SPARSE MMSE EQUALIZER FOR GTR-STBC IN AERONAUTICAL TELEMETRY

Abstract

This paper investigates the performance of sparse minimum mean squared error (MMSE) equalizer for generalized time-reversed space-time block codes (GTR-STBC) in aeronautical telemetry. GTR-STBC equipped with MMSE equalizer performs the best trade-off between the signal-tonoise ratio and inter-symbol interference by allocating unequal power over aeronautical telemetry channels. However, aeronautical telemetry channels are in general consists of larger delay spreads which make the MMSE equalization of aeronautical channels with GTR-STBC computationally complex. Interestingly enough, in spite of larger delays aeronautical channels are made of few sparsely distributed multipaths and therefore their MMSE equalizers are highly compressible. In this paper, compressed sensing based greedy algorithm is used for the design of sparse MMSE equalizer and a convex curve-fitting algorithm is used to find the sub-optimum power allocation parameter at the same sparsity level for GTR-STBC. Our simulation results show that 75-90% of the non-zero equalizer taps can be reduced with a slight relaxation of the mean-squared error (or equivalentlyslight degradationof bit-errorrate performance). It isalso observedthat the optimum transmitter power profile for the sparse MMSE equalizer is different than that of the non-sparse equalizer.