CN114757227A - User presence sensing technology based on channel state information - Google Patents
User presence sensing technology based on channel state information Download PDFInfo
- Publication number
- CN114757227A CN114757227A CN202210358904.6A CN202210358904A CN114757227A CN 114757227 A CN114757227 A CN 114757227A CN 202210358904 A CN202210358904 A CN 202210358904A CN 114757227 A CN114757227 A CN 114757227A
- Authority
- CN
- China
- Prior art keywords
- csi
- user
- state
- detection
- distribution
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a user presence sensing technology based on channel state information, belonging to the technical field of electrical engineering and automation thereof, and comprising the following steps of: selecting a CSI acquisition end of 1 antenna as a research object, setting a CSI amplitude matrix H of sampling N sub-carriers within T seconds at an S-sending rate, and performing data filtering on time sequence data of each sub-carrier in the CSI amplitude matrix H in order to avoid negative influence of adverse factors such as measurement errors, environmental noise and the like on personnel detection. The main advantages of the technology in the invention are: 1. the precision is high: a high-precision detection result is provided by adopting a data depiction method for correcting a Rayleigh distribution model; 2. the speed is high: due to the cooperative computing mode and the rolling computing method, the human or unmanned state can be detected within the second level; 3. the cost is low: personnel detection is carried out based on Wi-Fi signals, and extra equipment does not need to be introduced when the intelligent home network is guaranteed to be normally used.
Description
Technical Field
The invention belongs to the technical field of electrical engineering and automation thereof, and particularly relates to a user presence sensing technology based on channel state information.
Background
In order to realize the requirements of intrusion detection, unmanned exploration, old people nursing and the like, various devices adopting the modes of video monitoring, sound wave monitoring, infrared detection and the like are widely applied in the market, but the defect that additional devices are required to be introduced exists. Due to the universality of Wi-Fi equipment application, the Wi-Fi-based detection mode can remarkably overcome the cost increase and the installation difficulty caused by the introduction of additional equipment in the traditional scheme, and therefore, the Wi-Fi-based detection mode has the characteristic of equipment independence (device-free).
The channel state information is used as a main reflection index of a Wi-Fi multipath propagation result, and can fully reflect the change generated along with the action of a user in the signal propagation process, so that the channel state information is widely researched and paid attention to. For example, in the related technologies [1], [2] and [3], deep learning algorithms such as a support vector machine and a neural network are utilized to search for multi-classification rules of CSI when a user is in different states, and the multi-classification rules are used as an offline feature fingerprint library to be compared with real-time feature fingerprints, so that whether the user state exists or not is judged; the correlation technique [4] measures the difference degree of CSI of different behaviors of a user by using a Doppler velocity motion model so as to estimate the walking parameters of the user, and the correlation technique [5] judges whether the user exists or not according to the difference of correlation coefficients between sub-carrier channels in different user states.
The defects in the prior art are mainly that the general calculation amount is large, the special network card equipment is relied on, the cost is high, the delay is large, the precision is insufficient, and the perception of user information is not suitable. Specifically, on one hand, a common computing device such as an edge computing intelligent gateway ECG needs to occupy a large amount of computing power when performing off-line solving on a characteristic fingerprint, so that other intelligent control processes are delayed, and a computing chip at a CSI acquisition end is in an idle state; on the other hand, in the prior art, most of the decisions are made on the basis of the CSI sequences in a longer time window to perform people detection in real time, but due to the influence of network delay, demodulation and calculation time consumption, the decision time scale in the second level is difficult to achieve in practice.
Disclosure of Invention
The invention aims to: the technology for sensing the existence of the user based on the Channel State Information is characterized in that a communication module which is widely embedded in an intelligent home equipment network and consists of a gateway and a terminal is combined with Wi-Fi Channel State Information (CSI) to sense the existence Information of the user in a high-precision, quick and low-cost mode by adopting a modified Rayleigh distribution model in a terminal and gateway collaborative calculation mode.
