CN113612555A - Intelligent calibration algorithm and system based on mobile terminal wireless radio frequency signal intensity - Google Patents
Intelligent calibration algorithm and system based on mobile terminal wireless radio frequency signal intensity Download PDFInfo
- Publication number
- CN113612555A CN113612555A CN202110952490.5A CN202110952490A CN113612555A CN 113612555 A CN113612555 A CN 113612555A CN 202110952490 A CN202110952490 A CN 202110952490A CN 113612555 A CN113612555 A CN 113612555A
- Authority
- CN
- China
- Prior art keywords
- calibration
- mobile terminal
- neural network
- value
- calibration model
- 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
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/309—Measuring or estimating channel quality parameters
- H04B17/318—Received signal strength
-
- 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]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/10—Monitoring; Testing of transmitters
- H04B17/11—Monitoring; Testing of transmitters for calibration
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/20—Monitoring; Testing of receivers
- H04B17/21—Monitoring; Testing of receivers for calibration; for correcting measurements
-
- 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
- H04W64/006—Locating users or terminals or network equipment for network management purposes, e.g. mobility management with additional information processing, e.g. for direction or speed determination
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computing Systems (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Electromagnetism (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Monitoring And Testing Of Transmission In General (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention provides an intelligent calibration algorithm and system based on the wireless radio frequency signal intensity of a mobile terminal, wherein a calibration model base is established according to an improved BP neural network calibration model obtained by a proposed nonlinear convergence factor, and the model number of the terminal and corresponding calibration model parameters are recorded and stored; judging whether the model of the mobile terminal is in a calibration model library, if not, taking the original RSSI observed values of all APs acquired by the standard mobile terminal at all acquisition points as standard sampling data, and taking the original RSSI observed values of all corresponding APs of the mobile terminal at corresponding acquisition points as test sampling data; and taking an original RSSI observation value received by the mobile terminal at any indoor position as a calibration model input, processing the original RSSI observation value by an improved BP neural network calibration model, and taking the final output as a calibration value. The invention can avoid the algorithm from falling into local optimum, has wide applicability, can finish the calibration work only by a handheld mobile terminal, and effectively eliminates the heterogeneous difference of software and hardware of different mobile terminals.
Description
Technical Field
The invention relates to the field of wireless radio frequency signal intensity observed value calibration, in particular to an intelligent calibration algorithm and system based on mobile terminal wireless radio frequency signal intensity.
Background
In the field of indoor positioning, the environment sensing can be carried out by utilizing the multipath effect of the wireless radio frequency signal propagation of the mobile terminal, and fingerprint positioning is carried out by acquiring the signal characteristics. However, since software and hardware of different mobile terminals are different, the observed values of the radio frequency signal strength received by different mobile terminals at the same position have larger difference, which results in that fingerprint positioning is not available. In order to solve the above problem, the radio frequency signal intensity observed value needs to be calibrated, so that the radio frequency signal intensity observed values received by other mobile terminals and the standard mobile terminal at the same position tend to be consistent.
The invention provides an intelligent calibration algorithm and system based on mobile terminal wireless radio frequency signal intensity, and based on the intelligent calibration algorithm and system, in order to further accelerate establishment and optimization of the calibration algorithm and avoid falling into local optimization, the invention further provides a BP neural network calibration model with an improved nonlinear convergence factor a, so that radio frequency signal intensity observed values received by other mobile terminals and a standard mobile terminal at the same position tend to be consistent. A user can finish the calibration work only by holding the mobile terminal, the software and hardware heterogeneous difference of each mobile terminal is effectively eliminated, the operation is simple, the adaptability to the environment can be improved, the expandability is strong, and the applicability is wide.
Disclosure of Invention
The invention aims to solve the problems that in the prior art, the observed values of the received radio frequency signal strength are greatly different due to the fact that software and hardware of each mobile terminal are heterogeneous, and a traditional calibration method is complex in step, narrow in application range and low in expandability.
