A clutter suppression scheme for high frequenccy (HF) radar

Poon, Martin Wai Yee (1991) A clutter suppression scheme for high frequenccy (HF) radar. Masters thesis, Memorial University of Newfoundland.

[English] PDF (Migrated (PDF/A Conversion) from original format: (application/pdf)) - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (12MB) [English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. (Original Version)

Abstract

A clutter suppression scheme for high frequency (HF) radar is presented in this thesis. The HF radar has been developed for coastal surveillance and the remote sensing of the ocean. When the HF electromagnetic waves propagate over the ocean, the backscatter from the ocean surface has well defined frequencies, known as “Bragg" frequencies, shifted from the radar frequency. One of the characteristics of the HF backscatter is the high level of ocean clutter which hampers target detection. It is desirable to suppress the ocean clutter before target detection operation. The proposed scheme is developed based on the recognition of the time-varying behaviour of the ocean clutter that can be simply characterized by two narrowband frequency-modulated sinusoidal signals with their centre frequencies equal to the "Bragg" frequencies. The scheme consists of two parts. First, a time-varying technique referred to as Hankel rank reduction method is used to estimate the instantaneous frequencies of the clutter signals. The method states that a Hankel matrix of a time series data consisting of a finite number of sinusoids can be approximated to, via Singular Value Decomposition (SVD), a lower rank matrix defined by the finite number of the principal singular values, even if the frequencies of the sinusoids are varying slowly with time. The instantaneous frequencies are estimated from those principal singular values. The use of SVD is to decompose the Hankel matrix into a signal and a noise vector subspace. The signal subspace is identified by the largest singular values. Second, a process in which the frequency component of the clutter signals is removed from the reduced rank Hankel matrix instantaneously is developed to suppress the ocean clutter. Subsequently, another reduced rank Hankel matrix is constructed from which the target signal can be extracted. -- The performance of the scheme has been evaluated on computer-synthesized data and on some real data collected from a recently developed HF radar. The results from both cases showed that the instantaneous frequencies of the ocean clutter signals and the target signal were properly tracked by the Hankel rank reduction method and that a substantial level of the ocean clutter, in the range of 20 to 50 dB, could be suppressed by the scheme proposed.

Actions (login required)

View Item