CN115002703A - Passive indoor people number detection method based on Wi-Fi channel state information - Google Patents
Passive indoor people number detection method based on Wi-Fi channel state information Download PDFInfo
- Publication number
- CN115002703A CN115002703A CN202210834030.7A CN202210834030A CN115002703A CN 115002703 A CN115002703 A CN 115002703A CN 202210834030 A CN202210834030 A CN 202210834030A CN 115002703 A CN115002703 A CN 115002703A
- Authority
- CN
- China
- Prior art keywords
- csi
- data
- amplitude
- people
- signals
- 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
-
- 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
-
- 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
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/80—Services using short range communication, e.g. near-field communication [NFC], radio-frequency identification [RFID] or low energy communication
-
- 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)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a passive indoor people number detection method based on Wi-Fi channel state information, which comprises the following steps: step 1, collecting CSI original signals of different numbers of volunteers in the process of slowly walking at different positions; step 2, carrying out data preprocessing on the CSI signals acquired in the step 1; step 3, carrying out minimum-maximum normalization processing on the amplitude obtained in the step 2 to obtain standardized data; step 4, performing feature extraction and PCA dimension reduction on the obtained standardized data based on a time window method, thereby constructing a feature vector, formulating a corresponding label and constructing a data set; and 5, randomly dividing the obtained characteristic vector data set into a training set and a test set according to a proportion, training the training set by utilizing a probability density function-based method to improve a BP neural network, and verifying by using the test set to obtain a person number detection model capable of accurately detecting the number of people in the room. The invention has low cost, easy deployment and strong expansibility.
Description
Technical Field
The invention belongs to the technical field of realizing perception by using common wireless equipment, and relates to a passive indoor people number detection method based on Wi-Fi channel state information.
Background
With the development of science and technology, people perceive and explore the world more and more abundantly, various wireless sensing technologies emerge endlessly, and wireless signals not only can transmit data, but also can infer and perceive the change of the surrounding environment. Wi-Fi signals are ubiquitous, have the advantages of wide coverage range, insensitivity to illumination and shelters, privacy protection and the like, further promote the development of the human body perception technology, and research in the fields of smart homes, smart internet of things, smart buildings and the like is being conducted all over the world. The indoor crowd counting is a key step of multi-target perception, and has wide application prospects in the fields of tourist appreciation, crowd control and the like.
The traditional people number detection method is mostly based on computer vision technology, and people number counting is carried out by using a camera or special hardware equipment. The technology is widely applied to hospitals, schools and other practical scenes at present, although the visual perception technology based on the camera is achieved with great success, the camera perception technology has short boards which are limited in monitoring range, greatly influenced by light-receiving line components and easy to be shielded, and meanwhile, the Ultra-Wide Band (UWB) biological radar technology is also applied to the research of the people perception technology, but in the research, extra equipment is needed to operate a system, and the problems of privacy disclosure and the like exist. Special hardware systems such as WiTrack compute TOF (Time-of-Flight) through FMCW, providing coarse tracking of body parts but requiring the user to stay within direct line of sight of the device. Some researchers infer the number of people talking based on a smartphone platform, but in an indoor environment, it is inconvenient for the elderly and children to choose to equip with an FMCW signal generator and a mobile phone device. Wi-Fi, a wireless network based on the IEEE 802.11 protocol, has been deployed in homes and in public places. Conventional wireless sensing uses Received Signal Strength (RSS) to identify the number of people in a detected area. The RSSI is coarse-grained information, has a multipath effect, can be influenced by refraction, reflection, diffraction and the like of signals by an indoor environment, causes unstable positioning performance, is only suitable for message-level MAC layer information, and has the problems of insufficient detection accuracy, reliability, stability and the like. Meanwhile, sufficient recognition capability and robustness cannot be provided in a complicated indoor environment, and thus a counting error exists. Channel State Information (CSI) is used as a Wi-Fi physical layer signal, consists of amplitude and phase of subcarriers, provides richer data than RSSI of a Wi-Fi link layer, is a high-quality Wi-Fi signal sensing source, can obtain amplitude-frequency responses of a plurality of subcarriers, and is very important for realizing reliable and high data rate of multi-Channel wireless communication systems such as Wi-Fi and the like. Multiple Input Multiple Output (MIMO) techniques (e.g., IEEE 802.11n, 3GPP LTE, and mobile WiMAX systems) can suppress significant fading caused by CSI obtained by different subcarriers, and can improve data throughput and transmission distance without increasing bandwidth and transmission power, thereby improving communication quality. Therefore, the CSI has both the frequency diversity provided by OFDM and the spatial diversity provided by MIMO, can provide more environmental information about people's presence and movement, and is more suitable for people detection than RSSI data.
