Improved Parallel ZF-VBLAST Detector for MIMO System

The rapid growth of data traffic in the last decade is expected to continue in the next generation (5G) system. In order to service the high demand of data rates in 5G, complex communication architectures are needed such as multiple input multiple output (MIMO) technology. The main idea of MIMO is to increase the number of antennas in both sides (transmitter Tx and receiver Rx) that can achieve high throughput and energy efficiency. Receivers complexity and speed are some of the important requirements in MIMO systems. We focus on Vertical Bell Laboratories Layered Space Time (VBLAST) receiver. Using large number of antennas with VBLAST technique can increase the complexity of the system and the required time to recover the original signal. However, parallel computing can be used to divide the tasks of complex algorithm and distribute them among different microprocessors to run in an acceptable execution time. In this paper, we perform deep performance and time analysis to decide the opportunities of parallelization. We achieved speed up of 11.4X in the computation time of ZF-VBLAST algorithm for different sizes of MIMO systems using multi-threaded implementation.