Analysis of low probability of intercept (LPI) radar signals using cyclostationary processing

Abstract

LPI radar is a class of radar systems possessing certain performance characteristics that make them nearly undetectable by today's digital intercept receivers. This presents a significant tactical problem in the battle space. To detect these types of radar, new digital receivers that use sophisticated signal processing techniques are required. This thesis investigates the use of cyclostationary processing to extract the modulation parameters from a variety of continuous-wave (CW) low-probability-of-intercept (LPI) radar waveforms. The cyclostationary detection techniques described exploit the fact that digital signals vary in time with single or multiple periodicity, owing to their spectral correlation, namely non-zero correlation between certain frequency components, at certain frequency shifts. The use of cyclostationary signal processing in a non-cooperative intercept receiver can help identify the particular emitter and aid in the development of electronic attack signals. LPI CW waveforms examined include Frank codes, P1 through P4, Frequency Modulated CW (FMCW), Costas frequencies as well as several frequency-shift-keying/phase-shift-keying (FSK/PSK) waveforms. This thesis show that for signal-to-noise ratios of 0 dB and -6 dB, the cyclostationary signal processing can extract the modulation parameters necessary in order to distinguish between the various types of LPI modulations.