The invention aims to identify the indoor number of people with fine granularity, and has better robustness to the activities of people at different indoor positions under the condition of small indoor number scale.
Disclosure of Invention
The invention provides a passive indoor people number detection method based on Wi-Fi channel state information, which is characterized by comprising the following steps:
step 2, carrying out data preprocessing on the CSI signals acquired in the step 1;
step 3, carrying out minimum-maximum normalization processing on the amplitude obtained in the step 2 to obtain standardized data;
step 4, performing feature extraction and PCA dimension reduction on the standardized data obtained in the step 3 based on a time window method, so as to construct a feature vector, formulate a corresponding label and construct a data set;
and 5, randomly dividing the characteristic vector data set obtained in the step 4 into a training set and a test set according to a proportion, training the training set by improving a BP neural network by using a probability density function-based method, and verifying by using the test set to obtain a person number detection model capable of accurately detecting the number of people indoors.
Further, in the step 1, the CSI original amplitude signal is obtained specifically according to the following steps:
step 1.1, an Atheros AR9580 wireless network card is selected to collect CSI signals, two routers capable of supporting installation of OpenWRT firmware capable of receiving the CSI signals are utilized, one of the two routers serves as a transmitting end, and the other router serves as a receiving end.
Furthermore, the two routers belong to the same local area network, the two routers are controlled by the terminal to achieve data sending and receiving, the type of the router is TL-WDR4900, OpenWRT 15.05.1 is installed, and the packet sending frequency of the router is set to be 100 Hz.
Further, the method specifically comprises the steps of detecting and deleting abnormal points of the amplitude signal, and filtering amplitude noise in an original CSI observation value;
further, in step 2, the amplitude information after the preprocessing is obtained specifically according to the following steps:
and 2.1, reading the CSI original signals by using the CSI original signals of different numbers of volunteers in the step 1 and using a read _ log _ file function based on a Matlab software platform, wherein the CSI data is a two-dimensional matrix of N rows, and N is the number of data packets.
In the experiment, the transmitter has 3 transmitting antennas, the receiver has 3 receiving antennas, each receiving antenna has 56 subcarriers, and each signal sample contains 3 × 56 — 504 subcarrier CSI data, so that the obtained CSI data is a two-dimensional matrix of N × 540;
wherein the frequency response of the channel can be represented by the following formula:
wherein, H (f) i ) Represents a phase; h (f) i ) Is a center frequency of f i Channel state information, i.e., CSI, of the subcarriers of (a); n is a radical of s The number of subcarriers for a single antenna is 56 on the Atheros AR9580 wireless network card. Suppose that the number of antennas at the transmitting end is N tx The number of antennas at the receiving end is N rx Then the CSI data may be expressed as:
H ij CSI data for the ith antenna of a transmit antenna (TX) to the jth antenna of a receive antenna (RX), each antenna pair data comprising N s Sub-carriers, then H ij It can be specifically expressed as:
each signal sample containing N tx ×N rx ×N s Subcarrier CSI data;
step 2.2, abnormal point detection and removal are carried out on the CSI original signals read in the step 2.1;
and 2.3, removing the CSI signals of the abnormal amplitude points in the step 2.2, and filtering amplitude noise in the original CSI observation value by using a discrete wavelet transform method.