In order to achieve the purpose, the invention adopts the following technical scheme:
A user presence sensing technology based on channel state information specifically comprises the following steps:
s1, data filtering: selecting a CSI acquisition end of 1 antenna as a research object, setting a CSI amplitude matrix H of sampling N subcarriers within T seconds at an S packet sending rate, and performing data filtering on time sequence data of each subcarrier in the CSI amplitude matrix H in order to avoid negative effects of adverse factors such as measurement errors and environmental noises on personnel detection;
s2, based on S1, selecting Hampel filtering to eliminate abnormal points caused by measurement errors, and reducing environmental noises such as temperature and humidity changes by wavelet denoising;
s3, the distribution of CSI amplitude values is described by the corrected Rayleigh distribution: the sub-carriers of the CSI are orthogonal and do not interfere with each other, so that the sub-carriers can be assumed to accord with the generalized stationary incoherent scattering (WSSUS) criterion, and the amplitude of the synthesized signal is subject to Rayleigh distribution of different parameters;
s4, feature selection: the CSI amplitude characteristic is used for representing the human and unmanned states of the power consumers, so that the following characteristic quantities are provided: shape parameter sigma, right bias correction coefficient beta and subcarrier CSI amplitude change state delta.
S5, constructing a person detection characteristic fingerprint: on the basis of CSI amplitude characteristics, a corresponding personnel detection judgment matrix can be constructed, and under the influence of a frequency selectivity effect, each subcarrier has difference on CSI sensing capacity with or without user states, so that m sensitive subcarriers need to be screened out to serve as detection objects to improve the calculation rate, and meanwhile, once a CSI acquisition end is arranged, the position is hardly changed, so that only the sensitive subcarriers need to be screened when historical characteristic fingerprints are calculated in a preliminary preparation stage for the first time, and in view of the fact that the provided characteristic quantities are all related to sigma, the screening method is to select the first m subcarriers in which the sigma difference values in the historical characteristic fingerprints of the user states and the user states are sorted in a descending order;
S6, increasing the calculation rate based on the rolling calculation mode: a rolling calculation mode is provided to improve the timeliness of calculating a real-time characteristic fingerprint matrix, in a CSI amplitude matrix containing T data packets, the mode reduces the CSI amplitude matrix updated once by the original T data packets into k/s second updating, the k value is a value required by comprehensively considering calculation efficiency and detection precision, and is generally selected to be 50 data packets;
and S7, judging the existence state of the room personnel through a minimum distance criterion, describing the distance between the real-time characteristic fingerprint and the historical characteristic fingerprint matrix with or without the user state through the F norm of the difference value of the characteristic matrix in order to measure the difference degree of the current CSI data and the data distribution in the two states, and taking the state of the person close to the distance as the detection result of final judgment.
As a further description of the above technical solution:
in S2, it should be noted that temperature and humidity changes are usually high-frequency noise, and signal changes caused by user activities are mostly in a low-frequency range, so that the detection accuracy is increased by using the advantage of wavelet denoising in high-frequency filtering.
As a further description of the above technical solution:
in S4, σ and β are used as unique parameters describing CSI amplitude distribution, and user states can be accurately measured, where the former affects the form of probability density, and the latter is used for right-shifting the probability distribution.
As a further description of the above technical solution:
in S4, user states may be classified into a user static state (such as noon break) and a user action state (such as walking), and the user static state have no influence on multipath propagation of the signal, and the CSI amplitude distribution of the subcarriers is in a steady state, but the user action will disturb the distribution thereof, and in the case of walking or moving of the user, the shape parameter σ changes significantly, and therefore, the CSI amplitude change state δ of the subcarriers is defined to characterize the user action.
As a further description of the above technical solution:
in S6, the method improves timeliness while making full use of historical data. Furthermore, the proposed mode not only does not affect the calculation of the feature quantity in the user-unavailable state, but also can describe the time domain characteristics of the feature quantity in the user action state in a finer granularity.
In summary, due to the adoption of the technical scheme, the invention has the beneficial effects that:
in the invention, the judgment of the state of people in a room without people can be realized in the gateway-terminal intelligent home network which takes Wi-Fi as a main communication mode.