In order to solve the technical problem, the invention provides an intelligent calibration algorithm and system based on the wireless radio frequency signal intensity of mobile terminals, which can establish a corresponding improved BP neural network calibration model for each mobile terminal, wherein the method comprises the following steps:
establishing a calibration model library according to an improved BP neural network calibration model obtained by the proposed nonlinear convergence factor, and recording and storing the terminal model and corresponding calibration model parameters;
before a certain mobile terminal is held by hand for calibration, judging whether the model of the terminal is in a calibration model library;
if not, taking all original radio frequency signal intensity observed values of all APs of the standard mobile terminal on all acquisition points as standard sampling data, and taking all original radio frequency signal intensity observed values of all corresponding APs of the mobile terminal on corresponding acquisition points as test sampling data; establishing and training an improved BP neural network calibration model according to the standard sampling data and the test sampling data, and recording and storing the model number of the terminal and corresponding calibration model parameters;
if so, calibrating by directly utilizing the calibration model parameters corresponding to the calibration model library;
when the mobile terminal is held by hand for calibration, an original radio frequency signal intensity observation value received at any indoor position is used as an input of a calibration model, the calibration model is processed by an improved BP neural network calibration model, and finally obtained output is used as a calibration value to finish intelligent calibration of the mobile terminal.
An intelligent calibration algorithm based on the wireless radio frequency signal intensity of a mobile terminal utilizes an improved BP neural network calibration model to establish a calibration model library, and comprises the following steps:
step 1: randomly selecting a plurality of acquisition points indoors, acquiring the radio frequency signal intensity on all the acquisition points by a standard mobile phone, comprehensively representing the radio frequency signal intensity as initial standard sampling data, taking all original radio frequency signal intensity observed values of all corresponding APs on the corresponding acquisition points by a mobile terminal as test sampling data, taking the test sampling data as a real input value of an improved BP neural network calibration model, taking the initial standard sampling data as a real output value of the improved BP neural network calibration model, and randomly initializing each layer weight of the neural networkThreshold valueAs initial whale flock position vector XiSetting the whale population size N, setting the current whale population iteration number t to be 0, and setting the maximum iteration number t of the whale populationmaxCurrent BP neural network iteration number T, maximum BP neural network iteration number Tmax;
Step 2: when the iteration times t of the whale population are less than the maximum iteration times t of the whale populationmaxThen, the fitness value f (X) of each whale was calculatedi) Finding out the best fitness and the corresponding optimal whale position Xbest;
Wherein, yiThe real value of the ith RSSI of the standard sampling data is obtained, y is the predicted value of the ith RSSI of the test sampling data, and n is the number of samples;
and step 3: in order to accelerate the establishment and optimization of a calibration algorithm and improve the updating iteration speed, a nonlinear convergence factor a is provided for simulating the shrinkage behavior of a surrounding prey, and the convergence factor a only dynamically changes along with the current iteration time t, so that the situation that the surrounding prey falls into local optimum can be effectively avoided;
wherein t is the iteration times of the current whale population, and tmaxThe maximum iteration number of the whale population is obtained; updating the whale location parameter A, C;
A=2ar-a (3)
C=2r (4)
wherein r is a random number of [0,1 ];
and 4, step 4: randomly generating probability p, judging whether p is less than 0.5, if p is more than or equal to 0.5, updating the contraction surrounding position:
wherein t is the iteration times of the current whale population; xbestIs the optimal whale position; xiIs the current whale position; a and C are coefficient vectors obtained in the step 3;
if p <0.5, and spiral walk when | A | < 1:
wherein the content of the first and second substances,representing the distance of the current whale from the optimal position; b is a constant and defines the shape of a logarithmic spiral; l is [ -1,1 [ ]]A medium random number;
when | A | ≧ 1, random walk is performed according to the following formula:
wherein, XrandIs a randomly selected position vector;
and 5: the iteration times t of the whale population are increased automatically, and the optimal position is updated according to the comparison in the step 2; when reaching the maximum iteration number t of whale populationmaxThen, output XbestI.e. the optimal weight wijThreshold value thetajAnd is used as the optimal initial parameter of the BP neural network;
step 6: the BP neural network carries out the forward propagation process and passes the connection weight w between the neuronsijAnd neuron threshold θjAnd processing data, and obtaining a predicted output value by adopting a nonlinear Sigmoid activation function.
Wherein, wijThe connection weight from the neuron i to the neuron j is obtained; thetajIs neuron j threshold; i isjInputting a value for neuron j; o isjOutputting a value for neuron j; RSSIx,iIs the input value of neuron i.
