A method, device and system for detecting home intrusion based on WiFi signals

By setting up wireless AP devices within the range of wireless routers, using ICMP packets to obtain channel information and combining machine learning, the problem of expensive or insufficient sensitivity of existing non-visual intrusion detection devices is solved, and real-time camera-free and wireless signal oscilloscopes are realized , accurate intrusion detection, providing a cheap, stable and privacy-free intrusion identification solution.

CN113132992BActive Publication Date: 2025-01-10NANJING INTELLIGENT INNOVATION TECH CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN201911388830.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2019-12-30
Publication Date
2025-01-10
Estimated Expiration
2039-12-30

AI Technical Summary

Technical Problem

Existing non-visual intrusion detection equipment is expensive or insufficient detection sensitivity, which cannot effectively meet the needs of home intrusion detection.

Method used

By setting up a wireless AP device within 6m range of the wireless router, ICMP packets are used to continuously send and parse the channel information of RSSI and CSI, calculate the stability of the signal-to-noise ratio, and judge whether an intrusion occurs in combination with machine learning methods to realize intrusion detection .

Benefits of technology

It realizes the function of real-time and accurate identification of intrusions without the need for a camera or wireless signal oscilloscope. The device structure is simple, cheap and has no privacy issues, improving the stability and accuracy of detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113132992B_ABST
    Figure CN113132992B_ABST
Patent Text Reader

Abstract

A method, device and system for detecting home intrusion based on WiFi signals, wherein a wireless AP device is set within 6m of a wireless router, and a circle formed by the line connecting the wireless AP device and the router device as the diameter is located in the area where the intruder is predicted to pass; the wireless AP device is connected to the wireless network of the wireless router, the wireless AP device continuously sends ICMP data packets to the wireless router device at equal time intervals, and the wireless router returns the ICMP data packets to the wireless AP device, the wireless AP device parses the returned ICMP packet, obtains the time series data composed of the channel information data of RSSI and CSI, processes the time series data, calculates the stability of the signal-to-noise ratio in the time series data, and determines whether an intrusion has occurred. The invention has a simple structure, a cheap device, no privacy issues, and can accurately identify the occurrence of intrusion.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of network security, relates to home security technology, and is a home intrusion detection method based on WiFi signals. Background Art

[0002] Intrusion detection is an important means to protect the personal safety and property safety of family members. With the development of economy and the advancement of science and technology, intrusion detection in home environment has received more and more attention.

[0003] Existing intrusion detection technologies can be divided into two categories: vision-based intrusion detection and non-vision-based intrusion detection. Vision-based intrusion detection mainly relies on camera monitoring and image recognition of videos. The detection requires a lot of computation and requires cameras with night vision at night, which raises privacy issues. Among non-vision detection methods, the equipment based on WiFi raw signals is too expensive, and the method based on WiFi signal strength has low accuracy and insufficient sensitivity. Summary of the invention

[0004] The problem to be solved by the present invention is that the existing non-visual intrusion detection equipment is expensive or has insufficient detection sensitivity and cannot meet the use requirements.

[0005] The technical solution of the present invention is: a home intrusion detection method based on WiFi signals, wherein a wireless AP device is set within 6m of a wireless router, and a circle formed by a line connecting the wireless AP device and the router device as a diameter is located in an area where an intruder is predicted to pass; the wireless AP device is connected to the wireless network of the wireless router, the wireless AP device continuously sends ICMP data packets to the wireless router device at equal time intervals, and the wireless router returns the ICMP data packet to the wireless AP device, the wireless AP device parses the returned ICMP packet, obtains time series data composed of RSSI and CSI channel information data, processes the time series data, calculates the stability of the signal-to-noise ratio in the time series data, and determines whether an intrusion has occurred.

[0006] Furthermore, the calculation of the stability of the signal-to-noise ratio is as follows: the phase and amplitude representation of the complex representation of each subcarrier of each antenna of the CSI is converted into a signal-to-noise ratio SNR, the maximum and minimum SNR of each subcarrier within a period of time are calculated, the variance of the maximum and minimum values ​​within a set time is calculated, the variance of the RSSI of each antenna within a set time is calculated, and the two obtained variances are used as features, and a machine learning method is used for training, learning and judgment, and the obtained learning model is used to determine whether an intrusion occurs based on real-time features.