Further, in step 3, the normalized amplitude information is obtained specifically according to the following steps:
step 3.1, analyzing CSI signals of different numbers of volunteers walking slowly at different positions after amplitude data preprocessing in the step 2.3;
step 3.2, carrying out minimum-maximum normalization on the amplitude value preprocessed in the step 2, and mapping the data value between [0,1 ]; the specific formula is as follows:
the amplitude signal x after CSI noise reduction is normalized into an x' formula through the maximum and minimum values: where a represents the amplitude of the current subcarrier, and minA and maxA are the minimum and maximum values of the current subcarrier amplitude, respectively.
Further, in step 4, a feature vector of the amplitude is obtained and a corresponding class label is formulated specifically according to the following steps, and a data set is constructed:
step 4.1, selecting a proper time window size to provide a basis for constructing a feature vector in the next step;
step 4.2, according to the size of the time window selected in the step 4.1, taking the mean value, the range, the standard deviation and the root mean square of the amplitude as characteristic representation fluctuation conditions to form characteristic vectors;
featrure=[m,v,d,rs]
wherein m is a mean value, v is a range, d is a standard deviation, and rs is a root mean square value, and then constructing a feature vector;
and 4.3, carrying out PCA (principal component analysis) dimension reduction on the feature vector formed in the step 4.2, selecting the number of columns with obvious features, formulating corresponding category labels, and constructing a data set of the people number detection system.
Furthermore, in the step 5, a back propagation neural network improved by a method based on a probability density function is used for training the training set, and the number of people in the room can be accurately detected by establishing a people number detection model.
The invention aims at the problem that the activity states of people at different positions are not further researched in the prior art, the characteristic of the fluctuation state of the people is constructed by utilizing the amplitude information of the received subcarriers, and the innovation point is that a Back Propagation (BP) neural network is improved by utilizing a probability density function method and is used for identifying the number of people indoors with fine granularity. The method provided by the invention has the advantages of low cost and easy deployment, so that the method has strong expansibility and can have good robustness on activities of people at different indoor positions under the condition of small indoor people number scale.
Drawings
FIG. 1 is a block diagram of an implementation system of a passive indoor people detection method based on Wi-Fi channel state information.
Fig. 2 is a diagram of an indoor population counting experiment scene.
Fig. 3 is a structure diagram of a readable storage medium that can implement a passive people detection method based on Wi-Fi channel state information according to the present disclosure.
Detailed Description
The embodiments of the present disclosure are described below with specific examples, and other advantages and effects of the present disclosure will be readily apparent to those skilled in the art from the disclosure of the present disclosure. It is to be understood that the described embodiments are merely illustrative of some, and not restrictive, of the embodiments of the disclosure. The disclosure may be embodied or carried out in various other specific embodiments, and various modifications and changes may be made in the details within the description without departing from the spirit of the disclosure. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
It is noted that various aspects of the embodiments are described below within the scope of the appended claims. It should be apparent that the aspects described herein may be embodied in a wide variety of forms and that any specific structure and/or function described herein is merely illustrative. Based on the disclosure, one skilled in the art should appreciate that one aspect described herein may be implemented independently of any other aspects and that two or more of these aspects may be combined in various ways. For example, an apparatus may be implemented and/or a method practiced using any number of the aspects set forth herein. Additionally, such an apparatus may be implemented and/or such a method may be practiced using other structure and/or functionality in addition to one or more of the aspects set forth herein.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present disclosure, and the drawings only show the components related to the present disclosure rather than the number, shape and size of the components in actual implementation, and the type, amount and ratio of the components in actual implementation may be changed arbitrarily, and the layout of the components may be more complicated.
In addition, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects may be practiced without these specific details.
The invention is described in detail below with reference to the drawings and the detailed description.
Fig. 1 is an experimental overall framework diagram, which describes the overall operation flow of the patent in detail.
As shown in fig. 1, the invention relates to a passive indoor people number detection method based on Wi-Fi channel state information, which specifically comprises the following steps:
The transmitter and the receiver in the Wi-Fi detection environment are respectively a commercial standard 2.4GHz Wi-Fi b/g/n router, wherein the transmitter is provided with 3 transmitting antennas, the receiver is provided with 3 receiving antennas, each receiving antenna is provided with 56 subcarriers, an open source driver Atheros-CSI-Tool developed by Yaxing Xie and the like is used, and the packet sending frequency of the router is set to be 100 Hz; the transmitters and the receivers are arranged in a non-line-of-sight manner, and CSI raw signals of different numbers of volunteers (0, 1,2,3,4 and 5 persons) in the process of slowly walking at different positions are collected from experimental equipment.
