Deep Transfer Learning-Assisted Signal Detection for Ambient Backscatter Communications

@article{Liu2020DeepTL,
  title={Deep Transfer Learning-Assisted Signal Detection for Ambient Backscatter Communications},
  author={Chang Liu and Xuemeng Liu and Zhiqiang Wei and Derrick Wing Kwan Ng and Jinhong Yuan and Ying-Chang Liang},
  journal={GLOBECOM 2020 - 2020 IEEE Global Communications Conference},
  year={2020},
  pages={1-6}
}
Existing tag signal detection algorithms inevitably suffer from a high bit error rate (BER) due to the difficulties in estimating the channel state information (CSI). To eliminate the requirement of channel estimation and to improve the system performance, in this paper, we adopt a deep transfer learning (DTL) approach to implicitly extract the features of communication channel and directly recover tag symbols. Inspired by the powerful capability of convolutional neural networks (CNN) in… Expand

Figures and Tables from this paper

Deep Transfer Learning for Signal Detection in Ambient Backscatter Communications
TLDR
A deep transfer learning (DTL) approach to implicitly extract the features of channel and directly recover tag symbols and an asymptotic explicit expression is derived to characterize the properties of the proposed CNN-based method when the number of samples is sufficiently large. Expand
Deep Residual Learning-Assisted Channel Estimation in Ambient Backscatter Communications
TLDR
Channel estimation in AmBC is modeled as a denoising problem and a convolutional neural network-based deep residual learning denoiser (CRLD) is developed to directly recover the channel coefficients from the received noisy pilot signals to demonstrate the performance of the proposed method approaches the optimal minimum mean square error (MMSE) estimator. Expand
Energy Efficient MIMO-OFDM Spectrum Sensing Using Deep Stacked Spiking Delayed Feedback Reservoir Computing
  • K. Hamedani, Lingjia Liu, Y. Yi
  • Computer Science
  • IEEE Transactions on Green Communications and Networking
  • 2021
TLDR
A spiking reservoir computing (RC) based approach is introduced for spectrum sensing of MIMO-OFDM systems to take advantage of the spatial and temporal correlations of the environment and a spiking delayed feedback reservoir (DFR) is developed. Expand

References

SHOWING 1-10 OF 21 REFERENCES
Deep Transfer Learning for Signal Detection in Ambient Backscatter Communications
TLDR
A deep transfer learning (DTL) approach to implicitly extract the features of channel and directly recover tag symbols and an asymptotic explicit expression is derived to characterize the properties of the proposed CNN-based method when the number of samples is sufficiently large. Expand
Deep Residual Learning-Assisted Channel Estimation in Ambient Backscatter Communications
TLDR
Channel estimation in AmBC is modeled as a denoising problem and a convolutional neural network-based deep residual learning denoiser (CRLD) is developed to directly recover the channel coefficients from the received noisy pilot signals to demonstrate the performance of the proposed method approaches the optimal minimum mean square error (MMSE) estimator. Expand
Deep CNN for Spectrum Sensing in Cognitive Radio
TLDR
This paper uses a deep neural network to automatically learn features from data itself, and develops a data-driven detection approach, which outperforms the conventional maximum eigenvalue detection method and derives the cost function for offline training in the spectrum sensing model, which guarantees the optimality of the designed test statistic. Expand
Deep Residual Learning for Channel Estimation in Intelligent Reflecting Surface-Assisted Multi-User Communications
TLDR
This paper develops a versatile DReL-based channel estimation framework, a convolutional neural network (CNN)-based DRN (CDRN) is proposed for channel estimation in IRS-MUC systems, and an explicit expression of the proposed CDRN is derived and analyzed in terms of Bayesian estimation to characterize its properties theoretically. Expand
Deep CM-CNN for Spectrum Sensing in Cognitive Radio
TLDR
A deep neural network is used to intelligently explore the data-driven test statistic in spectrum sensing, where a DNN-based likelihood ratio test (DNN-LRT) is derived to guarantee the optimality of the designed test statistic. Expand
Machine Learning Based Signal Detection for Ambient Backscatter Communications
TLDR
A machine learning based detection method is proposed to detect the tag signals for an AmBC system by transforming the detection problem into a classification problem, which outperforms the traditional detection methods, especially in the low SNR regime. Expand
Noncoherent Detections for Ambient Backscatter System
TLDR
A practical transmission model for an ambient backscatter system, where a tag wishes to send some low-rate messages to a reader with the help of an ambient RF signal source, and then provide fundamental studies of noncoherent symbol detection when all channel state information of the system is unknown is formulated. Expand
Semi-Coherent Detection and Performance Analysis for Ambient Backscatter System
TLDR
This work designs the semi-coherent detection of radio frequency signals transmitted from an ambient source as both energy supply and information carrier to enable communications between low-power devices and proposes an effective approach to estimate detection-required parameters rather than channels themselves. Expand
Ambient Backscatter Communication Systems: Detection and Performance Analysis
TLDR
A new transmission model is formulated, the data detection algorithm is designed, and two closed-form detection thresholds are derived to approximately achieve the minimum sum bit error rate (BER). Expand
Signal detection and BER analysis for RF-powered devices utilizing ambient backscatter
TLDR
This paper builds up a theoretical model for a communication system which consists of one reader and one tag and designs an optimal detector which can minimize the BER and find the closed-form expression for the detection threshold. Expand
...
1
2
3
...