Deep Neural Network-Based Quantized Signal Reconstruction for DOA Estimation

3 May 2020  ·  Weifeng Han, Peng Chen, Zhenxin Cao ·

For a massive multiple-input-multiple-output (MIMO) system using intelligent reflecting surface (IRS) equipped with radio frequency (RF) chains, the multi-channel RF chains are expensive compared to passive IRS, especially, when the high-resolution and high-speed analog to digital converters (ADC) are used in each RF channel. In this letter, a direction of angle (DOA) estimation problem is investigated with low-cost ADC in IRS, and we propose a deep neural network (DNN) as a recovery method for the low-resolution sampled signal. Different from the existing denoising convolutional neural network (DnCNN) for Gaussian noise, the proposed DNN with fully connected (FC) layers estimates the quantization noise caused by the ADC. Then, the denoised signal is subjected to the DOA estimation, and the recovery performance for the quantized signal is evaluated by DOA estimation. Simulation results show that under the same training conditions, the better reconstruction performance is achieved by the proposed network than state-of-the-art methods. The performance of the DOA estimation using 1-bit ADC is improved to exceed that using 2-bit ADC.

PDF Abstract
No code implementations yet. Submit your code now

Datasets


  Add Datasets introduced or used in this paper

Results from the Paper


  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

Methods


No methods listed for this paper. Add relevant methods here