 1 is a flow chart of the construction of the wireless location fingerprint database according to the present invention. The off-line position fingerprint database construction process is as follows: in the acquisition stage, firstly, position fingerprint data acquisition is carried out on each point of a target area, then filtering processing is carried out, and an acquisition position fingerprint database is constructed; in the interpolation stage, firstly, points to be interpolated are selected, interpolation of each point, namely a received signal intensity value, is calculated by using a self-adaptive simulated annealing-particle swarm-kriging interpolation algorithm, and an interpolation position fingerprint database is constructed. The acquired position fingerprint database and the interpolation position fingerprint database are combined to form the off-line position fingerprint database provided by the invention.
FIG. 2 is a flow chart of the adaptive simulated annealing-particle swarm-kriging interpolation algorithm of the present invention, the flow of the interpolation algorithm is as follows:
1. measuring the WiFi receiving signal intensity value of each node in the target area, then calculating the distance between the nodes, and then calculating the actual variation function value of the kriging interpolation according to the following formula:
where M (h) is the number of point pairs at distance h.
2. After the actual variation function value is calculated, a spherical model is selected as a theoretical variation function model, wherein the spherical model is shown as the following formula:
wherein C is0Is the lump constant, C is the arch height, C is0+ C is the base value and a is the range.
3. Then, the parameters of the variation function model are obtained by using a self-adaptive simulated annealing-particle swarm optimization algorithm, and the self-adaptive simulated annealing particle swarm optimization algorithm comprises the following steps:
(1) setting boundary values of search space and search speed, and setting population size PopsizeAnd maximum number of iterations KmaxAnd initializing the particle group to produceGenerating initial positions and initial speeds of all particles;
(2) searching an optimal value of the current global particles, recording the optimal value as gbest, and setting the initial temperature of simulated annealing according to the following formula;
wherein T is the initial temperature, k is the iteration number, and E (gbest) is the fitness value of the current global optimum value;
(3) adaptively varying ω, c according to1，c2. The invention selects [ -4,4 [ -4 [ ]]The hyperbolic tangent curve between the two controls the change of the inertia weight coefficient. The adaptive weight coefficient function is as follows:
wherein, ω ismaxAnd ωminThe present invention takes 0.95 and 0.4 as the maximum and minimum values of the weight coefficient. k is the current iteration number, kmaxIs the maximum number of iterations. Self-learning factor c1And social learning factor c2The adaptive change strategy of (1) is as follows:
wherein, c1maxAnd c1minThe maximum value and the minimum value of the self-learning factor are expressed, and the maximum value and the minimum value of the self-learning factor are 2.5 and 1.25; c. C2maxAnd c2minThe maximum value and the minimum value of the social learning factor are shown, and the maximum value and the minimum value of the social learning factor are 1.25 and 2.5.
(4) And changing the particle speed according to the following formula, carrying out iterative optimization, and calculating the particle fitness value after movement.
vi＝ωvi+c1*rand()*(pbesti-xi)+c2*rand()*(gbesti-xi)
xi＝xi+vi
Where ω is an inertial weight factor, c1As a self-recognition factor, c2For social cognition factor, rand () is a random value of (0,1), pbestiFor the ith individual particle optimum, gbestiThe population optimal value is obtained. v. ofiAnd xiRespectively representing the velocity and position of the ith particle.
(5) Updating the individual and population optima for the particles according to the following formula:
(6) the probability P of receiving a new solution is calculated according toSA。
Wherein E isi(k) Representing the fitness value of the current particle at the kth iteration; egRepresenting the internal energy of the current particle optimal point, namely the current global optimal value; t isiIndicating the current temperature.
(7) Will PSAAnd comparing with rand (), judging whether the current global optimum value is replaced by the generated new solution, namely the current particle fitness value, carrying out annealing operation and updating the temperature.
(8) And (5) judging whether the maximum times is reached, and if not, returning to the step (3).
