Crypto-acoustic method based on MVDR

Abstract

The invention discloses a crypto-acoustic method based on MVDR. According to the method, firstly, short-time Fourier transform is utilized, a time domain signal received by a sensor array is decomposed into frequency domain sub-band signals, an MVDR-based crypto-acoustic filter is constructed on each sub-band, acoustic signals from a sound source within the crypto-acoustic distance are made to pass through the crypto-acoustic filters without attenuation, and lastly, an estimated signal is finally obtained through inverse STFT. The method is advantaged in that the relatively good crypto-acoustic effect is realized.

Description

technical field [0001] The invention relates to a sound hiding technology of a microphone array, in particular to an MVDR-based sound hiding method. Background technique [0002] The research on hidden sound technology has a long history. People have mainly explored two methods in the field of acoustic signal sensing: bone conduction microphone and ultrasonic microphone, and explored signal separation technology and differential microphone array in the field of acoustic signal processing. [0003] Bone conduction microphones use the slight vibration of the head and neck bones caused by human speech to collect sound signals and convert them into electrical signals. Because it is different from traditional microphones that pick up sound through air conduction, it can also transmit high-definition sound in a very noisy environment. As early as hundreds of years ago, people had many practical applications of bone conduction technology, but the voice quality of early bone conduc...

AI Technical Summary

Problems solved by technology

Although bone conduction microphone technology has made a breakthrough, the current promotion of its communication earphones is not very ideal. There are several problems: 1) poor dynamic effect; 2) high cost; 3) poor packaging effect; 4) poor sound quality

However, the separation performance achieved by the current separation technology is still very limited, which cannot meet the needs of hidden sound applications.

Images