A wireless local area network flow analysis method and system

Abstract

The invention provides a wireless local area network flow analysis method and system. The method comprises the following steps that: equipment is uniquely determined by utilizing a radio frequency identification technology according to radio frequency fingerprint characteristics of the equipment; the fingerprint characteristics of the equipment are applied and used as identification identifiers ofthe equipment, and user sensitive data such as MAC and IP are not used as the identification identifiers, so that the privacy of a user is protected to a great extent; based on machine learning, a ReliefF algorithm is adopted to carry out feature selection on parameter vectors of collected data, a PCA algorithm is utilized to carry out dimension reduction on the feature vectors, and finally a K nearest neighbor algorithm is adopted to carry out classification identification, so that the test training time is shortened, and meanwhile, the identification speed and accuracy of equipment are improved.

Description

technical field [0001] The present invention relates to the technical field of media communication, and in particular, to a method and system for analyzing the flow of a wireless local area network. Background technique [0002] In the existing WIFI network traffic analysis, the method of identifying a device generally adopts the wifi probe technology. The wifi probe actually refers to the probe frame. When we generally connect to a wireless network, we first need to select the corresponding wireless network, that is, according to the name of the wireless network (SSID). Well, there are two ways to know the name, active scanning and passive scanning, in which the probe frame is used in the technology of active scanning. The device node actively sends the probe request frame to the AP (that is, the route), the AP then feeds back the probe response, and then uses some information in the frame, the node continues to complete the WIFI network access process. In this way, the ...

AI Technical Summary

Problems solved by technology

[0008] 1. Every legitimate user needs to be added to the RF fingerprint feature vector white list. When the number of users increases or the user group changes frequently, updating the RF fingerprint feature vector white list and user management of the white list will become difficult. Frequent changes Legitimate users add more workload to whitelist management

[0009] 2. The uniqueness of the legitimate user needs to be guaranteed by both the radio frequency fingerprint feature and the legal user's biometric fingerprint feature. If a legitimate user changes the device, it may happen that the radio frequency fingerprint feature is not in the white list and the legitimate user cannot access the wireless network.

[0010] 3. The entire system intervention process needs to enter user information and biometric fingerprints, which is equivalent to obtaining part of the user's private information. The protection of private information increases the burden on the system, and the security of private information has become the entry point for the entire system to be attacked

Although this patent has the characteristics of small delay and low complexity, it needs to manually set the threshold, and the selection of the threshold is difficult.

Images