Single-Channel Blind Source Separation of Spatial Aliasing Signal Based on Stacked-LSTM
Open Access
Abstract
Aiming at the problem of insufficient separation accuracy of aliased signals in space Internet satellite-ground communication scenarios, a stacked long short-term memory network (Stacked-LSTM) separation method based on deep learning is proposed. First, the coding feature representation of the mixed signal is extracted. Then, the long sequence input is divided into smaller blocks through the Stacked-LSTM network with the attention mechanism of the SE module, and the deep feature mask of the source signal is trained to obtain the Hadamard product of the mask of each source and the coding feature of the mixed signal, which is the encoding feature representation of the source signal. Finally, characteristics of the source signal is decoded by 1-D convolution to to obtain the original waveform. The negative scale-invariant source-to-noise ratio (SISNR) is used as the loss function of network training, that is, the evaluation index of single-channel blind source separation performance. The results show that in the single-channel separation of spatially aliased signals, the Stacked-LSTM method improves SISNR by 10.09∼38.17 dB compared with the two classic separation algorithms of ICA and NMF and the three deep learning separation methods of TasNet, Conv-TasNet and Wave-U-Net. The Stacked-LSTM method has better separation accuracy and noise robustness.Keywords
Funding Information
- Chinese Academy of Sciences Space Science Strategic Leading Science and Technology Project (XDA15060100)
- Chinese Academy of Sciences Strategic High-Tech Innovation Fund (GQRC-19-14)
This publication has 49 references indexed in Scilit:
- Multitalker Speech Separation With Utterance-Level Permutation Invariant Training of Deep Recurrent Neural NetworksIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2017
- Xception: Deep Learning with Depthwise Separable Convolutions2017
- Sparse component analysis‐based under‐determined blind source separation for bearing fault feature extraction in wind turbine gearboxIET Renewable Power Generation, 2017
- Convolutional Radio Modulation Recognition Networks2016
- Ultra-lightweight PANiNF/MWCNT-functionalized separators with synergistic suppression of polysulfide migration for Li–S batteries with pure sulfur cathodesJournal of Materials Chemistry A, 2015
- An Experimental Study on Speech Enhancement Based on Deep Neural NetworksIEEE Signal Processing Letters, 2013
- Under-Determined Reverberant Audio Source Separation Using a Full-Rank Spatial Covariance ModelIEEE Transactions on Audio, Speech, and Language Processing, 2010
- Source separation using single channel ICASignal Processing, 2007
- Underdetermined blind source separation using sparse representationsSignal Processing, 2001
- Learning to Forget: Continual Prediction with LSTMNeural Computation, 2000