Experimental carrier detection of BPSK and QPSK direct sequence spread spectrum signals

This paper reports on an experimental investigation of the detection of BPSK and QPSK direct sequence (DS) spread spectrum signals above and below the noise floor using nonlinear carrier detection methods. These processes produce carrier harmonics 2f₀ and 4f ₀ for BPSK and QPSK signals respectively. The signal-to-noise ratio of these harmonics depends on the input signal signal-to-noise ratio, SNR (chip symbol energy/noise power density), normalized input bandwidth, η(input filter bandwidth/chip rate) and interceptor´s process gain, R/B (chip rate/detection bandwidth). Measurements agree well with analytic expressions. At low input SNR, the signal-to-noise ratio of the carrier harmonics is maximized for η=1.0 and varies as SNR² for BPSK signals and in the range of SNR³ to SNR⁴ for QPSK signals (the theoretical limit ∝SNR⁴ is only sometimes reached in practice.) QPSK signals generated from typical commercial quadrature modulators may also be detected at If ₀ from mixer LO leak-through and at 2f₀ from I/Q channel imbalance. For LPI use of QPSK DS signals, the LO leak-through must be carefully suppressed to <-55 dB and the two channels balanced within 0.2 dB. For a typical R/B=60 dB (10 MHz chip rate detected in a 10 Hz detection bandwidth), the threshold of detection is an SNR of -23 dB for BPSK signals and -8 dB for QPSK signals