A Hardware-Efficient Massive MIMO Detector Using Improved Quasi-Newton Method

The quasi-newton methods are one of the most effective for solving unconstrained optimization problems. This letter provides a hardware-efficient massive multiple-input multiple-output (MIMO) detector using an improved (IQN) method. Due to similarity in stepsize calculation Barzilai-Borwein and limited-memory BFGS, two deeply fused proposed IQN algorithm higher convergence speed. corresponding efficient architecture is also given, which dual-track systolic array employed diminish number required Processing Elements (PE) by nearly half with less computational delay. Furthermore, approximate divider based on Goldschmidt method designed further reduce hardware overhead. Simulation results show that achieves better Bit-Error-Ratio (BER) performance under different antenna configurations, FPGA implementation validate superiority terms efficiency over state-of-the-art (SOTA) detectors.