The main advantages of the technology are: 1. the precision is high: a data characterization method for correcting a Rayleigh distribution model is adopted to provide a high-precision detection result; 2. the speed is high: due to the cooperative computing mode and the rolling computing method, the human or unmanned state can be detected within the second level; 3. the cost is low: personnel detection is carried out based on Wi-Fi signals, and extra equipment does not need to be introduced when the intelligent home network is guaranteed to be normally used.
Drawings
Fig. 1 is a schematic diagram of a CSI-based personnel detection scheme of a user presence sensing technology based on channel state information according to the present invention;
fig. 2 is a schematic diagram of a topology of a related device of a user presence sensing technology based on channel state information according to the present invention;
fig. 3 is a schematic diagram of a collaborative computing mode of an ECG and CSI collection terminal of a user presence sensing technology based on channel state information according to the present invention;
fig. 4 is a schematic structural diagram of a real-time human detection process in the user presence sensing technology based on channel state information according to the present invention;
fig. 5 is a schematic diagram of a parameter algorithm estimation process in a user presence sensing technology based on channel state information according to the present invention;
fig. 6 is a schematic diagram of CSI amplitude distribution of subcarriers when there is no user status in the user presence sensing technology based on channel status information according to the present invention;
fig. 7 is a schematic diagram of a modified rayleigh distribution with and without user states of a user presence sensing technique based on channel state information according to the present invention;
fig. 8 is a schematic diagram of a rolling calculation mode of a characteristic fingerprint matrix of a user presence sensing technology based on channel state information according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1-8, the present invention provides a technical solution: a user presence sensing technology based on channel state information specifically comprises the following steps:
s1, data filtering: selecting a CSI acquisition end of 1 antenna as a research object, setting a CSI amplitude matrix H of sampling N subcarriers within T seconds at an S packet sending rate, and performing data filtering on time sequence data of each subcarrier in the CSI amplitude matrix H in order to avoid negative effects of adverse factors such as measurement errors and environmental noises on personnel detection;
matrix H is as follows:
H＝[hk,t]N×T'
in the formula: h isk,tThe CSI amplitude of the subcarrier k at the time t; t ═ T × S;
s2, based on S1, selecting Hampel filtering to eliminate abnormal points caused by measurement errors, and reducing environmental noises such as temperature and humidity changes by wavelet denoising, wherein in S2, it needs to be noted that the temperature and humidity changes are usually high-frequency noises, and signal changes caused by user activities are mostly in a low-frequency range, so that the detection precision is increased by utilizing the advantages of the wavelet denoising in the aspect of high-frequency filtering;
S3, the distribution of CSI amplitude values is described by the corrected Rayleigh distribution: the sub-carriers of the CSI are orthogonal and do not interfere with each other, so that the sub-carriers can be assumed to accord with the generalized stationary incoherent scattering (WSSUS) criterion, and the amplitude of the synthesized signal is subject to Rayleigh distribution of different parameters;
the probability density is as follows:
wherein σ is Rayleigh distribution shape parameter, and standard Rayleigh distribution is represented by all observed values xiIs defined as
Theoretically, a series of CSI amplitude distribution characteristics such as the maximum probability density, the mean probability density, the sigma probability density, and the like can be obtained by estimating the CSI subcarrier amplitude sequence, and then the rayleigh distribution parameters required for detecting the user state can be easily obtained. However, the CSI amplitude actually received by the CSI acquisition end may generate a corresponding offset, so that the present document proposes a CSI correction probability density expression based on rayleigh distribution:
in the formula: beta is a right-bias correction coefficient, and sigma is a shape parameter.
As shown in fig. 6. Therefore, the CSI amplitude distribution condition of the subcarriers is highly consistent with the corrected Rayleigh distribution, and compared with the Gaussian kernel density estimation corrected Rayleigh distribution, the distribution form of the whole data without considering noise can be further described.