In the step 1, the CSI original amplitude signal is obtained specifically according to the following steps:
step 1.1, selecting an Atheros AR9580 wireless network card to acquire CSI signals, and utilizing two routers which can support installation of OpenWRT firmware capable of receiving the CSI signals (OpenWRT is a highly modularized and highly automated embedded Linux system with powerful network components and expansibility), wherein one of the routers is used as an Access Point (AP) and the other router is used as a receiving end (MP);
the two routers belong to the same local area network, the two routers are controlled by a terminal (PC) to realize data sending and receiving, the routers are provided with OpenWRT 15.05.1 firmware, and the packet sending frequency of the routers is set to be 100 Hz.
Compared with the mode that the antenna is drawn out by using the PC, the router is more flexible, and the PC end is not required to be configured with another environment from the beginning.
And 2, performing data preprocessing on the CSI signals acquired in the step 1.
And detecting and deleting abnormal points of the amplitude signal, and filtering amplitude noise in the original CSI observation value.
Due to the fact that abnormal points and interference of random noise exist in the detection environment, the amplitude of the CSI signals is preprocessed by the CSI original signals of different numbers of volunteers collected in the step 1 on the basis of the Matlab software platform, accuracy of feature extraction and accuracy of a classifier are improved, and counting errors caused by the noise and the abnormal values are effectively reduced.
In the step 2, the amplitude information after the preprocessing is obtained specifically according to the following steps:
the step 2 specifically comprises the following steps:
and 2.1, reading the CSI original signals by using read _ log _ file functions based on a Matlab software platform by using the CSI original signals of the volunteers with different numbers in the step 1, wherein the CSI data is a two-dimensional matrix with N rows, and N is the number of data packets.
Since the transmitter has 3 transmit antennas and the receiver has 3 receive antennas in the experiment, each receive antenna has 56 subcarriers, and each signal sample contains 3 × 56 × 504 subcarrier CSI data, the CSI data obtained is a two-dimensional matrix of N × 540.
In one embodiment, since the transmitter has 3 transmit antennas, the receiver has 3 receive antennas, each receive antenna has 56 subcarriers, and each signal sample contains 3 × 56 × 504 subcarrier CSI data, the CSI data obtained is a two-dimensional matrix of N × 540(3 × 56), where N is the number of data packets.
The CSI data describe: the frequency response of the channel may be represented by:
wherein, H (f) i ) Represents a phase; h (f) i ) Is the channel state information, i.e., CSI, for the subcarriers whose center frequencies are fi. N is a radical of hydrogen s The number of subcarriers for a single antenna is 56 on the Atheros AR9580 wireless network card. Suppose the number of antennas at the transmitting end is N tx The number of antennas at the receiving end is N rx Then the CSI data may be expressed as:
H ij CSI data for the ith antenna of a transmit antenna (TX) to the jth antenna of a receive antenna (RX), each antenna pair data comprising N s Sub-carriers, then H ij It can be specifically expressed as:
each signal sample containing N tx ×N rx ×N s Subcarrier CSI data.
And 2.2, carrying out abnormal point detection and removal on the CSI original signal read in the step 2.1.
After reading the CSI original signal, the amplitude signal is detected and outliers are removed using a hanpell (Hampel) filter.
And 2.3, removing the CSI signals of the abnormal amplitude points in the step 2.2, and filtering amplitude noise in the original CSI observation value by using a discrete wavelet transform method. Wherein, db3 wavelet function is adopted to carry out 5-layer decomposition, and the fluctuation relation of amplitude signals along with the number of people is obtained.
After the signals are filtered, the graph curve becomes smoother, and frequent random fluctuation is filtered.