[0007] The present invention also proposes a home intrusion detection device based on WiFi signals, including a wireless router and a wireless AP device. The wireless router is used to provide a wireless network and return corresponding ICMP data packets to the ICMP data packets sent by the wireless AP device; the wireless AP device is used to continuously send ICMP data packets to the wireless router device at equal time intervals, and parse the ICMP data packets returned by the wireless router to obtain channel information.

[0008] Furthermore, the wireless AP device is also used to send the obtained channel information to a designated external data processing platform, and the external data processing platform is used to process the channel information to determine whether an intrusion occurs.

[0009] As a preferred method, the wireless AP device includes a wireless AP mainboard with a PCIE interface, a wireless network card, a power interface and three antennas. The wireless network card is inserted into the PCIE interface of the wireless AP mainboard. The wireless network card is used to send ICMP data packets to the wireless router. The power interface is connected to the wireless AP mainboard, and the antenna is connected to the wireless network card.

[0010] The present invention also proposes a home intrusion detection system based on WiFi signals, comprising the above-mentioned intrusion detection device and a smart phone, wherein the intrusion detection device is used to obtain the channel information of RSSI and CSI according to the above-mentioned detection method, and the smart phone is used to process the channel information of RSSI and CSI according to the above-mentioned detection method to determine whether an intrusion occurs.

[0011] Although some intrusion identification methods based on CSI data have been proposed in the prior art, the identification processing of such methods is relatively simple and may lead to unstable identification. The detection method proposed in the present invention comprehensively applies the channel information of RSSI and CSI to stably identify the occurrence of intrusion. The present invention provides a device and method for intrusion detection, which can perform real-time intrusion detection and alarm without the need for equipment such as cameras and wireless signal oscilloscopes. It has a simple structure, a cheap device, no privacy issues, and can accurately identify the occurrence of intrusion. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 The figure is a flow chart of the intrusion detection method according to an embodiment of the present invention.

[0013] Figure 2 The figure is a flow chart of channel information processing in intrusion detection according to an embodiment of the present invention.

[0014] Figure 3 FIG. 4 is a schematic diagram of a wireless AP device in an embodiment of the present invention. DETAILED DESCRIPTION

[0015] like Figure 1As shown, the present invention proposes a home intrusion detection method based on WiFi signals, in which a wireless AP device is set within 6m of a wireless router, and a circle formed by a line connecting the wireless AP device and the router device as a diameter is located in an area where an intruder is predicted to pass; the wireless AP device is connected to the wireless network of the wireless router, the wireless AP device continuously sends ICMP data packets to the wireless router device at equal time intervals, and the wireless router returns ICMP data packets to the wireless AP device, the wireless AP device parses the returned ICMP packet, obtains time series data composed of RSSI and CSI channel information data, processes the time series data, calculates the stability of the signal-to-noise ratio in the time series data, and determines whether an intrusion has occurred.

[0016] like Figure 2 As shown, the calculation of the stability of the signal-to-noise ratio is as follows: the phase and amplitude representation of the complex representation of each subcarrier of each antenna of the CSI is converted into a signal-to-noise ratio SNR, the maximum and minimum values ​​of the SNR of each subcarrier within a period of time are calculated, the variance of the maximum and minimum values ​​within a set time is calculated, the variance of the RSSI of each antenna within a set time is calculated, and the two obtained variances are used as features, and a machine learning method is used for training, learning and judgment. The obtained learning model is used to determine whether an intrusion occurs based on real-time features.