For the above formula, the x value range of the modified rayleigh distribution is first made as follows Discussion: for a standard Rayleigh distribution, x is for any x>0, the rayleigh distribution definitional equation only satisfies the normalization of the probability density function when x takes on the value 0, + ∞). Therefore, for the corrected rayleigh distribution shown in the formula, it can be known that the corrected rayleigh distribution is a standard rayleigh distribution right-shifted β with respect to the x-axis, and therefore, for the corrected rayleigh distribution, there should be a case where the normalization of the probability density function is satisfied
The estimation of beta and sigma can adopt a maximum likelihood estimation method, and the solution of the two is as follows:
as can be seen from the above formula, the expression of β relates to an n +1 th-order polynomial about β, and the solution is complicated, so to adapt to the computational capability of the terminal chip, a solution value estimation method based on a heuristic method and a bisection method is adopted: continuously increasing a small value to probe the zero-crossing interval of g (beta) at the position, and then combining a dichotomy to obtain a more accurate solution, wherein the algorithm flow is shown in figure 5;
the method can quickly and accurately estimate the parameters for correcting the Rayleigh distribution, and practice tests prove that the more accurate solution can be obtained when the integral iteration times are always within five times, so that a basis is provided for real-time calculation in a terminal.
S4, feature selection: by representing the presence or absence of a power consumer by using the CSI amplitude characteristics, the following features are proposed: shape parameter sigma, right bias correction coefficient beta and subcarrier CSI amplitude change state delta, wherein in S4, sigma and beta are used as unique parameters for describing CSI amplitude distribution conditions, so that the user state can be accurately measured, the former influences the form of probability density, and the latter is used for right shift probability distribution conditions; (ii) a
In S4, user states may be divided into a user static state (e.g., noon break) and a user action state (e.g., walking), and the user static state and the user non-user state have no influence on multipath propagation of signals, and the CSI amplitude distribution of the subcarriers is in a steady state, but the user action disturbs the distribution thereof, as shown in fig. 7, in the case of user walking or user action, the shape parameter σ changes significantly, and therefore, the CSI amplitude change state δ of the subcarriers is defined to characterize the user action;
considering that user actions are a dynamic process, while users are static and no user is a static situation, the values of the pre-story preparation phase are as follows:
in the real-time detection phase, δ may be determined by determining the difference between adjacent detection periods σ:
in the formula: sigmatA Rayleigh distribution parameter is a detection time period t; Δ is the sensitivity threshold of the neighboring state, and its range is discussed later.
S5, constructing a person detection characteristic fingerprint: on the basis of CSI amplitude characteristics, a corresponding personnel detection judgment matrix can be constructed, and under the influence of a frequency selectivity effect, each subcarrier has difference on CSI sensing capacity with or without user states, so that m sensitive subcarriers need to be screened out to serve as detection objects to improve the calculation rate, and meanwhile, once a CSI acquisition end is arranged, the position is hardly changed, so that only the sensitive subcarriers need to be screened when historical characteristic fingerprints are calculated in a preliminary preparation stage for the first time, and in view of the fact that the provided characteristic quantities are all related to sigma, the screening method is to select the first m subcarriers in which the sigma difference values in the historical characteristic fingerprints of the user states and the user states are sorted in a descending order;
Constructing a characteristic fingerprint criterion by using the characteristic quantity of the sensitive subcarrier, wherein a characteristic fingerprint matrix after dimension is normalized and eliminated is as follows:
in the formula: CFm×3A characteristic fingerprint matrix formed by m sensitive subcarriers;
and
respectively a Rayleigh distribution parameter, a right bias correction coefficient and a CSI amplitude change state of the sensitive subcarrier i. Meanwhile, in order to ensure that the influence degrees of all the indexes on the judgment result are the same, the shape parameter and the right deviation correction parameter need to be carried out by [0,1 ]]Normalization;
s6, increasing the calculation rate based on the rolling calculation mode: a rolling calculation mode is provided to improve the timeliness of calculating a real-time characteristic fingerprint matrix, in a CSI amplitude matrix containing T data packets, the mode reduces the CSI amplitude matrix updated once by the original T data packets into k/S second updating, the k value is a value required by comprehensively considering calculation efficiency and detection precision, generally 50 data packets are selected, and in S6, the timeliness is improved by the method under the condition of fully utilizing historical data. Furthermore, the extracted mode does not influence the calculation of the characteristic quantity in a user-free state, and can depict the time domain characteristic of the characteristic quantity in a user action state in a finer granularity;
In the real-time detection stage, the characteristic fingerprint matrix is calculated according to the CSI amplitude distribution characteristics of subcarriers, and accumulation of a certain amount of time sequence CSI amplitude sequences is a premise of calculation of CSI amplitude distribution, so that a certain time acquisition process is required for achieving the accumulated data volume in most researches. In addition, most low-cost CSI acquisition terminals have a low packet sending rate S (such as 50 packet/S), which seriously affects the timeliness of personnel detection. Therefore, a rolling calculation mode is provided to improve the timeliness of calculating the real-time feature fingerprint matrix.