And 3, analyzing the fluctuation relation of the amplitude signal preprocessed in the step 2 along with the number of people and standardizing the preprocessed amplitude.
The step 3 specifically comprises the following steps:
step 3.1, analyzing CSI signals of different numbers of volunteers (0, 1,2,3,4 and 5 persons) walking slowly at different positions after amplitude data preprocessing in the step 2.3;
along with the increase of people, the CSI amplitude fluctuation range is larger and larger, and the change effect is more and more obvious, so that the working principle of identifying people by using the CSI amplitude fluctuation is well proved.
And 3.2, carrying out minimum-maximum normalization on the amplitude values preprocessed in the step 2, namely carrying out linear transformation on the data, and mapping the data values between [0 and 1 ]. The specific formula is as follows:
the amplitude signal x after CSI noise reduction is normalized into an x' formula through the maximum and minimum values: (A denotes the amplitude of the current subcarrier, and maxA are the minimum and maximum values of the current subcarrier amplitude, respectively).
And 4, performing feature extraction, PCA dimension reduction, feature vector construction and corresponding label formulation on the amplitude subjected to the data preprocessing in the step 3 based on a time window method.
The method comprises the steps of adopting the average value, the range, the standard deviation, the root mean square and the like of amplitude as characteristic representation fluctuation conditions to form characteristic vectors, formulating corresponding class labels, calculating data statistical characteristics of the number of packets received at intervals of a fixed time window 5s, extracting amplitude characteristic information of slow walking of different numbers of volunteers (0, 1,2,3,4 and 5 individuals) at different positions, formulating corresponding class labels and constructing a data set.
Step 4 specifically comprises the following steps:
and 4.1, selecting a proper time window size to provide a basis for constructing a feature vector in the next step.
For the indoor activities of people, the pace is generally not too fast, and if the time window is too small, the selected data is not enough to represent the overall characteristics; if the time window is too large, the system may not be sensitive to the variation of the number of people, and a large amount of training data is needed, and finally the time window is selected to be 5s according to the amplitude fluctuation effect, that is, 500 data packets are taken as one sample (each sample can be regarded as a row vector of 1 × 504);
and 4.2, according to the size of the time window selected in the step 4.1, taking the mean value, the range, the standard deviation and the root mean square of the amplitude as characteristic representation fluctuation conditions to form a characteristic vector.
featrure=[m,v,d,rs]
Wherein m is the mean, v is the range, d is the standard deviation, rs is the root mean square value, and then the feature vector is constructed.
For example, the data features are calculated according to the number of packets received every fixed time window 5s, and the amplitude features of different numbers of volunteers (0, 1,2,3,4, 5 persons) walking slowly at different positions form a feature vector.
And 4.3, carrying out PCA (principal component analysis) dimension reduction on the feature vector formed in the step 4.2, selecting the number of columns with obvious features, formulating corresponding category labels, and constructing a data set of the people number detection system.
And 5, establishing a crowd counting detection model according to a machine learning algorithm.
