Compressive sensing and reception for MIMO-OFDM based cognitive radio

Abstract

This paper explores a novel receiver architecture for multiple-input multiple-output (MIMO) and orthogonal frequency division multiplexing (OFDM) based cognitive radio (CR) that utilizes compressive sensing (CS) technique. Assuming that a limited number of subcarriers are used simultaneously in one MIMO-OFDM channel, we show that the conventional MIMO receiver can be replaced with the proposed receiver to compressively sample signals. In the proposed reception architecture, signals from multiple antennas are mixed and sampled with less hardware by exploiting the sparsity in the OFDM channel usage. Applying the CS technology to the receiver directly reduces the power consumption in mixed signal circuit which is attributable to less number of analog-to-digital converters (ADCs). A new streamlined algorithm for digital signal processing (DSP) to recover the compressively sensed data is also devised. Besides the simplification of the signal sensing, the simulation results also show that the reception fidelity of the proposed architecture outperforms that of the conventional maximum likelihood (ML) MIMO detector when the channel is lightly loaded.