S7, judging the existence state of the room personnel through a minimum distance criterion, describing the distance between the real-time characteristic fingerprint and the historical characteristic fingerprint matrix with or without the user state through the F norm of the characteristic matrix difference value in order to measure the difference degree of the current CSI data and the data distribution under the two states, and taking the state of a person close to the distance as the detection result of final judgment:
in the formula: the state is a real-time result of personnel detection, 1 represents that people exist, and 0 represents that no people exist; CF (compact flash)real、CF1And CF0Respectively real-time characteristic fingerprints, presence and absence of user state historical characteristic fingerprints; | | is F norm.
In this embodiment: the basic framework of the CSI-based sounding method is explained here.
The CSI, which may reflect the channel properties of Wi-Fi multipath propagation under the ieee802.11a/g/n protocol, decomposes the channel measurements into a series of orthogonal subcarriers and estimates the frequency response of each subcarrier, thereby obtaining a finer-grained channel description in the frequency domain.
In the formula, Hk(t) is a CSI sampling value on the kth subcarrier at the moment t, k is less than or equal to N, and N is the total number of subcarriers;
Wi-Fi signal propagation paths and their propagation can thus be analyzed from a finer-grained point of view, and the human presence detection framework based thereon is shown in fig. 1.
The technology provided by the scheme is realized in a common intelligent home hardware system, and hardware equipment of the intelligent home hardware system mainly comprises a smart gateway, a socket, an infrared common intelligent terminal and the like.
Each terminal adopts a low-cost ESP32-C3 chip, the chip has better storage performance and operational capability, and can be simultaneously used as an MCU, a communication module and CSI acquisition equipment to be integrated in terminals such as an intelligent socket in an intelligent power utilization network, and the introduction of the multifunctional multiplexing and expanding overall system architecture is realized as follows:
1. CSI-related device topology
The topology of the personnel detection system of the CSI acquisition end and the ECG and the like related to personnel detection is shown in FIG. 2.
In the system, a CSI acquisition PCB is embedded into an intelligent switch, an intelligent socket or an air conditioner control panel, and is installed in various power utilization scenes together with the equipment, and the equipment interacts with ECG through the existing WiFi module of the intelligent equipment. As can be seen from fig. 2, the topology structure multiplexes the Wi-Fi communication module of the sensing layer, thereby improving the economy.
2. EGC-terminal cooperative information interaction and calculation mechanism
According to the scheme, the calculation task allocation strategy of the ECG and the CSI acquisition terminal is designed by utilizing the residual calculation power of the CSI acquisition terminal.
The cooperative computing mode of the ECG and the CSI acquisition end emphasizes the localization of personnel real-time detection at the CSI acquisition end, the CSI acquisition end realizes the real-time detection based on the criterion threshold value trained by the ECG, and then sends the detection result to the ECG, as shown in fig. 3.
