Convolutionally encoded FFH-MFSK: demodulation and frequency tracking over frequency-varying jammed channels

A novel soft decision metric for the Viterbi demodulation and decoding of convolutionally encoded fast-frequency-hopped M-ary frequency shift keying (FFH-MFSK) is described. The effectiveness of two possible hop-by-hop AGC methods is evaluated for use with metric calculation methods to optimise the worst-case partial-band-jamming performance. The metric is shown to improve the worst-case jamming bit-error-rate (BER) performance for M>or=8 in these situations and to alter the worst-case jamming scenario from one of sparsely separated tones to either full-band AWGN or densely spaced jamming tones. The worst-case partial-band tone jamming BER performance is consistently close to the broadband AWGN case. The Viterbi frequency-tracking method is shown to provide fast tracking of frequency offsets whilst degrading the BER performance of 8-FSK in AWGN or worst-case partial-band tone jamming by 0.5 dB for a tracking range of 16 possible carrier-frequency positions.<>