Efficient Soft-Output Demodulation for mmWave Hybrid-MIMO Systems with Low-Resolution ADCs

In this paper, an efficient soft-output demodulation algorithm is proposed for mmWave hybrid-Multiple-Input Multiple-Output (MIMO) systems with low-resolution ADCs. The mmWave communications are suitable to satisfy the capacity requirement of the next-generation networks due to the utilization of wide bandwidth. In practical applications, a hybrid-MIMO system with low-resolution ADCs is considered to reduce the hardware complexity and power consumption. If existing demodulation methods are applied to such a system, demodulation performance is severely degraded due to high quantization noise and large correlation of channels. The proposed soft-output demodulation method introduces the linear approximated quantization model and quantized log-likelihood-ratio (LLR) calculation to reduce the effect of the quantization noise with low additional complexity. Additionally, the proposed algorithm adopts the lattice reduction-aided soft-output tree-searching which provides the enhanced performance. The proposed method has a tree-searching structure similar to the existing sub-6Ghz demodulators, so the additional implementation complexity is low and can support various bit resolutions. The BER performance results present that the proposed algorithm achieves an enhanced performance compared to existing soft-output methods with negligible additional complexity.