And 7: the BP neural network carries out an error back propagation process, obtains a predicted output value through the forward propagation process, and obtains a loss function E of the current iteration times according to the difference between the predicted output value of the intelligent equipment and the real output value of the standard intelligent equipmentjAnd reversely transmitting the error to the neuron on the upper layer to obtain the error on the layer, transmitting the error layer by layer until the hidden layer on the uppermost layer, and continuously adjusting the connection weight and the threshold value based on a gradient descent method.
Wherein the RSSIy,jTrue output values for neuron j; RSSI'y,jPredicting an output value for output layer neuron j; w'ijIs the updated weight value; theta'jIs an updated threshold; η ∈ (0,1) is a learning rate, and if the value is large, convergence is fast but local optimum is likely to be achieved, and if the value is small, convergence is slow but global optimum is approached.
And 8: after repeated learning and training, when the current iteration number T of the BP neural network reaches the maximum iteration number T of the BP neural networkmaxThen, selecting a loss function EjAnd taking the minimum BP neural network as a final calibration model, and storing the parameters of the current improved BP neural network calibration model.
And step 9: and (4) repeating the steps 1 to 8 with a plurality of mobile terminals, and further establishing a calibration model library.
The invention also provides an intelligent calibration system based on the wireless radio frequency signal intensity of the mobile terminal, wherein the system comprises:
the data acquisition module takes all original radio frequency signal intensity observed values of all APs of the standard mobile terminal on all acquisition points as standard sampling data, and takes all original radio frequency signal intensity observed values of all corresponding APs of other mobile terminals on corresponding acquisition points as test sampling data;
the model establishing module is used for establishing and training an improved BP neural network calibration model according to the standard sampling data and the test sampling data, and recording and storing the terminal model and corresponding calibration model parameters;
and the intelligent calibration module takes the radio frequency signal value received by the mobile terminal at any indoor position as the input of a calibration model, processes the radio frequency signal value through the improved BP neural network calibration model, and finally obtains the output as a calibration value to finish the intelligent calibration of the mobile terminal.
Compared with the prior art, the intelligent calibration algorithm and system based on the wireless radio frequency signal intensity of the mobile terminal, provided by the invention, can avoid the algorithm from falling into local optimization, improve the adaptability to the environment, have strong expandability and wide applicability, can finish the calibration work only through the handheld mobile terminal, effectively eliminate the software and hardware heterogeneous differences of different mobile terminals, and are simple to operate.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
FIG. 1 is a schematic block diagram of embodiment 1 of the present invention;
FIG. 2 is a flowchart of example 1 of the present invention;
fig. 3 is a schematic structural diagram of embodiment 2 of the present invention.
Detailed Description
To facilitate an understanding of the invention, the invention will now be described more fully with reference to the accompanying drawings. Preferred embodiments of the present invention are shown in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
Example 1
Referring to fig. 1 and fig. 2, for an intelligent calibration algorithm based on the wireless rf signal strength of a mobile terminal according to a first embodiment of the present invention, the specific implementation manner includes the following steps:
s101, establishing a calibration model library according to an improved BP neural network calibration model obtained by the proposed nonlinear convergence factor, and recording and storing the terminal model and corresponding calibration model parameters;
s102, before the handheld mobile terminal is calibrated, judging whether the model of the terminal is in a calibration model library, if not, taking all original radio frequency signal intensity observed values of all APs of a standard mobile terminal on all acquisition points as standard sampling data, and taking all original radio frequency signal intensity observed values of all corresponding APs of the mobile terminal on corresponding acquisition points as test sampling data;
s103, an original radio frequency signal intensity observed value received by the mobile terminal at any indoor position is used as an input of a calibration model, the original radio frequency signal intensity observed value is processed by the improved BP neural network calibration model, and finally obtained output is used as a calibration value to finish intelligent calibration of the mobile terminal.
The calibration algorithm is used for establishing a calibration model between a standard mobile terminal and other mobile terminals, one mobile terminal only has one corresponding calibration model at last and has no direct relation with acquisition points, namely sampling data of the calibration model can be acquired at one acquisition point or a plurality of acquisition points, and radio frequency signals refer to WiFi, Bluetooth and the like. The acquisition points in the standard sampling data and the test sampling data are in one-to-one correspondence, and the APs and all the RSSIs thereof are also in one-to-one correspondence.
