Hardware-Efficient Beamspace Direction-of-Arrival Estimator for Unequal-Sized Subarrays

Hybrid antenna array consisting of analog subarrays has been widely used for beamforming in millimeter-wave communication systems. Due to the hybrid structure, direction-of-arrival (DoA) estimation, which is essential to beam adjustment, has to be performed at the beamspace. In our previous work, unequal-sized subarrays has been proposed for beamspace DoA estimation because it can resolve the ambiguity problem, i.e., the DoA estimate at the beamspace may be non-unique. In this brief, we develop and implement a hardware-efficient beamspace DoA estimator dedicated for unequal-sized subarrays. Specifically, the proposed algorithm is tailored for hardware implementation such that the number of multiplications is reduced by 75% without significant performance loss. The proposed hardware architecture is layouted in 28-nm technology with a core area of 0.09 mm(2), a hardware complexity of 132 kilogate equivalents (kGE), and a clock frequency of 2.17 GHz.

Automated summary

This article introduces an efficient hardware beamspace DoA estimator for unequal-sized subarrays. By optimizing the algorithm and hardware design, the number of multiplication operations is reduced by 75% without significant performance loss. Experimental results show that the proposed hardware architecture achieves a small core area and high clock frequency in 28-nm technology.