(9) And outputting the current optimal particles, and terminating the algorithm.
The invention optimizes the spherical model of the theoretical variation function of the kriging interpolation, and the adopted fitness function is shown as the following formula:
wherein, gamma (h)m) Is a spherical model of a theoretical variation function, gamma' (h)m) For the actual variation function, M is the number of separations, hmIs the mth separation distance.
4. The kriging equation set is constructed as follows, and the weight coefficient λ is calculated.
Where μ is the Lagrangian function factor, this equation can be represented by:
can be abbreviated as:
AX＝B
wherein, A, X and B respectively correspond to the left matrix, the middle matrix and the right matrix. And the weight coefficient vector X may be calculated by:
X＝A`B
wherein gamma (x) in the A matrixi-yi) The value of (b) can be obtained from the corresponding variogram of the spherical theoretical model.
5. And calculating the received signal strength value of the node to be interpolated, namely the final interpolation result.
Fig. 3 is a distribution diagram of wireless APs of the urban underground comprehensive pipe rack, which can be obtained from fig. 1, the wireless APs are distributed in a single manner in the long and narrow underground pipe rack, and the staff can only receive a single WiFi received signal intensity in the process of moving.
Claims (3)
1. An off-line position fingerprint database construction method based on self-adaptive simulated annealing-particle swarm-kriging interpolation algorithm is characterized by comprising the following steps:
s1, collecting position fingerprint data of each point of the target area;
s2, carrying out mean value filtering processing on the acquired data to construct an acquisition position fingerprint database;
s3, selecting points to be interpolated, calculating interpolation of each point by using the self-adaptive simulated annealing-particle swarm-kriging interpolation algorithm, namely, receiving signal intensity value, and constructing an interpolation position fingerprint database;
s4 combines the collected location fingerprint database with the interpolated location fingerprint database to form the offline location fingerprint database of the present invention.
2. The method for constructing the offline position fingerprint database based on the adaptive simulated annealing-particle swarm-kriging interpolation algorithm according to claim 1, wherein the interpolation algorithm comprises the following steps:
s1 measures the WiFi received signal strength value of each node in the target area, then calculates the distance between the nodes, and then calculates the actual variance function value of the kriging interpolation according to the following formula:
where M (h) is the number of point pairs at distance h.
S2, after calculating the actual variation function value, selects a spherical model as the theoretical variation function model, where the spherical model is represented by the following formula:
wherein C is0Is the lump constant, C is the arch height, C is0+ C is the base value and a is the range.
S3, calculating the variation function model parameter by using the self-adaptive simulated annealing-particle swarm optimization algorithm, optimizing the theoretical variation function spherical model of the kriging interpolation, wherein the fitness function is shown as the following formula:
wherein, gamma (h)m) Is a spherical model of a theoretical variation function, gamma' (h)m) For the actual variation function, M is the number of separations, hmIs the mth separation distance.
S4 constructs a kriging equation set as follows, and calculates the weight coefficient λ.
Where μ is the Lagrangian function factor, this equation can be represented by:
can be abbreviated as:
AX＝B
wherein, A, X and B respectively correspond to the left matrix, the middle matrix and the right matrix. And the weight coefficient vector X may be calculated by:
X＝A`B
wherein gamma (x) in the A matrixi-yi) The value of (b) can be obtained from the corresponding variogram of the spherical theoretical model.
S5 calculates the received signal strength value of the node to be interpolated, i.e. the final interpolation result.
