Multichannel HHT (Hilbert-Huang Transform) based UWB (Ultra Wideband) radar human body motion micro-Doppler feature extraction method

Abstract

The invention discloses a multichannel HHT (Hilbert-Huang Transform) based UWB (Ultra Wideband) radar human body motion micro-Doppler feature extraction method, which comprises the steps of 1, acquiring UWB radar signals; 2, preprocessing the acquired UWB radar signals; 3, performing effective channel selection; and 4, carrying out multichannel time-frequency spectrum fusion. The micro-Doppler time-frequency analysis method provided by the invention has higher time-frequency resolution, micro-Doppler features corresponding to different components are more obvious in component differentiation, and transient features and detail changes of human body motions can be better reflected. The time-frequency analysis method provided by the invention is high in anti-interference ability, and obvious micro-Doppler features can still be captured under the conditions of long-distance through-wall detection and sharp signal attenuation. The time-frequency analysis method provided by the invention has better denoising capability, the acquired time-frequency spectrum is higher in signal-to-noise ratio, and micro-Doppler feature components are more obvious.

Description

technical field [0001] The invention belongs to the technical field of biological radar or radar-type life detection and identification, and particularly relates to a UWB radar human motion micro-Doppler feature extraction method based on multi-channel HHT. Background technique [0002] Ultra Wideband (UWB) radar, as a new type of non-contact life detection technology, has high range resolution and anti-interference, and can penetrate non-metallic media (brick walls, ruins, etc.) of a certain thickness , long-distance, non-contact detection and identification of human target movement characteristics, gradually used in anti-terrorism, hostage rescue, urban street fighting, border security and post-disaster search and rescue, etc. [0003] The micro-Doppler feature means that in addition to the movement of the main body of the target, the micro-motion of each component will produce differential Doppler modulation on the radar wave. The micro-motion information contained in the...

AI Technical Summary

Problems solved by technology

However, for the above-mentioned application scenarios of penetration detection, the current micro-Doppler analysis and extraction methods of human motion ultra-wideband radar still have the following problems: the target motion micro-Doppler features will be distributed in multiple range units of the ultra-wideband radar signal, At present, there is no method for comprehensive analysis and extraction of micro-Doppler features for ultra-wideband radar; the micro-Doppler attenuation is rapid after the target moves through the wall, and the micro-Doppler features will interact with strong wall clutter and background noise. Interleaved or even submerged, making the signal-to-noise ratio of micro-Doppler features extremely low and difficult to extract; the traditional time-frequency transform extraction method for micro-Doppler features has a low time-frequency resolution, and the corresponding micro-Doppler features caused by different components indistinguishable extraction

Images