In S102, whether the calibration model of the model exists in other mobile terminals is discussed in two cases:
a. if the calibration model of the mobile terminal model exists, jumping to step S103, before calibrating the mobile terminal, when judging that the terminal model exists in a calibration library, using an original radio frequency signal intensity observed value received by the mobile terminal at any indoor position as a calibration model input, processing the calibration model after the original radio frequency signal intensity observed value is processed by an improved BP neural network calibration model, and finally obtaining an output as a calibration value to finish the intelligent calibration of the mobile terminal;
b. if the terminal model is not in the calibration library, all original radio frequency signal intensity observed values of all APs of a standard mobile terminal on all acquisition points are used as standard sampling data, all original radio frequency signal intensity observed values of all APs of the terminal on corresponding acquisition points are used as test sampling data, then the step S101 is skipped to, namely an improved BP neural network calibration model is obtained by establishing and training according to the standard sampling data and the test sampling data, the terminal model and corresponding calibration model parameters are recorded and stored, and finally the step S103 is skipped to complete the intelligent calibration of the mobile terminal.
It should be noted here that the acquired radio frequency signal is a signal source existing in the indoor environment, and an AP does not need to be specially arranged again.
In addition, in order to ensure that the standard sampling data and the test sampling data are stable and reliable, the sampling time is set to be 10 minutes and the sampling frequency is set to be 5 seconds at each acquisition point.
Further, the original data transmission format of the standard sample data obtained from the standard mobile terminal or the test sample data obtained from the other mobile terminals is:
{P1{(AP11,RSSI1,...,RSSIi,...,RSSIk),...,(AP1n,RSSI1,...,RSSIi,...,RSSIk)},
......
Pi{(APi1,RSSI1,...,RSSIi,...,RSSIk),...,(APin,RSSI1,...,RSSIi,...,RSSIk)},
......
Pj{(APj1,RSSI1,...,RSSIi,...,RSSIk),...,(APjn,RSSI1,...,RSSIi,...,RSSIk)}}
wherein, PiIs the ith acquisition point position, i is 1,2inIs at Pi(n) th AP, RSSI received at the acquisition pointkAnd obtaining the k-th observed value of the original radio frequency signal intensity.
S101, establishing a calibration model base based on the improved BP neural network calibration model obtained according to the provided nonlinear convergence factor, and recording and storing the terminal model and the corresponding calibration model parameters.
In this step, the specific implementation process is as follows:
step 1: randomly selecting a plurality of acquisition points indoors (if a plurality of environments such as corridors, rooms, staircases and the like exist indoors, a plurality of acquisition points are selected to represent the scene under each scene, and the indoor environment is represented comprehensively), acquiring the radio frequency signal intensity on all the acquisition points by using a standard mobile phone, and comprehensively representing the radio frequency signal intensity as initial standard sampling data (for example, n1 data are acquired at a point, n2 data are acquired at a point b, and n1+ n2 data are comprehensive), and taking all original radio frequency signal intensity observed values of all corresponding APs on the corresponding acquisition points of the mobile terminal as test sampling data;
randomly initializing each layer weight of the neural network by taking the test sampling data as the real input value of the improved BP neural network calibration model and the initial standard sampling data as the real output value of the improved BP neural network calibration modelThreshold valueAs initial whale flock position vector XiSetting the whale population size N, setting the current whale population iteration time t to be 0, and setting the maximum iteration time t of the whale populationmaxCurrent BP neural network iteration number T, maximum BP neural network iteration number Tmax;
Step 2: calculate fitness value f (X) of each whalei) Finding out the best fitness and the corresponding optimal whale position Xbest;
Wherein, yiThe real value of the ith RSSI of the standard sampling data is obtained, y is the predicted value of the ith RSSI of the test sampling data, and n is the number of samples;
and step 3: in order to accelerate the establishment and optimization of a calibration algorithm and improve the updating iteration speed, a nonlinear convergence factor a is provided for simulating the contraction behavior of a surrounding prey, the convergence factor a only dynamically changes along with the current iteration time t, so that the situation that the prey is locally optimal can be effectively avoided, and whale position parameters a, A and C are updated:
A=2ar-a (3)
C=2r (4)
wherein t is the iteration times of the current whale population, and tmaxR is [0,1] for the maximum number of iterations]A random number of (2);
and 4, step 4: randomly generating probability p, judging whether p is less than 0.5, if p is more than or equal to 0.5, updating the contraction surrounding position:
wherein t is the iteration times of the current whale population; xbestIs the optimal whale position; xiIs the current whale position; a and C are coefficient vectors obtained in the step 3;
if p <0.5, and spiral walk when | A | < 1:
wherein the content of the first and second substances,representing the distance of the current whale from the optimal position; b is a constant and defines the shape of a logarithmic spiral; l is [ -1,1 [ ]]A medium random number;
when | A | ≧ 1, random walk is performed according to the following formula:
wherein, XrandIs a randomly selected position vector;
and 5: the iteration times t of the whale population are increased automatically, and the optimal position is updated according to the comparison in the step 2; when reaching the maximum iteration number t of whale populationmaxThen, output XbestI.e. the optimal weight wijThreshold value thetajAnd is used as the optimal initial parameter of the BP neural network;
step 6: the BP neural network carries out the forward propagation process and passes the connection weight w between the neuronsijAnd neuron threshold θjAnd processing data, and obtaining a predicted output value by adopting a nonlinear Sigmoid activation function.
