Millimeter Wave Multi-Beam Combining Algorithm for Efficient 5G Cell Search

The New Radio (NR) 5G technology uses communication in millimeter wave (mmWave) frequency bands to provide significantly high throughput and low latency. Since mmWave frequencies suffer from higher path-loss, NR employs directional communication generated using beamforming techniques. Higher the directionality, better the coverage, but also larger is the number of beams required to cover all directions. In scenarios such as cell search during initial access or neighbor cell measurements, typically directional information is unavailable. Hence, user equipment (UE) needs to blindly search in all receive directions until a successful cell detection. Thus, in case of cell search, usage of directional communication results in a trade off between coverage performance and detection latency. In this paper, we propose an efficient NR UE cell search algorithm to mitigate this problem. To improve latency, we propose a beam scheduling algorithm using receive beam power measurements, which selects a subset of beam candidates and their scan order, and we further optimize detection performance by combining powers from multiple receive directions. Using a 3GPP simulation setup, we show that the proposed beam combining algorithm can yield significant gains, up to 2dB for detection performance, and beam scheduling algorithm can achieve a mean latency reduction of up to 50% compared to the conventional approach while reducing the overall processing overhead.