3. CSI-based real-time personnel detection process
The CSI measured value is easily influenced by the change of exogenous factors such as ambient temperature and humidity, indoor object layout and the like, so that the ECG needs to update the criterion threshold value calculated value at a fixed frequency and transmit the criterion threshold value calculated value to the CSI acquisition end in time, and the detection precision is maintained on a higher reference. The calculation of the criterion threshold value needs to occupy certain computing power, memory and other hardware performances, and the criterion threshold value is updated at half-day frequency in engineering. Most of the updating threshold processes of the ECG request the CSI acquisition terminal to send the CSI data without user states in the morning and at night regularly, and request the CSI acquisition terminal to send the CSI data with user states in the morning and in the afternoon, and the ECG carries out criterion threshold calculation based on the data sets.
In the cooperative computing mode of the ECG and CSI acquiring end, the real-time detection of the user status is divided into a preparation-in-advance phase and a real-time detection phase, as shown in fig. 4. In the pre-preparation stage, the ECG requests the CSI acquisition end to send a signal in the specific time period, and a historical characteristic fingerprint matrix with or without user states is generated through the multi-index characteristic fingerprint algorithm. In the detection implementation stage, the CSI acquiring end needs to respectively solve the F norm between the real-time characteristic fingerprint matrix and the historical characteristic fingerprint matrix with or without the user state, so as to measure the difference between the real-time user state and the user state with or without the user state, and finally take the user state with a smaller difference value as the detection result.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.
Claims (5)
1. A user presence sensing technology based on channel state information is characterized by comprising the following steps:
s1, data filtering: selecting a CSI acquisition end of 1 antenna as a research object, setting a CSI amplitude matrix H of sampling N subcarriers within T seconds at an S packet sending rate, and performing data filtering on time sequence data of each subcarrier in the CSI amplitude matrix H in order to avoid negative effects of adverse factors such as measurement errors and environmental noises on personnel detection;
s2, based on S1, selecting Hampel filtering to eliminate abnormal points caused by measurement errors, and reducing environmental noises such as temperature and humidity changes by wavelet denoising;
s3, the distribution of CSI amplitude values is described by the corrected Rayleigh distribution: the subcarriers of the CSI are orthogonal and do not interfere with each other, so that the CSI can be assumed to meet the generalized stationary incoherent scattering (WSSUS) criterion, and the amplitude of a synthesized signal is subjected to Rayleigh distribution of different parameters;
S4, feature selection: by representing the presence or absence of a power consumer by using the CSI amplitude characteristics, the following features are proposed: shape parameter sigma, right bias correction coefficient beta and subcarrier CSI amplitude change state delta.
S5, constructing a person detection characteristic fingerprint: on the basis of CSI amplitude characteristics, a corresponding personnel detection judgment matrix can be constructed, and under the influence of a frequency selectivity effect, each subcarrier has difference on CSI sensing capacity with or without user states, so that m sensitive subcarriers need to be screened out to serve as detection objects to improve the calculation rate, and meanwhile, once a CSI acquisition end is arranged, the position is hardly changed, so that only the sensitive subcarriers need to be screened when historical characteristic fingerprints are calculated in a preliminary preparation stage for the first time, and in view of the fact that the provided characteristic quantities are all related to sigma, the screening method is to select the first m subcarriers in which the sigma difference values in the historical characteristic fingerprints of the user states and the user states are sorted in a descending order;
s6, increasing the calculation rate based on the rolling calculation mode: a rolling calculation mode is provided to improve the timeliness of calculating a real-time characteristic fingerprint matrix, in a CSI amplitude matrix containing T data packets, the mode reduces the CSI amplitude matrix updated once by the original T data packets into k/s second updating, the k value is a value required by comprehensively considering calculation efficiency and detection precision, and is generally selected to be 50 data packets;
And S7, judging the existence state of the room personnel through a minimum distance criterion, describing the distance between the real-time characteristic fingerprint and the historical characteristic fingerprint matrix of the user existence state or the user nonexistence state through the F norm of the characteristic matrix difference value in order to measure the difference degree of the current CSI data and the data distribution under the two states, and taking the state of the person close to the distance as the detection result of final judgment.
2. The technology as claimed in claim 1, wherein in S2, it should be noted that temperature and humidity changes are usually high-frequency noise, and signal changes caused by user activities are mostly in a low-frequency range, so that the advantage of wavelet denoising in high-frequency filtering is utilized to increase detection accuracy.
