Review and analysis of DOA estimation techniques for wideband LFM signals: Extended report

Date

2018

Authors

Mulinde, R.
Dissanayake, D.W.
Ho, S.W.
Chan, T.
Attygalle, M.
Aziz, S.M.

Editors

Advisors

Journal Title

Journal ISSN

Volume Title

Type:

Report

Citation

Statement of Responsibility

Conference Name

Abstract

This report presents the results from phase 1 of the collaborative research project conducted by University of South Australia (UniSA) and the Systems Group of the Cyber and Electronic Warfare Division of the Defence Science and Technology Group (DSTG). The primary focus of this phase is to identify beamforming techniques suitable for identifying wideband signals of interest (SOI) in real-time. Particular requirements include achieving high detection gain in identifying signals with low signal-to-noise ratio (SNR) and low probability of interception. Among the latter, wideband linear frequency modulated (LFM) or chirp signals are very prominent. Hence this phase of the project investigates efficient techniques to identify wideband LFM signals with high detection gain. The problem can be defined as one of estimating the direction of arrival (DOA) of the signals of interest using an array of spatially distributed sensors or array elements. In this project, a uniform lineararray (ULA) is considered. The goal is to perform spatial filtering on the signals received by the antenna elements to enhance the signals of interest while suppressing the unwanted interfering signals and noise or clutter.A large body of literature exists in the area of narrowband beamforming, where the received signals only occupy a small fraction of the available bandwidth and they do not vary with time or vary very slowly. In contrast, for wideband signals, capturing the variation of signal with frequency is an important aspect. Many of the conventional techniques proposed for wideband beamforming are modifications/extensions of the narrowband beamforming techniques. These techniques consider processing of the received signals either in time domain or frequency domain These techniques assume that the signals of interest are stationary. However, this assumption is not valid for LFM or chirp signals considered in this project, and therefore conventional DOA estimation techniques yield low spatial resolution. Hence, estimation techniques that involve processing the combination of time-frequency distributions (TFDs) and array processing to exploit time, frequency as well as spatial diversity of the received signals would be of special interest for this phase of the project.To aid with identifying suitable techniques and algorithms for detection of wideband LFM signals, this phase of the project conducted an extensive review of conventional beamforming literature. The first report of this phase (D1) of the project contained this review. Further review was conducted during the course of the project while also analysing the detection performance and computational complexity of candidate algorithms in parallel. This report briefly summarises the conventional beamforming algorithms for completeness and also presents techniques suitable for detection of wideband LFM signals

School/Discipline

Dissertation Note

Provenance

Description

Access Status

Rights

Copyright 2018 The Authors

License

Grant ID

Published Version

Call number

Persistent link to this record