The invention discloses an active person number estimation method based on channel state information and mainly solves the problem that the active person number estimation result of an existing method is inaccurate and large in fluctuation. The active person number estimation method comprises an offline training stage and an online estimation stage, wherein the offline training stage specifically includes the steps of firstly, collecting the channel state information of known active person number; secondly, extracting the channel state information features of the known active person number; thirdly, training a support vector machine classifier; the online estimation stage specifically includes the steps of fourthly, collecting the channel state information of to-be-estimated active person number; fifthly, extracting the channel state information features of the to-be-estimated active person number; sixthly, estimating active person number. The active person number estimation method has the advantages that the channel state information features of different active person numbers can be accurately distinguished, and the accuracy and stability of active person number estimation are increased effectively.

Active person number estimation method based on channel state information

The invention provides a non-contact breathing detection method, including the steps: receiving radio frequency signal through at least two receiving antennas; conjugating multiplication of the firstchannel state information and the second channel state information corresponding to the radio frequency signals received by any two receiving antennas to construct a third channel state information; Analyzing features in the third channel state information to determine an optimal respiration detection feature; and calculating a respiratory frequency according to the optimal respiratory detection feature. The invention utilizes the radio frequency signal transmitted by the commercial WiFi equipment to realize the detection of the target breathing frequency. The conjugate multiplication method is used to solve the phase shift problem of commercial WiFi, and the amplitude feature and phase feature are combined for respiration detection, which realizes the respiration detection of commercial WiFi without blind area. In addition, the technical proposal provided by the invention also has the advantages of non-intrusion, convenience, low cost and the like in practical application.

Non-contact breathing detection method

The invention discloses an indoor activity detection method and system. The indoor activity detection method comprises the following steps: receiving wireless signals at two receiving points to obtainchannel state information; making difference between the phases in the channel state information of the two receiving points to obtain a phase difference signal sequence, and then performing interpolation processing on the phase difference; performing band-pass filtering on the phase difference after the interpolation processing; extracting features from the phase difference signal after the band-pass filtering, and calculating the standard deviation of the phase difference that the signal sequence is normalized through the manner of sliding window; and judging whether an activity exists in the current environment or not according to the standard deviation. According to the indoor activity detection method, the real-time mobile behavior detection can be realized by means of a commercial WiFi network card; and when the environment changes, the indoor activity detection method can automatically adapt to the environmental changes through the manner of self-learning without manual assistance; in addition, a number of novel processing mechanisms are provided by the invention to eliminate the interference caused by common scenes in life, therefore, the robustness of the mobile behaviordetection method is increased in general.

Indoor activity detection method and system

The invention discloses a WIFI module radio frequency calibration method. The method comprises the following steps: (11) selecting a modulation mode from all modulation modes of a WIFI module to serve as a reference modulation mode, performing test and calibration on the power of working channels in the reference modulation mode, and recording change rates of transmitting power values that change with the changes of power gain control parameters of the working channels; (12) calculating calibration values (the formula is as shown in the specification) of the power gain control parameters of the working channels in other modulation modes; (13) writing the (the formula is as shown in the specification) in the WIFI module; (14) controlling the (the formula is as shown in the specification) working channel in the (the formula is as shown in the specification) modulation mode to work according to the written (the formula is as shown in the specification). According to the WIFI module radio frequency calibration method disclosed by the invention, the property that the chip parameter, the resistance-capacitance, the PCB wiring and other parasitic parameters of the same board are invariable is adopted, all channels are calibrated by a certain modulation mode, meanwhile the corresponding power gain factors are obtained, the calibration parameters of other modulation modes can be directly calculated, and thus the radio frequency calibration time is greatly shortened.

