Abstract
We compare different combinations of the repetition diversity order L and code rate R for turbo-coded Frequency-Hopped Spread-Spectrum (FH/SS) communication systems in the presence of fading and partial-band Gaussian interference. For a fixed overall channel code rate R/L we show that using the lowest code rate and no repetition diversity always performs better than using a higher code rate and some repetition for both coherent and non-coherent schemes. We then propose a simple maximum-likelihood-based method for signal-to-noise-ratio (SNR) estimation in Non-Coherent Binary Frequency Shift Keying (NCBFSK) without training symbols. Except for impractically small hop sizes of 8 bits or less we obtain performance virtually equal to that of perfect SNR knowledge but with much less complexity than iterative schemes previously proposed. For the case of Coherent Binary Phase Shift Keying (CBPSK) we derive the Expectation Maximization (EM) estimate of the SNR without training symbols and iteratively feed the estimator with the extrinsic information from the turbo decoder. The performance for CBPSK is near that of perfect SNR knowledge for hop sizes of 64 bits or more. Unlike previously proposed methods for CBPSK the EM estimate of SNR does not require knowledge of the noise and interference variance, received bit energy, or the fading channel model.
Paper version not known (
Free)
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call