The method specifically comprises the following steps:
step 5.1, dividing the data set constructed in the step 4.3 according to a conventional method for dividing the data set by machine learning, wherein the data set is divided into 7: 3, randomly dividing the ratio into a training set and a testing set;
step 5.2, improving a Back Propagation (BP) neural network by using a method based on a probability density function, and training a training set;
in the forward propagation process, the BPNN model calculates the prediction output mapped by the current input according to the current input and the weight w and offset b values updated after the last backward propagation; in the process of back propagation, continuously updating w and b values from back to front according to the error of the current layer, wherein the updating process of w and b in the back propagation is as follows:
the BPNN model has an input layer, two hidden layers, and an output layer, where the parameter y represents an input vector of the input layer, specifically, a statistical characteristic value of the subcarriers after the dimension reduction processing in step 4.3, such as a mean value, a range, a standard deviation, a root mean square, and the like;the system is an output vector of an output layer, specifically a forecast number classification category; wherein the content of the first and second substances,representing a linear result acting on a jth neuron in a kth layer; parameter(s)Output, parameter representing the jth neuron activation function in the kth layerRepresenting the connection weight between the ith neuron in the k-1 layer pointing to the jth neuron in the k layer,representing the bias of the jth neuron at layer k. Obtaining a prediction error E by using a least square method total Analyzing and continuously optimizing the BPNN model according to the output number classification result of people, wherein the parameter sigma is k,j Representing the error of the jth neuron in the kth layer. (i, j, k ═ 1,2,3,4)
The activation functions of the neurons in the same layer of the traditional BPNN model are the same and common activation functions such as a linear function, a ReLU function and a sigmoid function are used, while the activation functions of the neurons in the same layer of the improved BPNN model are different, four neurons in the same layer use four different probability density functions as activation functions (exponential distribution, logarithmic distribution, normal distribution and gamma functions) to act on each neuron, and the different activation functions of the neurons in the same layer are realized, wherein the specific four probability density functions are as follows:
f 1 is an exponential distribution, f 2 Normal distribution, f 3 Lognormal distribution, f 4 Gamma function, f 1 ,f 2 ,f 3 ,f 4 Is an activation function.Are parameters.For the purpose of the lower incomplete gamma function,for the gamma function, E is calculated using the gradient descent method for parameter setting total And (4) relative to the partial differential of each parameter, as shown in the following formula, after the updating is completed, the trained BPNN model can be obtained.
And 5.3, verifying by using the test set to obtain a person number detection model which can accurately detect the number of people in the room.
For example, in an embodiment, subsequent validation with a test set results in a good people detection model that can accurately detect the number of people in a room without exceeding 4 people.
Fig. 2 is an experimental scenario design diagram. The terminal (PC) realizes that the main router sends CSI signal data and receives the CSI signal data from the router by controlling the wired connection of the main router, and the corresponding IP address is 192.168.1.180; the method is characterized in that an Atheros AR9580 wireless network card is configured in a master-slave router for collecting CSI signals, OpenWRT 15.05.1 firmware is configured and installed, the master router serves as a transmitting end, the corresponding IP address is 192.168.1.1, the MAC address is 9C:21:6A: F4: C1:6E, the slave router serves as a receiving end, the IP address is 192.168.1.233, the MAC address is D8:15:0D: FE:30:01, the master-slave router is in wireless connection with the same slave local area network, the master router receives CSI data from the router through Secure Shell (ssh) remote control, and compared with the mode that the router is used for receiving the CSI signals, the mode that the antenna is more flexible is used for drawing the PC, and another environment is not required to be configured for the PC end from the beginning.
Fig. 3 is a schematic diagram of a computer-readable storage medium according to an embodiment of the present disclosure. As shown in fig. 3, a computer-readable storage medium 40 according to an embodiment of the present disclosure has non-transitory computer-readable instructions 41 stored thereon. The non-transitory computer readable instructions 41, when executed by a processor, perform all or some of the steps of the aforementioned access control method of the removable storage device of the embodiments of the present disclosure.
It should be noted that the computer readable medium in the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present disclosure, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device.
Computer program code for carrying out operations for aspects of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described in the embodiments of the present disclosure may be implemented by software or hardware. Where the name of a unit does not in some cases constitute a limitation of the unit itself, for example, the first retrieving unit may also be described as a "unit for retrieving at least two internet protocol addresses".
The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the disclosure herein is not limited to the particular combination of features described above, but also encompasses other embodiments in which any combination of the features described above or their equivalents does not depart from the spirit of the disclosure. For example, the above features and the technical features disclosed in the present disclosure (but not limited to) having similar functions are replaced with each other to form the technical solution.
Claims (8)
1. A passive indoor people number detection method based on Wi-Fi channel state information is characterized by comprising the following steps:
step 1, building a Wi-Fi detection environment with a single transmitter and a single receiver, and collecting CSI original signals of different numbers of volunteers in the process of slowly walking at different positions;
step 2, carrying out data preprocessing on the CSI signals acquired in the step 1;
step 3, carrying out minimum-maximum normalization processing on the amplitude obtained in the step 2 to obtain standardized data;
step 4, performing feature extraction and PCA dimension reduction on the standardized data obtained in the step 3 based on a time window method, so as to construct a feature vector, formulate a corresponding label and construct a data set;
and 5, randomly dividing the characteristic vector data set obtained in the step 4 into a training set and a test set according to a proportion, training the training set by improving a BP neural network by using a probability density function-based method, and verifying by using the test set to obtain a person number detection model capable of accurately detecting the number of people indoors.
