Abstract

Frequency domain and multiple correlation analysis of sonar signals have not been extensively used in the past. A possible reason may be that the level of processing required is high. However, with the availability of digital systems and single chip digital signal processors with a built in microcomputer chip that can be preprogramed to perform some specific signal processing function, frequency domain and correlation analysis can be an attractive new development area for sonar applications. Using frequency domain or correlation processing of the sonar signal can result in two main advantages, first a true 3-D image can be generated without having to use multiple beams and second the sonar signal need not be of the usual high frequency pinging type, but band limited white noise can be used. There is an advantage in using band limited white noise over pinging type noise in that it is more difficult to detect. The principle of operation and the signal processing required of such a 3-D imaging system can be outlined as follows. A sector is insonified with band limited white noise by an electronically scanned beam with a narrow azimuth beamwidth and a wide vertical beamwidth. The backscattered signal is received by a multi-element linear transducer where the signal from each element is independently processed. Thus for each section of the insonified sector a matrix of space-time data is continuously being received. Cross spectral density functions are first evaluated between the received signals by each element of the receiver transducer and the transmitted signal. That is the space-time matrix is transformed to a space-frequency matrix. Two additional forward Fourier transforms are then applied to the data, one along the frequency domain and one along the space domain. The resulting matrix contains angle/range information for the particular section of the sector determined from the angle of the projected acoustic beam. A similar procedure is implemented if cross correlation analysis is used. In this paper the implementation of these techniques are described together with some evaluation of the requirements of the transducers to cover a desired range and vertical and horizontal view.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call