3. The technique of claim 1, wherein σ and β are unique parameters describing CSI amplitude distribution in S4, and the former affects probability density morphology and the latter is used to right shift probability distribution.
4. The technique of claim 1, wherein in S4, user states are divided into user static state (such as noon break) and user action state (such as walking), and the user static state have no influence on multipath propagation of the signal, the CSI amplitude distribution of the sub-carriers is in a steady state, but the user action will disturb the distribution, and in the case of user walking or action, the shape parameter σ changes significantly, and for this reason, the sub-carrier CSI amplitude variation state δ is defined to characterize the user action.
5. The channel state information-based user presence awareness technology according to claim 1, wherein in S6, the method improves timeliness while fully utilizing historical data. Furthermore, the extracted mode does not affect the calculation of the feature quantity in a user-free state, and the time domain characteristics of the feature quantity in a user action state can be described in a finer granularity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210358904.6A CN114757227A (en) | 2022-06-13 | 2022-06-13 | User presence sensing technology based on channel state information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210358904.6A CN114757227A (en) | 2022-06-13 | 2022-06-13 | User presence sensing technology based on channel state information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114757227A true CN114757227A (en) | 2022-07-15 |
Family
ID=82329015
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210358904.6A Pending CN114757227A (en) | 2022-06-13 | 2022-06-13 | User presence sensing technology based on channel state information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114757227A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115586581A (en) * | 2022-12-02 | 2023-01-10 | 荣耀终端有限公司 | Personnel detection method and electronic equipment |
-
2022
- 2022-06-13 CN CN202210358904.6A patent/CN114757227A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115586581A (en) * | 2022-12-02 | 2023-01-10 | 荣耀终端有限公司 | Personnel detection method and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108650628B (en) | Indoor positioning method combining distance measurement and fingerprint based on Wi-Fi network | |
| Kjærgaard | Indoor location fingerprinting with heterogeneous clients | |
| US7327245B2 (en) | Sensing and analysis of ambient contextual signals for discriminating between indoor and outdoor locations | |
| CN109672485B (en) | Indoor personnel real-time invasion and movement speed detection method based on channel state information | |
| CN107786959B (en) | Compressed data collection method in wireless sensor network based on adaptive measuring | |
| KR100948412B1 (en) | Method and system for estimating the location using a Received Signal Strength Indication | |
| CN110830946B (en) | Mixed type online data anomaly detection method | |
| CN106162720B (en) | Cognitive wireless self-organizing network node stability evaluation method based on multi-attribute decision | |
| CN108038419A (en) | Wi-Fi-based indoor personnel passive detection method | |
| Al-Husseiny et al. | RF-based traffic detection and identification | |
| CN110569876A (en) | Non-invasive load identification method and device and computing equipment | |
| Yang et al. | Research on Wi-Fi indoor positioning in a smart exhibition hall based on received signal strength indication | |
| CN112381063A (en) | Channel state information-based people counting method | |
| CN109274438B (en) | Self-adaptive double-threshold cooperative interference detection method and device | |
| CN114757227A (en) | User presence sensing technology based on channel state information | |
| CN114355298B (en) | Radar composite modulation pulse signal identification method | |
| CN114222202A (en) | Environment self-adaptive activity detection method and system based on WiFi CSI | |
| CN112949487A (en) | WiFi-based ship cab personnel number detection method and system | |
| Zhang et al. | Riemannian mean shift-based data fusion scheme for multi-antenna cooperative spectrum sensing | |
| CN110289926B (en) | Spectrum sensing method based on symmetric peak values of cyclic autocorrelation function of modulation signal | |
| CN110472188A (en) | A kind of abnormal patterns detection method of facing sensing data | |
| Yang et al. | An identity perception algorithm based on WiFi channel state information | |
| CN114676783A (en) | Load identification method based on single classification and fuzzy width learning | |
| CN102195720B (en) | Method and system for measuring bottom noise of machine | |
| CN110575176A (en) | Action segmentation method based on two-layer sliding window threshold |
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 |