2. The method of claim 1, wherein the Wi-Fi channel status information is used to detect the number of people in a room,
in the step 1, the CSI original amplitude signal is obtained specifically according to the following steps:
step 1.1, an Atheros AR9580 wireless network card is selected to collect CSI signals, two routers capable of supporting installation of OpenWRT firmware capable of receiving the CSI signals are utilized, one of the two routers serves as a transmitting end, and the other router serves as a receiving end.
3. The method of claim 2, wherein the Wi-Fi channel status information is used to detect the number of people in a room,
the two routers belong to the same local area network, the two routers are controlled by the terminal to realize data sending and receiving, the type of the router is TL-WDR4900, OpenWRT 15.05.1 is installed, and the packet sending frequency of the router is set to be 100 Hz.
4. The method as claimed in claim 1, wherein the step 2 specifically includes detecting an amplitude signal, deleting outliers, and filtering out amplitude noise in the raw CSI observation.
5. The method of claim 4, wherein the Wi-Fi channel status information is used to detect the number of people in a room,
in step 2, the amplitude information after the preprocessing is obtained specifically according to the following steps:
step 2.1, reading the CSI original signals by using the CSI original signals of different numbers of volunteers in the step 1 and using a read _ log _ file function based on a Matlab software platform, wherein the CSI data is a two-dimensional matrix of N rows, and N is the number of data packets;
in the experiment, the transmitter has 3 transmitting antennas, the receiver has 3 receiving antennas, each receiving antenna has 56 subcarriers, and each signal sample contains 3 × 56 — 504 subcarrier CSI data, so that the obtained CSI data is a two-dimensional matrix of N × 540;
wherein the frequency response of the channel can be represented by the following formula:
wherein, H (f) i ) Represents a phase; h (f) i ) Is a center frequency of f i Channel state information, i.e., CSI, of the subcarriers of (a); n is a radical of s 56 subcarriers of a single antenna are on the Atheros AR9580 wireless network card; suppose the number of antennas at the transmitting end is N tx The number of antennas at the receiving end is N rx Then the CSI data may be expressed as:
H ij CSI data for the ith antenna of a transmit antenna (TX) to the jth antenna of a receive antenna (RX), each antenna pair data comprising N s Sub-carriers, then H ij It can be specifically expressed as:
each signal sample containing N tx ×N rx ×N s Subcarrier CSI data;
step 2.2, abnormal point detection and removal are carried out on the CSI original signal read in the step 2.1;
and 2.3, removing the CSI signals of the abnormal amplitude points in the step 2.2, and filtering amplitude noise in the original CSI observation value by using a discrete wavelet transform method.
6. The method for detecting the number of the people in the passive room based on the Wi-Fi channel state information according to claim 1, wherein in the step 3, the normalized amplitude information is obtained specifically according to the following steps:
step 3.1, analyzing CSI signals of different numbers of volunteers walking slowly at different positions after amplitude data preprocessing in the step 2.3;
step 3.2, carrying out minimum-maximum normalization on the amplitude value preprocessed in the step 2, and mapping the data value to the position between [0 and 1 ]; the concrete formula is as follows:
the amplitude signal x after CSI noise reduction is normalized into an x' formula through the maximum and minimum: where a represents the amplitude of the current subcarrier, and minA and maxA are the minimum and maximum values of the current subcarrier amplitude, respectively.
