Acousto-optic processors for the detection of spread spectrum radar signals

Newer radar systems using such techniques as direct-sequence phase modulation, frequency hopping and linear FM (chirp) require large transmission bandwidth and increasingly sophisticated receivers to intercept and classify their returns. These signals are difficult to distinguish from background noise and are hence said to have low probability of intercept (LPI). Current electronic receiver systems work well at detecting and characterizing narrowband pulse signals but are not very effective with spread spectrum signals. Acousto-optic (AO) processors have shown great potential for dealing with wideband signals and offer the ability to detect and analyze various LPI waveforms. In this thesis we present a theoretical, numerical and an experimental study of a time integrating acousto-optic processor, with an electronically inserted reference tone, to detect and characterize linear FM and frequency hopped spread spectrum LPI signals corrupted by additive noise and narrowband interferers. Electronic and optical components were designed and the processor assembled. Analytical expressions for chirp and frequency hopped signal correlations with the processor were then derived and numerically simulated. The processing gain of the processor was subsequently derived using a stochastic analysis approach for a number of SNR scenarios. It is shown that for relatively low SNR intercept signals, the laser noise is de-emphasized leading to a processing gain that is proportional to the square of the detector dynamic range. At high SNR levels, the laser intensity noise becomes the predominant factor. Experimental evaluation of the noise loading and effect of the narrow-band interferers on the processor output were then carried out. A near real-time method based on digitally tunable notch filters was developed to excise the narrowband interferer energy prior to correlation. The method uses a unique space-integrating electronically inserted reference tone arrangement. Finally, a novel way of estimating the overall optical misalignment in the processor hardware was proposed and demonstrated. It represents a simple and robust alternative to many purely optical procedures. The resolution accuracy of the method is shown to be limited by the CCD pixel dimensions. As a side result, the scaling law that applies to the output signal when using CCD detection was also derived.