A Low-Cost MMSE-SIC Detector for the MIMO System: Algorithm and Hardware Implementation

Abstract

We consider the minimum mean-squared error with successive interference cancellation (MMSE-SIC) detection of a frame of data in the spatially multiplexed multiple-input–multiple-output (MIMO) system. A complete MMSE-SIC detector needs to compute at both the preprocessing and SIC-detection stages. Since the SIC detection, which can be regarded as a backward substitution, is inevitable, we develop a computationally efficient preprocessing algorithm that relies on the backward substitution. We then propose a low-cost hardware architecture with a commonly shared backward-substitution module to work at both the preprocessing and detection stages. The very-large-scale-integration implementation results of our architecture for the four-by-four MIMO system using the 0.18- <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$\mu\hbox{m}$</tex></formula> complementary metal–oxide–semiconductor technology reveal that our architecture requires the fewest 79-K gates, provides the high throughput rate of 416 Mb/s, and works with the smallest preprocessing latency of 64 clock cycles. Our MMSE-SIC detector is a cheap solution for MIMO detection.