Gaussian-Distributed Spread-Spectrum for Covert Communications.

Area of Science:

• Electrical Engineering
• Information Security
• Signal Processing

Background:

• Covert communication is vital for secure wireless transmissions in military and commercial sectors.
• Direct-sequence spread-spectrum (DSSS) offers some security but has exploitable cyclostationary properties.
• Adversaries can detect and exploit DSSS signals using spectral analysis, increasing vulnerability to jamming.

Purpose of the Study:

• To propose a new covert communication scheme, Gaussian distributed spread-spectrum (GDSS), to overcome DSSS limitations.
• To randomize transmitted signals, reducing cyclic features and mimicking thermal noise.
• To evaluate GDSS performance against DSSS using various detection methods without receiver knowledge of noise parameters.

Main Methods:

• Developed the Gaussian distributed spread-spectrum (GDSS) scheme to generate noise-like signals.
• Employed three detectors: high-order moments, modulation stripping, and spectral correlation.
• Assessed signal detectability under varying signal-to-noise ratios (SNRs) and spreading factors.

Main Results:

• The moment-based detector failed to detect GDSS signals at any SNR (N=256), unlike DSSS up to -12 dB.
• Modulation stripping showed no significant phase convergence for GDSS, similar to noise-only signals.
• Spectral correlation analysis revealed no identifiable peaks for GDSS at -12 dB SNR, confirming its covert nature.

Conclusions:

• GDSS effectively generates noise-like signals with reduced detectable features, enhancing covert communication security.
• GDSS proves superior to DSSS in resisting detection by common signal analysis techniques.
• The GDSS scheme incurs an approximate 2 dB SNR penalty for its enhanced security.