A passive human body identification method based on narrow-band radio frequency link sampling

Abstract

The invention relates to a passive human body identification method based on narrowband radio frequency link sampling. The method comprises the following steps: collecting biological characteristic information generated by walking of a pedestrian to be identified through a plurality of radio frequency links, matching the biological characteristic information with a radio frequency signal receivingintensity sequence, fusing the radio frequency signal receiving intensity sequences of all the radio frequency links to obtain a fused radio frequency signal receiving intensity value vector; carrying out maximum likelihood probability estimation matching on the fusion radio frequency signal receiving intensity value vector and each Gaussian mixture model in an identity database, and carrying outidentity recognition on the to-be-recognized pedestrian according to an output probability value. According to the invention, biological characteristic information generated by walking of a pedestrian to be identified is sampled and acquired through the constructed multi-angle interleaved radio frequency links; forced cooperation and visual privacy invasion of a user in a limited environment areeffectively avoided, the dependence of correct recognition on illumination and different wearing clothes is remarkably reduced, and the advantages of being free of invasion, difficult to hide, difficult to camouflage and the like are achieved.

Description

technical field [0001] The invention relates to the technical field of biological feature identification by radio frequency technology, in particular to a passive human body identification method based on narrowband radio frequency link sampling. Background technique [0002] At present, obtaining biometric information represented by the inherent physiological structure and dynamic behavior of the human body has important research and application significance in the fields of identification and authentication, criminal investigation, and medical diagnosis. From the perspective of biological information acquisition methods, a large number of researches and applications can be divided into two categories: contact and non-contact. [0003] In the traditional contact or short-distance contact acquisition means, the collection of biometric information such as human fingerprints, palmprints, and irises has been proved to be able to perfectly match individual characteristics and ha...

AI Technical Summary

Problems solved by technology

However, the method of using visual equipment to obtain gait features has some inevitable limitations in the large-scale monitoring network environment: first, the image or video itself has spatio-temporal redundancy, foreground object extraction and information compression are necessary Network transmission and storage for data

Existing methods have good recognition efficiency for small groups of people, but it is difficult to distinguish people with similar height and body shape

In addition, in the case of occlusion, the recognition rate is difficult to meet the demand

Images