Passive detection of scatterer using autocorrelation of surf noise

Abstract

When a scatterer is located in a diffuse noise field, time domain Green's function between two different receivers can be extracted from cross-correlation of ambient noise which is recorded by the two receivers so that target detection can be implemented. However, the method based on cross-correlation strongly depends on timing synchronization of each receiver, otherwise there will be a time drift in the cross-correlation result, which can bring error in the positioning detection. Besides, two receivers that are far from each other must communicate with each other to implement cross-correlation in real-time data processing, but big data transmission is difficult in the ocean. Compared with cross-correlation, autocorrelation means that each receiver works independently and only the final autocorrelation result is to be transmitted. Actually, the scattered wave of target is always so weak that it is submerged in the autocorrelation result of the ambient noise. In this paper, we propose a method of processing the autocorrelation of the ambient noise. When the averaging noise autocorrelation of all receivers is subtracted from the autocorrelation result of the noise recorded by each receiver, the signalnoise ratio of the scattered wave will be significantly enhanced. With the help of Kirchhoff migration algorithm, detection of a scatterer can be implemented. We have conducted a scatterer passive detection experiment in Shilaoren beach, Qingdao, and accurately detected the position of a polyviny chloride pipe (about 8 m away from the nearest receiver) using only 12 min surf noise data. The experimental result shows that the processing of autocorrelation could replace cross-correlation in passive target detection when the ambient noise is time steady and the statistical characteristics of the background noise at different receivers are the same. Unlike Green's function extracted from cross-correlation of ambient noise, each receiver can work independently without considering the problems of massive data transmission and timing synchronization, which may be suitable for target detection using multi-receivers and mobile platform.