A Deep-Learning-Based Distributed Compressive Sensing in UWB Soil Signals

Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 571)


Various studies that address the distributed compressed sensing (DCS) problem are based on jointly sparse prior (Sarvotham et al Asilomar conference on signals, systems, and computers, pp 1537–1541 (2006), [1]). In this paper, we relax this condition and propose a data-driven method to reconstruct UWB soil echo signals from compressed sensing (CS) random measurements. To this end, we use a long short-term memory (LSTM) network architecture which takes in DCS measurements as input and outputs reconstruction signals. The proposed method is LSTM-DCS. On a dataset of UWB soil echo signals, we show that the LSTM-DCS significantly outperforms traditional DCS solvers.


Distributed compressed sensing Long short-term memory Deep learning UWB signals 



This work was supported by the National Natural Science Foundation of China (61671138, 61731006), and was partly supported by the 111 Project No. B17008.


  1. 1.
    Sarvotham S, Baron D, Wakin M, Duarte MF, Baraniuk RG (2006) Distributed compressed sensing of jointly sparse signals. In: Asilomar conference on signals, systems, and computers, pp 1537–1541Google Scholar
  2. 2.
    Liu X, Yu X, Liang J (2016) Soil moisture retrieval via non-singleton fuzzy logic with UWB echoes. In: International conference in communications, signal processing, and systems. Springer, pp 219–228Google Scholar
  3. 3.
    Wang T, Liang J, Liu X (2018) Soil moisture retrieval algorithm based on tfa and cnn. IEEE Access 7:597–604CrossRefGoogle Scholar
  4. 4.
    Donoho DL et al (2006) Compressed sensing. IEEE Trans Inf Theory 52(4):1289–1306MathSciNetCrossRefGoogle Scholar
  5. 5.
    Mousavi A, Patel AB, Baraniuk RG (2015) A deep learning approach to structured signal recovery. In: 2015 53rd annual allerton conference on communication, control, and computing (Allerton). IEEE, pp 1336–1343Google Scholar
  6. 6.
    Yao H, Dai F, Zhang D, Ma Y, Zhang S, Zhang Y, Tian Q (2017) Dr2-net: deep residual reconstruction network for image compressive sensing. arXiv:1702.05743
  7. 7.
    Kulkarni K, Lohit S, Turaga P, Kerviche R, Ashok A (2016) Reconnet: non-iterative reconstruction of images from compressively sensed measurements. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 449–458Google Scholar
  8. 8.
    Metzler C, Mousavi A, Baraniuk R (2017) Learned d-amp: principled neural network based compressive image recovery. In: Advances in neural information processing systems, pp 1772–1783Google Scholar
  9. 9.
    Baron D, Duarte MF, Wakin MB, Sarvotham S, Baraniuk RG (2009) Distributed compressive sensing. arXiv:0901.3403
  10. 10.
    Zhang Z, Rao BD (2011) Sparse signal recovery with temporally correlated source vectors using sparse bayesian learning. IEEE J Sel Topics Signal Process 5(5):912–926CrossRefGoogle Scholar
  11. 11.
    Ji S, Xue Y, Carin L et al (2008) Bayesian compressive sensing. IEEE Trans Signal Process 56(6):2346Google Scholar
  12. 12.
    Palangi H, Ward R, Deng L (2016) Distributed compressive sensing: a deep learning approach. IEEE Trans Signal Process 64(17):4504–4518MathSciNetCrossRefGoogle Scholar
  13. 13.
    Liang J, Liu X, Liao K (2018) Soil moisture retrieval using uwb echoes via fuzzy logic and machine learning. IEEE Internet Things J 5(5):3344–3352CrossRefGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  1. 1.School of Information and Communication EngineeringUniversity of Electronic Science and Technology of ChinaChengduChina

Personalised recommendations