Abstract

In this work the effect of perfect and imperfect synchronization on the performance of single-link and cooperative communication is investigated. A feedforward nondata-aided near maximum likelihood (NDA-NML) timing estimator which is effective for an additive white Gaussian noise (AWGN) channel and also for a flat-fading channel, is developed. The Cramer Rao bound (CRB) and modified Cramer Rao bound (MCRB) for the estimator for a single-link transmission over an AWGN channel is derived. A closed form expression for the probability distribution of the timing estimator is also derived. The bit-error-rate (BER) degradation of the NDA-NML timing estimator with raised cosine pulse shaping for static timing errors over an AWGN channel is characterized. A closed form expression is derived for the conditional bit error probability (BEP) with static timing errors of binary phase shift keying modulation over a Rayleigh fading channel using rectangular pulse shaping. The NDA-NML timing estimator is applied to a cooperative communication system with a source, a relay and a destination. A CRB for the estimator for asymptotically low signal-to-noise-ratio case is derived. The timing complexity of the NDA-NML estimator is derived and compared with a feedforward correlation based data-aided maximum likelihood (DA-ML) estimator. The BER performance of this system operating with a detect-and-forward relaying is studied, where the symbol timings are estimated independently for each channel. A feedforward data and channel aided maximum likelihood (DCA-ML) symbol timing estimator for cooperative communication operating over flat fading channels is then developed. For more severe fading the DCA-ML estimator performs better than the NDANML estimator and the DA-ML estimator. The performance gains of the DCA-ML estimator over that of the DA-ML estimator become more significant in cooperative transmission than in single-link node-to-node transmission. The NDA-NML symbol timing estimator is applied to three-node cooperative communication in fast flat-fading conditions with various signal constellations. It is found that timing errors have significant effect on performance in fast flat-fading

Full Text
Published version (Free)

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