Quantized Distributed Reception for MIMO Wireless Systems Using Spatial Multiplexing

We study a quantized distributed reception scenario in which a transmitter equipped with multiple antennas sends multiple streams via spatial multiplexing to a large number of geographically separated single antenna receive nodes. This approach is applicable to scenarios such as those enabled by the Internet of Things (IoT) which holds much commercial potential and could facilitate distributed multiple-input multiple-output (MIMO) communication in future systems. The receive nodes quantize their received signals and forward the quantized received signals to a receive fusion center. With global channel knowledge and forwarded quantized information from the receive nodes, the fusion center attempts to decode the transmitted symbols. We assume the transmit vector consists of arbitrary constellation points, and each receive node quantizes its received signal with one bit for each of the real and imaginary parts of the signal to minimize the transmission overhead between the receive nodes and the fusion center. Fusing this data is a nontrivial problem because the receive nodes cannot decode the transmitted symbols before quantization. We develop an optimal maximum likelihood (ML) receiver and a low-complexity zero-forcing (ZF)-type receiver at the fusion center. Despite its suboptimality, the ZF-type receiver is simple to implement and shows comparable performance with the ML receiver in the low signal-to-noise ratio (SNR) regime but experiences an error rate floor at high SNR. It is shown that this error floor can be overcome by increasing the number of receive nodes.