Wherein, wijThe connection weight from the neuron i to the neuron j is obtained; thetajIs neuron j threshold; i isjInputting a value for neuron j; o isjOutputting a value for neuron j; RSSIx,iIs the input value of neuron i.
And 7: the BP neural network carries out an error back propagation process, obtains a predicted output value through the forward propagation process, and obtains a loss function E of the current iteration times according to the difference between the predicted output value of the intelligent equipment and the real output value of the standard intelligent equipmentjAnd reversely transmitting the error to the neuron on the upper layer to obtain the error on the layer, transmitting the error layer by layer until the hidden layer on the uppermost layer, and continuously adjusting the connection weight and the threshold value based on a gradient descent method.
Wherein the RSSIy,jTrue output values for neuron j; RSSI'y,jPredicting an output value for output layer neuron j; w'ijIs the updated weight value; theta'jIs an updated threshold; η ∈ (0,1) is a learning rate, and if the value is large, convergence is fast but local optimum is likely to be achieved, and if the value is small, convergence is slow but global optimum is approached.
And 8: after repeated learning and training, when the iteration number T of the BP neural network reaches the maximum iteration number T of the BP neural networkmaxThen, selecting a loss function EjAnd taking the minimum BP neural network as a final calibration model, and storing the parameters of the current improved BP neural network calibration model.
And step 9: and (4) repeating the steps 1 to 8 with a plurality of mobile terminals, and further establishing a calibration model library.
When a pedestrian holds the mobile terminal and needs to calibrate, firstly, the system can automatically match the model of the terminal in a calibration database of the server, if the model exists, the system can hold the mobile terminal to calibrate the received radio frequency signal intensity observed value at any indoor position, otherwise, test sampling data are collected at a collection point, and calibration is carried out after an improved BP neural network calibration model of the terminal is established.
It should be added that any indoor position may be an acquisition point of the sampled data, or may be other positions, and the mobile terminal receives the original radio frequency signal intensity observation values of all APs at any position and performs calibration immediately.
Example 2
Referring to fig. 3, an intelligent calibration system based on wireless rf signal strength of a mobile terminal according to a second embodiment of the present invention includes a data acquisition module 11, a model building module 12, and an intelligent calibration module 13, which are connected in sequence;
the data acquisition module 11 is specifically configured to:
taking all original radio frequency signal intensity observed values of all APs of a standard mobile terminal on all acquisition points as standard sampling data, and taking all original radio frequency signal intensity observed values of all corresponding APs of the mobile terminal on corresponding acquisition points as test sampling data;
wherein the model building module 12 is specifically configured to:
establishing and training an improved BP neural network calibration model according to the standard sampling data and the test sampling data, and recording and storing the model number of the terminal and corresponding calibration model parameters;
the intelligent calibration module 13 is specifically configured to:
the radio frequency signal value received by the mobile terminal at any indoor position is used as an input of a calibration model, the radio frequency signal value is processed by the improved BP neural network calibration model, and finally obtained output is used as a calibration value to finish intelligent calibration of the mobile terminal.