3. The adaptive simulated annealing-particle swarm-kriging interpolation method according to claim 2, wherein the adaptive simulated annealing particle swarm optimization algorithm comprises the following steps:
s1 boundary values of search space and search speed are set, and population size Pop is setsizeAnd maximum number of iterations KmaxAnd initializing the particle group to produceGenerating initial positions and initial speeds of all particles;
s2, searching an optimal value of the current global particles, recording the optimal value as gbest, and setting the initial temperature of simulated annealing according to the following formula;
wherein T is the initial temperature, k is the iteration number, and E (gbest) is the fitness value of the current global optimum value;
s3 adaptively changing omega, c according to the following formula1，c2. The invention selects [ -4,4 [ -4 [ ]]The hyperbolic tangent curve between the two controls the change of the inertia weight coefficient. The adaptive weight coefficient function is as follows:
wherein, ω ismaxAnd ωminThe present invention takes 0.95 and 0.4 as the maximum and minimum values of the weight coefficient. k is the current iteration number, kmaxIs the maximum number of iterations. Self-learning factor c1And social learning factor c2The adaptive change strategy of (1) is as follows:
wherein, c1maxAnd c1minThe maximum value and the minimum value of the self-learning factor are expressed, and the maximum value and the minimum value of the self-learning factor are 2.5 and 1.25; c. C2maxAnd c2minThe maximum value and the minimum value of the social learning factor are shown, and the maximum value and the minimum value of the social learning factor are 1.25 and 2.5.
S4, changing the particle speed according to the following formula, carrying out iterative optimization, and calculating the particle fitness value after movement.
vi＝ωvi+c1*rand()*(pbesti-xi)+c2+rand()*(gbesti-xi)
xi＝xi+vi
Where ω is an inertial weight factor, c1As a self-recognition factor, c2For social cognition factor, rand () is a random value of (0,1), pbestiFor the ith individual particle optimum, gbestiThe population optimal value is obtained. v. ofiAnd xiRespectively representing the velocity and position of the ith particle.
S5 updating the individual and population optima for the particles according to:
s6 calculating the probability P of receiving a new solution according to the following formulaSA。
Wherein E isi(k) Representing the fitness value of the current particle at the kth iteration; egRepresenting the internal energy of the current particle optimal point, namely the current global optimal value; t isiIndicating the current temperature.
S7 reaction of PSAAnd comparing with rand (), judging whether the current global optimum value is replaced by the generated new solution, namely the current particle fitness value, carrying out annealing operation and updating the temperature.
S8 judges whether the maximum number of times is reached, if not, it returns to S3.
S9 outputs the current optimal particle and the algorithm terminates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110871465.4A CN113590587A (en) | 2021-07-30 | 2021-07-30 | Offline position fingerprint database construction method based on self-adaptive simulated annealing-particle swarm-kriging interpolation algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110871465.4A CN113590587A (en) | 2021-07-30 | 2021-07-30 | Offline position fingerprint database construction method based on self-adaptive simulated annealing-particle swarm-kriging interpolation algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113590587A true CN113590587A (en) | 2021-11-02 |
Family
ID=78252654
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110871465.4A Pending CN113590587A (en) | 2021-07-30 | 2021-07-30 | Offline position fingerprint database construction method based on self-adaptive simulated annealing-particle swarm-kriging interpolation algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113590587A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114422952A (en) * | 2022-01-29 | 2022-04-29 | 南京邮电大学 | Indoor fingerprint positioning method based on improved LSSVR |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018072351A1 (en) * | 2016-10-20 | 2018-04-26 | 北京工业大学 | Method for optimizing support vector machine on basis of particle swarm optimization algorithm |
