Spatial multiplexing multiple-input-multiple-output (MIMO) communication systems have recently drawn significant attention as a means to achieve tremendous gains in wireless system capacity and link reliability. The optimal hard decision detection for MIMO wireless systems is the maximum likelihood (ML) detector. ML detection is attractive due to its superior performance (in terms of BER). However, direct implementation grows exponentially with the number of antennas and the modulation scheme,

more »

... aking its ASIC or FPGA implementation infeasible for all but low-density modulation schemes using a small number of antennas. Sphere decoding (SD) solves the ML detection problem in a computationally efficient manner. However, even with this complexity reduction, real-time implementation on a DSP processor is generally not feasible and high-performance parallel computing platforms such as FPGAs are increasingly being employed for this class of applications. The sphere detection problem affords many opportunities for algorithm and micro-architecture optimizations and tradeoffs. This paper provides an overview of techniques to simplify and minimize FPGA resource utilization of sphere detectors for highperformance low-latency systems.