Those skilled in the art will appreciate that all or part of the steps in the method for implementing the above embodiments may be implemented by a program instructing the relevant hardware. The program may be stored in a computer-readable storage medium. Which when executed comprises the steps of the method described above. The storage medium includes: ROM/RAM, magnetic disk, optical disk, etc.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (3)
1. Intelligent calibration algorithm based on mobile terminal wireless radio frequency signal intensity is characterized in that: the calibration algorithm establishes a corresponding improved BP neural network calibration model for each mobile terminal, and comprises the following steps:
s1, establishing a calibration model library according to the improved BP neural network calibration model obtained by the proposed nonlinear convergence factor, and recording and storing the terminal model and the corresponding calibration model parameters;
s2, before a certain mobile terminal is held by hand for calibration, judging whether the terminal model is in a calibration model library;
if not, taking all original radio frequency signal intensity observed values of all APs of the standard mobile terminal on all acquisition points as standard sampling data, taking all original radio frequency signal intensity observed values of all APs of the mobile terminal on corresponding acquisition points as test sampling data, taking the test sampling data as a real input value of the improved BP neural network calibration model, taking the standard sampling data as a real output value of the improved BP neural network calibration model, repeating the step S1 to establish and train to obtain calibration model parameters of the terminal model in the improved BP neural network calibration model, and then entering the step S3;
if yes, directly entering step S3 to calibrate by using the calibration model parameters corresponding to the calibration model library;
s3, when the mobile terminal is held by hand for calibration, the original radio frequency signal intensity observed value received at any indoor position is used as the input of a calibration model, the calibration model is processed by the improved BP neural network calibration model, and the finally obtained output is used as the calibration value to finish the intelligent calibration of the mobile terminal.
2. The intelligent calibration algorithm based on the wireless radio frequency signal strength of the mobile terminal according to claim 1, wherein the method for establishing the calibration model library according to the improved BP neural network calibration model obtained from the proposed nonlinear convergence factor comprises the following steps:
s1-1: a plurality of acquisition points are randomly selected indoors, the standard mobile phone acquires the radio frequency signal intensity on all the acquisition points and comprehensively expresses the radio frequency signal intensity as initial standard sampling data,taking all original radio frequency signal intensity observed values of all corresponding APs on a corresponding acquisition point of a mobile terminal as test sampling data, taking the test sampling data as a real input value of an improved BP neural network calibration model, and taking the initial standard sampling data as a real output value of the improved BP neural network calibration model; randomly initializing weights of each layer of neural networkThreshold valueAs initial whale flock position vector XiSetting the whale population size N, setting the current whale population iteration time t to be 0, and setting the maximum iteration time t of the whale populationmaxCurrent BP neural network iteration number T, maximum BP neural network iteration number Tmax;
S1-2: when the current iteration times t of the whale population is less than the maximum iteration times t of the whale populationmaxThen, the fitness value f (X) of each whale was calculatedi) Finding out the best fitness and the corresponding optimal whale position Xbest:
In the formula, yiThe real value of the ith RSSI of the standard sampling data is obtained, y is the predicted value of the ith RSSI of the test sampling data, and n is the number of samples;
s1-3: in order to accelerate the establishment and optimization of a calibration algorithm and improve the updating iteration speed, a nonlinear convergence factor a is provided for simulating the shrinkage behavior of a surrounding prey, the convergence factor a only dynamically changes along with the current iteration time t, the algorithm can be effectively prevented from falling into local optimum, and the calculation formula of the nonlinear convergence factor a is as follows:
wherein t is the iteration number of the current whale population, and tmaxThe maximum iteration number of the whale population is obtained; the whale location parameter A, C is updated as follows:
A=2ar-a (3)
C=2r (4)
wherein r is a random number of [0,1 ];
s1-4: randomly generating probability p, judging whether p is less than 0.5, if p is more than or equal to 0.5, updating the contraction surrounding position:
in the formula, t is the iteration times of the current whale population; xbestIs the optimal whale position; xiIs the current whale position; a and C are coefficient vectors obtained in the step S1-3;