[0017] The present invention also proposes a home intrusion detection device based on WiFi signals, including a wireless router and a wireless AP device, wherein the wireless router is used to provide a wireless network and return a corresponding ICMP data packet to the ICMP data packet sent by the wireless AP device; the wireless AP device is used to continuously send ICMP data packets to the wireless router device at equal time intervals, and parse the ICMP data packets returned by the wireless router to obtain the original data including the signal strength and channel information of RSSI and CSI. The wireless AP device is also used to send the obtained channel information to a designated external data processing platform, and the external data processing platform is used to process the channel information, determine whether an intrusion has occurred, and issue an alarm. The external data processing platform can be a networked data device such as a user's smart phone.

[0018] The wireless AP device includes a wireless AP motherboard with a PCIE interface, a wireless network card, a power interface and three antennas. Figure 3As shown, the wireless network card is inserted into the PCIE interface of the wireless AP motherboard. The wireless network card is used to send ICMP data packets to the wireless router. The power interface is connected to the wireless AP motherboard, and the antenna is connected to the wireless network card. The wireless AP device must be within 6m of the wireless router so that the wireless AP device can receive the data packets returned by the wireless router when collecting data. The three antennas are generally fixed upward to ensure that when the wireless AP device collects RSSI and CSI information, the specific settings are set based on the principle that the signal is not blocked.

[0019] The present invention also proposes a home intrusion detection system based on WiFi signals, comprising the above-mentioned intrusion detection device and a smart phone, wherein the intrusion detection device is used to obtain the channel information of RSSI and CSI according to the above-mentioned detection method, and the smart phone is used to process the channel information of RSSI and CSI according to the above-mentioned detection method to determine whether an intrusion occurs.

Claims

1. A method for detecting home intrusion based on WiFi signals, characterized in that Set up a wireless AP device within 6 meters of the wireless router, and the circle formed by the line connecting the wireless AP device and the router device is located in the area where the intruder is predicted to pass; The wireless AP device is connected to the wireless network of the wireless router. The wireless AP device continuously sends ICMP data packets to the wireless router device at equal time intervals, and the wireless router returns ICMP data packets to the wireless AP device. The wireless AP device parses the returned ICMP packets, obtains the time series data composed of RSSI and CSI channel information data, processes the time series data, calculates the stability of the signal-to-noise ratio in the time series data, and determines whether an intrusion has occurred; The calculation of the stability of the signal-to-noise ratio is as follows: the phase and amplitude representation of the complex representation of each subcarrier of each antenna of the CSI is converted into a signal-to-noise ratio SNR, the maximum and minimum SNR of each subcarrier within a period of time are calculated, the variance of the maximum and minimum values ​​within a set time is calculated, the variance of the RSSI of each antenna within a set time is calculated, and the two obtained variances are used as features, and a machine learning method is used for training, learning and judgment. The obtained learning model is used to judge whether an intrusion occurs based on real-time features.

2. A home intrusion detection system based on WiFi signals, characterized in that It includes an intrusion detection device and a smart phone, wherein the intrusion detection device is used to obtain RSSI and CSI channel information according to the detection method of claim 1, and the smart phone is used to process RSSI and CSI channel information according to the detection method of claim 1 to determine whether an intrusion occurs; The intrusion detection device includes a wireless router and a wireless AP device. The wireless router is used to provide a wireless network and return corresponding ICMP data packets to the ICMP data packets sent by the wireless AP device; the wireless AP device is used to continuously send ICMP data packets to the wireless router device at equal time intervals, and parse the ICMP data packets returned by the wireless router to obtain channel information.

3. A home intrusion detection system based on WiFi signals according to claim 2, characterized in that The wireless AP device is also used to send the obtained channel information to a designated external data processing platform, and the external data processing platform is used to process the channel information to determine whether an intrusion occurs.

4. A home intrusion detection system based on WiFi signals according to claim 2 or 3, characterized in that The wireless AP device includes a wireless AP mainboard with a PCIE interface, a wireless network card, a power interface and three antennas. The wireless network card is inserted into the PCIE interface of the wireless AP mainboard. The wireless network card is used to send ICMP data packets to the wireless router. The power interface is connected to the wireless AP mainboard, and the antenna is connected to the wireless network card.

Citation Information

Patent Citations

  • Coal mine underground intrusion detection method based on channel state information and random forest

    CN108631890A