WIFI module radio frequency calibration method

The invention provides a human body indoor positioning automatic detection method and system, and belongs to the technical field of positioning. The method provided by the invention includes the following steps: S1. a wireless receiving end receives a wireless signal transmitted by a wireless signal transmitter, and uploads wireless signal information to a server; S2. the server receives the wireless signal information of the wireless receiving end, and assesses channel state information; S3. the server detects positioning information of a human body according to change of the channel state information; and S4. the server returns a calculated positioning result to the wireless receiving end. The method also includes a parameter correcting step A: using a sensor carried by a human body to correct parameters used in an algorithm detected in Step S3. The method provided by the invention can perform accurate poisoning of a human body indoors, utilize a self-learning function of a system to handle a false alarm situation, and further reduce a false alarm rate; and can be used in any environment where a wireless network exists, and a detected person does not need to carry any extra sensing equipment, which is very convenient.

Human body indoor positioning automatic detection method and system

The invention relates to a fall-down detection method and system. The fall-down detection method includes receiving a first WiFi signal flow passing through the environment via a first reception antenna; receiving a second WiFi signal flow passing through the environment via a second reception antenna; determining the physical layer channel state information flow, namely a first CSI flow, of the first WiFi signal flow; determining the physical layer channel state information flow, namely a second CSI flow, of the second WiFi signal flow; determining the phase difference, namely the CSI phase difference, between the physical layer channel state information flow of the first WiFi signal flow and the physical layer channel state information flow of the second WiFi signal flow in the corresponding state at the same time to form a CSI phase difference flow; and determining the fall-down event based on the CSI flows and the CSI phase difference flow.

Fall-down detection method and system

The invention discloses a method of detecting the walking direction of a person through utilization of radio frequency signals A signal transmitting device Tx and two signal receiving devices R*1 and R*2 are respectively placed at known fixed positions; A connecting line between the Tx and R*1 is as perpendicular to a connecting line between Tx and R*2 as possible, and no block exists between the device pairs of Tx-R*1 and Tx-R*2 ; signals are subjected to OFDM modulation, and information of signal strength changes is obtained by measuring channel state information; and multi-frequency wireless channel changes are extracted through measuring radio frequency (RF) signals so as to obtain the walking direction of the person through calculating The technical scheme of the invention is not limited by the fact in sensing methods based on sensing devices that the walking direction of a person can only be detected when the person carries a sensing device The method can use existing wireless channel CSI values, so that no hardware change in wireless transceivers is needed, and the cost is saved The devices of the method can continuously operate in real time, and the method is advantaged by being convenient and practical, being small in detection error and high in precision

A method of detecting the walking direction of a person through utilization of radio frequency signals

The invention discloses a respiration detection method. The method can calibrate the position where human respiration can be detected. Human respiration at any position can be detected through RF by adjusting the position of RF receiving and transmitting equipment. The method includes the steps that the radio frequency signal receiving and transmitting equipment is placed, the spatial arrangement of Fresnel zones determined by the receiving and transmitting equipment is acquired, the human body is made to be located in the middle region of a certain Fresnel zone in the spatial arrangement of the Fresnel zones, and the respiratory rate is obtained through calculation according to radio frequency signals received by a receiving end. According to the technical scheme, the problem about how to achieve optimal respiration detection when the human body is located at different positions can be solved. In addition, the human body does not need to carry or adhere to any equipment, the advantages of being free of intrusion, convenient to use and low in cost are achieved, the detection result is definite and reliable, the prominent effect is achieved in the technical field of respiration detection, and the method is particularly suitable for long-term respiration detection.

Respiration detection method

The present invention relates to a fall detection method and system The fall detection method comprises: receiving, by a first receiving antenna, a first WiFi signal stream propagating through an environment; receiving, by a second receiving antenna, a second WiFi signal stream propagating through the environment; determining a physical layer Channel State Information (CSI) stream, namely, a first CSI stream, of the first WiFi signal stream; determining a physical layer CSI stream, namely, a second CSI stream, of the second WiFi signal; determining a phase difference, namely, a CSI phase difference, between respective phase of the physical layer CSI stream of the first WiFi signal stream and the physical layer CSI stream of the second WiFi signal stream at the same time point, to form a CSI phase difference stream; and determining, according to the CSI streams and the CSI phase difference stream, a fall event

Wang Yasha

Papers

Fall-down detection method and system

A method of detecting the walking direction of a person through utilization of radio frequency signals

Respiration detection method

Fall detection method and system