if p <0.5, and when | A | <1, a helical walk is performed:
in the formula (I), the compound is shown in the specification,representing the distance of the current whale from the optimal position; b is a constant and defines the shape of a logarithmic spiral; l is [ -1,1 [ ]]A medium random number;
when | A | ≧ 1, random walk is performed according to equation (6):
in the formula, XrandIs a randomly selected position vector;
s1-5: the iteration times t of the whale population are increased automatically, and the optimal position is updated according to the comparison in the step S1-2; maximum iteration when whale population is reachedNumber of times tmaxThen, output XbestI.e. the optimal weight wijThreshold value thetajAnd is used as the optimal initial parameter of the BP neural network;
s1-6: the BP neural network carries out the forward propagation process and passes the connection weight w between the neuronsijAnd neuron threshold θjAnd (3) processing data, and obtaining a predicted output value by adopting a nonlinear Sigmoid activation function:
in the formula, wijThe connection weight from the neuron i to the neuron j is obtained; thetajIs neuron j threshold; i isjInputting a value for neuron j; o isjOutputting a value for neuron j; RSSIx,iIs the input value of neuron i;
s1-7: the BP neural network carries out an error back propagation process, obtains a predicted output value through the forward propagation process, and obtains a loss function E of the current iteration times according to the difference between the predicted output value of the intelligent equipment and the real output value of the standard intelligent equipmentjAnd reversely propagating the error to the neuron at the upper layer to obtain the error at the layer, transmitting the error layer by layer until the hidden layer at the uppermost layer, and continuously adjusting the connection weight and the threshold value based on a gradient descent method:
wherein RSSIy,jTrue output values for neuron j; RSSI'y,jPredicting an output value for output layer neuron j; w'ijIs the updated weight value; theta'jIs an updated threshold; eta belongs to (0,1) as a learning rate, if the value is larger, the convergence is fast but the local optimum is easy to fall into, and if the value is smaller, the convergence is slow but the global optimum is approached;
s1-8: after repeated learning and training, when the iteration number T of the BP neural network reaches the maximum iteration number T of the BP neural networkmaxThen, selecting a loss function EjThe minimum BP neural network is used as a final calibration model, and the parameters of the current improved BP neural network calibration model and the corresponding mobile terminal model are saved;
s1-9: and repeating the steps S1-1 to S1-8 on a plurality of mobile terminals, and further establishing a standard model library.
3. Intelligent calibration system based on mobile terminal radio frequency signal intensity, its characterized in that: the calibration system comprises a data acquisition module, a model establishment module and an intelligent calibration module;
the data acquisition module is used for taking all original radio frequency signal intensity observed values of all APs of a standard mobile terminal on all acquisition points as standard sampling data, and taking all original radio frequency signal intensity observed values of all corresponding APs of other mobile terminals on corresponding acquisition points as test sampling data;
the model establishing module is used for establishing and training an improved BP neural network calibration model according to the standard sampling data and the test sampling data, and recording and storing the model number of the terminal and corresponding calibration model parameters;
the intelligent calibration module is used for taking a radio frequency signal value received by the mobile terminal at any indoor position as a calibration model input, processing the radio frequency signal value through the improved BP neural network calibration model, and taking the finally obtained output as a calibration value to finish the intelligent calibration of the mobile terminal.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110952490.5A CN113612555B (en) | 2021-08-19 | 2021-08-19 | Intelligent calibration method and system based on mobile terminal wireless radio frequency signal intensity |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110952490.5A CN113612555B (en) | 2021-08-19 | 2021-08-19 | Intelligent calibration method and system based on mobile terminal wireless radio frequency signal intensity |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113612555A true CN113612555A (en) | 2021-11-05 |
CN113612555B CN113612555B (en) | 2022-06-07 |
Family
ID=78341189
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110952490.5A Active CN113612555B (en) | 2021-08-19 | 2021-08-19 | Intelligent calibration method and system based on mobile terminal wireless radio frequency signal intensity |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113612555B (en) |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130165143A1 (en) * | 2011-06-24 | 2013-06-27 | Russell Ziskind | Training pattern recognition systems for determining user device locations |
