Title | ||
---|---|---|
Low Complexity Receiver Design Using Deep Neural Network Based on Compact Sparse AutoEncoder |
Abstract | ||
---|---|---|
In this letter, we demonstrate a novel strategy for designing a low complexity deep neural network (DNN) receiver. The compact-stacked Autoencoder (CSAE) receiver is designed based on the proposed neuron and layer numbers selection (NLNS) methodology. Compared with other DNN-based receiver, the CSAE receiver has low complexity but can achieve superior performance. Simulation results show that CSAE receiver can achieve or even outperform state of the art accuracy, and furthermore, the proposed receiver provides a robust performance against the phase offset, carrier frequency offset (CFO), and imperfect channel state information (ICSI). |
Year | DOI | Venue |
---|---|---|
2020 | 10.1109/LCOMM.2020.3011560 | IEEE Communications Letters |
Keywords | DocType | Volume |
Deep neural network (DNN),compact-stacked autoencoder (CSAE) receiver,MIMO,carrier frequency offset,imperfect channel state information (ICSI) | Journal | 24 |
Issue | ISSN | Citations |
12 | 1089-7798 | 0 |
PageRank | References | Authors |
0.34 | 0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Chongzheng Hao | 1 | 0 | 0.34 |
Xiaoyu Dang | 2 | 38 | 10.69 |
Maqsood Hussain Shah | 3 | 0 | 0.34 |
Meng Wang | 4 | 37 | 14.43 |
Xiangbin Yu | 5 | 170 | 39.05 |