7. The method as claimed in claim 1, wherein in the step 4, the feature vector of the amplitude is obtained and a corresponding class label is formulated to construct a data set according to the following steps:
step 4.1, selecting a proper time window size to provide a basis for constructing a feature vector in the next step;
step 4.2, according to the size of the time window selected in the step 4.1, taking the mean value, the range, the standard deviation and the root mean square of the amplitude as characteristic representation fluctuation conditions to form a characteristic vector;
featrure=[m,v,d,rs]
wherein m is a mean value, v is a range, d is a standard deviation, and rs is a root mean square value, and then constructing a feature vector;
and 4.3, carrying out PCA (principal component analysis) dimensionality reduction on the feature vector formed in the step 4.2, selecting the column number with obvious features, formulating corresponding category labels, and constructing a data set of the people number detection system.
8. The method as claimed in claim 1, wherein in the step 5, the training set is trained by using a back propagation neural network improved by a method based on a probability density function, and the people number can be accurately detected by establishing a people number detection model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210834030.7A CN115002703A (en) | 2022-07-14 | 2022-07-14 | Passive indoor people number detection method based on Wi-Fi channel state information |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210834030.7A CN115002703A (en) | 2022-07-14 | 2022-07-14 | Passive indoor people number detection method based on Wi-Fi channel state information |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115002703A true CN115002703A (en) | 2022-09-02 |
Family
ID=83022810
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210834030.7A Pending CN115002703A (en) | 2022-07-14 | 2022-07-14 | Passive indoor people number detection method based on Wi-Fi channel state information |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115002703A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116304844A (en) * | 2023-05-23 | 2023-06-23 | 山东科技大学 | Personnel entry and exit counting and counting system and method |
-
2022
- 2022-07-14 CN CN202210834030.7A patent/CN115002703A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116304844A (en) * | 2023-05-23 | 2023-06-23 | 山东科技大学 | Personnel entry and exit counting and counting system and method |
CN116304844B (en) * | 2023-05-23 | 2023-09-01 | 山东科技大学 | Personnel entry and exit counting and counting system and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Hsieh et al. | Deep learning-based indoor localization using received signal strength and channel state information | |
Zou et al. | Deepsense: Device-free human activity recognition via autoencoder long-term recurrent convolutional network | |
JP7154295B2 (en) | Motion detection based on machine learning of radio signal characteristics | |
CN108629380B (en) | Cross-scene wireless signal sensing method based on transfer learning | |
Khan et al. | A deep learning framework using passive WiFi sensing for respiration monitoring | |
CN109672485B (en) | Indoor personnel real-time invasion and movement speed detection method based on channel state information | |
US11962437B2 (en) | Filtering channel responses for motion detection | |
CN107480699A (en) | A kind of intrusion detection method based on channel condition information and SVMs | |
CN106792808A (en) | Los path recognition methods under a kind of indoor environment based on channel condition information | |
Arshad et al. | Leveraging transfer learning in multiple human activity recognition using WiFi signal | |
CN114359738B (en) | Cross-scene robust indoor people number wireless detection method and system | |
CN110933628B (en) | Fingerprint indoor positioning method based on twin network | |
CN107992882A (en) | A kind of occupancy statistical method based on WiFi channel condition informations and support vector machines | |
CN111901028B (en) | Human body behavior identification method based on CSI (channel State information) on multiple antennas | |
CN114222202A (en) | Environment self-adaptive activity detection method and system based on WiFi CSI | |
CN110109090A (en) | Circumstances not known multi-target detection method and device based on microwave radar | |
CN115002703A (en) | Passive indoor people number detection method based on Wi-Fi channel state information | |
CN114423034A (en) | Indoor personnel action identification method, system, medium, equipment and terminal | |
CN114639169B (en) | Human motion recognition system based on attention mechanism feature fusion and irrelevant to position | |
CN106230544A (en) | The monitoring identification of a kind of automobile remote-control interference signal and localization method | |
CN111817803A (en) | Frequency spectrum sensing method and system based on correlation coefficient and K-means clustering algorithm and computer readable storage medium | |
Choi et al. | Simultaneous crowd estimation in counting and localization using WiFi CSI | |
CN114676727B (en) | CSI-based human body activity recognition method irrelevant to position | |
Natarajan et al. | A machine learning approach to passive human motion detection using WiFi measurements from commodity IoT devices | |
CN111093163B (en) | Passive target classification method based on channel state information |
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 |