Detection of Frequency-Hopped Signals Embedded in Interference Waveforms

摘要

Many military communications systems utilize frequency-hopped spread spectrum waveforms to protect against jamming and enemy detection. These waveforms may be subjected not only to intentional jamming but may also be unintentionally jammed by other communications signals. While some systems can overcome inband interference with more signal power, covert systems may be limited to small amounts of transmitted power. The objective of this thesis was to investigate a method for resolving a frequency-hopped signal embedded in interference waveforms. With exponential averaging in the frequency domain, the spectra of the interfering signals can be estimated as long as they are present over a period longer than that of the frequency-hopped signal. Certain FFT sizes and weights are more beneficial to achieving this estimate than others. The interference estimate can be used to extract the desired frequency-hopped signal through spectral division of the received signal with the estimate. This technique is designated as noise-normalization. Simulations in MATLAB demonstrate the use of the technique and show how the desired signal can be resolved.