Adaptive Sparsity Based Channel Estimator for 1-Bit ADC MIMO-Constant Envelope Modulation

Abstract

The 1-bit ADC multi-input-multi-output constant envelope modulation (MIMO-CEM) was introduced to acclimate with the power constrains of the new cutting edge wireless communication technologies (i.e. IoT and wireless sensor networks). Despite of its ability to fulfil the PHY layer constrains of those technologies (i.e. high-power efficiency and low hardware complexity), the 1-bit ADC MIMO-CEM received signal is exposed to severe quantization distortion. This quantization distortion is sorely affected on the efficiency of the MIMO-CEM channel estimation since, estimation operation depends on the received signal amplitude information. This is in turn adds serious threat to the MIMO-CEM decoder quality which, is depending on the accuracy of the estimated channel. Therefore, a low complex and spectral efficient MIMO-CEM channel estimation algorithm is proposed in this paper. In which, the MIMO-CEM channel sparsity is exploited to propose an adaptive compressive sensing (ACS) channel estimation algorithm to cope with the 1-bit ADC severe quantization distortion. In the proposed ACS, the sensing (measurement) matrix is adaptively updated (changed) and resized in order to speed up (low computational complexity) the channel estimation process with minimum error constrain at low preamble signal length. However, the proposed algorithm is spectrally efficient (very low preamble signal length) and it is computational complexity is very low compared to the recently MIMO-CEM channel estimators. Eventually, the proposed algorithm introduces computational complexity reduction up to 90 % and it gives up to 72 % spectrum saving compared to the conventional MIMO-CEM channel estimator.