| CN110967761A (en) * | 2018-09-30 | 2020-04-07 | 中国石油化工股份有限公司 | Geostatistical stochastic inversion method and system based on quantum annealing algorithm |
| CN111125885A (en) * | 2019-12-03 | 2020-05-08 | 杭州电子科技大学 | ASF correction table construction method based on improved kriging interpolation algorithm |
| CN111260118A (en) * | 2020-01-10 | 2020-06-09 | 天津理工大学 | Vehicle networking traffic flow prediction method based on quantum particle swarm optimization strategy |
-
2021
- 2021-07-30 CN CN202110871465.4A patent/CN113590587A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018072351A1 (en) * | 2016-10-20 | 2018-04-26 | 北京工业大学 | Method for optimizing support vector machine on basis of particle swarm optimization algorithm |
| CN110967761A (en) * | 2018-09-30 | 2020-04-07 | 中国石油化工股份有限公司 | Geostatistical stochastic inversion method and system based on quantum annealing algorithm |
| CN111125885A (en) * | 2019-12-03 | 2020-05-08 | 杭州电子科技大学 | ASF correction table construction method based on improved kriging interpolation algorithm |
| CN111260118A (en) * | 2020-01-10 | 2020-06-09 | 天津理工大学 | Vehicle networking traffic flow prediction method based on quantum particle swarm optimization strategy |
Non-Patent Citations (2)
| Title |
|---|
| 刘辉元;马金辉;黄琼;: "基于改进克里金插值的室内定位位置指纹库构建方法", 重庆邮电大学学报(自然科学版), no. 06 * |
| 王存华;王伟;: "基于模拟退火优化BP算法的指纹地图构建方法", 国外电子测量技术, no. 03 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114422952A (en) * | 2022-01-29 | 2022-04-29 | 南京邮电大学 | Indoor fingerprint positioning method based on improved LSSVR |
| CN114422952B (en) * | 2022-01-29 | 2024-05-03 | 南京邮电大学 | Indoor fingerprint positioning method based on improved LSSVR |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111294921B (en) | RSSI wireless sensor network three-dimensional cooperative positioning method | |
| CN109672973B (en) | Indoor positioning fusion method based on strongest AP | |
| CN106851571B (en) | Decision tree-based rapid KNN indoor WiFi positioning method | |
| CN103561463A (en) | RBF neural network indoor positioning method based on sample clustering | |
| CN113596989B (en) | Indoor positioning method and system for intelligent workshop | |
| CN101977436B (en) | WLAN indoor positioning-based mobile subscriber position coordinate correction method | |
| Siyang et al. | WKNN indoor Wi-Fi localization method using k-means clustering based radio mapping | |
| CN113590587A (en) | Offline position fingerprint database construction method based on self-adaptive simulated annealing-particle swarm-kriging interpolation algorithm | |
| Chai et al. | A new indoor positioning algorithm of cellular and Wi-Fi networks | |
| CN113365214B (en) | Wireless sensor network node positioning method and device based on trilateral positioning improvement | |
| CN113056001B (en) | Differential correction weighted centroid positioning method based on hybrid filtering | |
| Gou et al. | Three-dimensional DV-hop localization algorithm based on hop size correction and improved sparrow search | |
| CN110662167A (en) | Indoor heterogeneous network cooperative positioning method and system and readable storage medium | |
| CN116709240A (en) | Hierarchical sensor deployment method based on whale optimization algorithm | |
| KR20120048375A (en) | Knn/pcm hybrid mehod using gath-geva method for indoor location determination in waln | |
| CN115099385A (en) | Spectrum map construction method based on sensor layout optimization and adaptive Kriging model | |
| CN110032070B (en) | Target tracking method of mobile wireless sensor network based on particle swarm fuzzy tree | |
| Minu et al. | Node localization in wireless sensor networks by artificial immune system | |
| Pan | Node localization method for massive sensor networks based on clustering particle swarm optimization in cloud computing environment | |
| CN111447579B (en) | DV-hop indoor positioning method based on RSSI average hop distance and path loss | |
| Jegede et al. | A genetic algorithm for node localization in wireless sensor networks | |
| CN114745674B (en) | Ranging model positioning algorithm based on improved BP neural network | |
| Xianhao et al. | Node self-localization algorithm based on rssi in wireless sensor networks outdoor | |
| Liu et al. | A novel dimension reduction algorithm for fingerprint positioning based on GrDOP and geometric constraints | |
| Li | Weighted centroid localization algorithm based on mea-bp neural network and dbscan clustering |
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 |