US20190150006A1 (en) * | 2017-11-15 | 2019-05-16 | Futurewei Technologies, Inc. | Predicting received signal strength in a telecommunication network using deep neural networks |
CN109948791A (en) * | 2017-12-21 | 2019-06-28 | 河北科技大学 | Utilize the method for genetic algorithm optimization BP neural network and its application in positioning |
US10655971B1 (en) * | 2019-05-01 | 2020-05-19 | Mapsted Corp. | Maintaining a trained neural network for mobile device RSS fingerprint based indoor navigation |
US20200259570A1 (en) * | 2020-04-29 | 2020-08-13 | Intel Corporation | Indoor localization with beacon technology based on signal strength distribution and deep learning techniques |
WO2021007812A1 (en) * | 2019-07-17 | 2021-01-21 | 深圳大学 | Deep neural network hyperparameter optimization method, electronic device and storage medium |
CN112333652A (en) * | 2019-07-16 | 2021-02-05 | ***通信集团浙江有限公司 | WLAN indoor positioning method and device and electronic equipment |
CN112700060A (en) * | 2021-01-08 | 2021-04-23 | 佳源科技股份有限公司 | Station terminal load prediction method and prediction device |
CN113163484A (en) * | 2021-01-08 | 2021-07-23 | 广东工业大学 | Indoor positioning method |
-
2021
- 2021-08-19 CN CN202110952490.5A patent/CN113612555B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130165143A1 (en) * | 2011-06-24 | 2013-06-27 | Russell Ziskind | Training pattern recognition systems for determining user device locations |
US20190150006A1 (en) * | 2017-11-15 | 2019-05-16 | Futurewei Technologies, Inc. | Predicting received signal strength in a telecommunication network using deep neural networks |
CN109948791A (en) * | 2017-12-21 | 2019-06-28 | 河北科技大学 | Utilize the method for genetic algorithm optimization BP neural network and its application in positioning |
US10655971B1 (en) * | 2019-05-01 | 2020-05-19 | Mapsted Corp. | Maintaining a trained neural network for mobile device RSS fingerprint based indoor navigation |
CN112333652A (en) * | 2019-07-16 | 2021-02-05 | ***通信集团浙江有限公司 | WLAN indoor positioning method and device and electronic equipment |
WO2021007812A1 (en) * | 2019-07-17 | 2021-01-21 | 深圳大学 | Deep neural network hyperparameter optimization method, electronic device and storage medium |
US20200259570A1 (en) * | 2020-04-29 | 2020-08-13 | Intel Corporation | Indoor localization with beacon technology based on signal strength distribution and deep learning techniques |
CN112700060A (en) * | 2021-01-08 | 2021-04-23 | 佳源科技股份有限公司 | Station terminal load prediction method and prediction device |
CN113163484A (en) * | 2021-01-08 | 2021-07-23 | 广东工业大学 | Indoor positioning method |
Also Published As
Publication number | Publication date |
---|---|
CN113612555B (en) | 2022-06-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109523017B (en) | Gesture detection method, device, equipment and storage medium | |
US9542938B2 (en) | Scene recognition method, device and mobile terminal based on ambient sound | |
US8738032B2 (en) | Hybrid location using a weighted average of location readings and signal strengths of wireless access points | |
US8774830B2 (en) | Training pattern recognition systems for determining user device locations | |
CN110166154B (en) | Software radio frequency spectrum monitoring and identifying method based on neural network | |
WO2020098256A1 (en) | Speech enhancement method based on fully convolutional neural network, device, and storage medium | |
CN109993938A (en) | Fire monitoring method, apparatus, equipment and medium | |
CN112874469B (en) | Vehicle unlocking and locking control method and device | |
CN110349185B (en) | RGBT target tracking model training method and device | |
CN107113764A (en) | The method and apparatus for improving artificial neural network positioning performance | |
CN109657539B (en) | Face value evaluation method and device, readable storage medium and electronic equipment | |
US8214312B2 (en) | System and method for calibrating radio frequency power of communication devices | |
CN112348100B (en) | Rock recognition method, device, equipment and storage medium | |
CN111310902A (en) | Training method of neural network model, slope displacement prediction method and related device | |
CN113435568A (en) | Computer-implemented method and electronic system for training neural networks | |
Cui et al. | Improved genetic algorithm to optimize the Wi-Fi indoor positioning based on artificial neural network | |
CN113612555B (en) | Intelligent calibration method and system based on mobile terminal wireless radio frequency signal intensity | |
CN112738709B (en) | Particle swarm search indoor positioning method and system based on UWB signals | |
CN117037258B (en) | Face image detection method and device, storage medium and electronic equipment | |
CN112766402A (en) | Algorithm selection method and device and electronic equipment | |
CN113065513A (en) | Method, device and equipment for optimizing self-training confidence threshold of intelligent camera | |
CN110660399A (en) | Training method and device for voiceprint recognition, terminal and computer storage medium | |
CN115567871A (en) | WiFi fingerprint indoor floor identification and position estimation method | |
CN113676857B (en) | Channel state information feature extraction method for indoor Wi-Fi positioning | |
CN113408571B (en) | Image classification method and device based on model distillation, storage medium and terminal |
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 |