A Receiver-Oriented Rate-Adaptation Strategy for Improving Network Efficiency in Mobile Ad-Hoc Networks

This article investigates the efficacy of a novel receiver-oriented adaptive-rate system for sustaining the quality of a randomly fluctuating wireless link at a prescribed level in generalized multipath fading channels. Multilevel hierarchical modulations inherently have a limited ability to facilitate passive rate-adaptation, and this attribute is exploited to improve the effective bit transmission rate at low and moderate carrier-to-noise ratios. In our scheme, the multiresolution source data is first matched to the unequal bit error protections of a multiresolution modulation, and in the presence of adverse channel conditions, the "less important" bits from the demodulated symbols are dropped in order to maintain the link bit error rate at the desired level. One of the salient features of our proposed link adaptation strategy is that it neither rely nor require any feedback of channel quality estimates from the receiver to the sender, and the performance improvement is achieved without dynamically altering the signal constellation size at the transmitter or by using error control coding with incremental redundancy. Moreover, the proposed link adaptation strategy can still be effective even in a rapidly time-varying environment since the received signal strength of each symbol at the demodulator input is used to decide on which demodulated bits of that symbol should be retained, and thus our rate adaptation is operated at the symbol level. The duality between the problems of finding 'optimal signaling in a delay constrained or unknown time-varying channels' and 'optimal signaling for broadcast/multicast channels' is also established.