Abstract
In this paper, we propose a noise enhanced neural network based detector. The proposed method can detect the known weak signal in additive non-Gaussian noise. Carefully injected noise in a neural network enhances the weak signal detection performance. During training, the back-propagation algorithm achieves less error and it converges faster with the addition of the external noise. The optimum value of external noise is calculated theoretically and justified by simulation. This method excels over the traditional neural network based detectors in terms of its performance characteristics i.e., the probability of detection (\(P_D\)) at some specified probability of false alarm (\(P_{FA}\)). Performance of the noise enhanced neural network based detector under several signal-to-noise ratio environments are also compared with state-of-the-art detectors.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.References
Lehmann EL (1986) Testing statistical hypothesis, 2nd edn. Wiley, New York
Poor HV (1988) An introduction to signal detection and estimation. Springer-Verlag, New York
Kassam SA (1988) Signal detection in non-Gaussian noise. Springer-Verlag, New York
Watterson JW (1990) An optimum multilayer perceptron neural receiver for signal detection. IEEE Trans Neural Netw 1(4):298–300
Lippmann RP, Beckman P (1989) Adaptive neural net preprocessing for signal detection in non-Gaussian noise. In: Advances in neural information processing systems, vol 1
Michalopoulou Z, Nolta L, Alexandrou D (1995) Performance evaluation of multilayer perceptrons in signal detection and classification. IEEE Trans Neural Netw 6(2):381
Gandhi PP, Ramamurti V (1997) Neural networks for signal detection in non-Gaussian noise. IEEE Trans Signal Process 45(11):2846
Halay N, Todros K, Hero AO (2017) Binary hypothesis testing via measure transformed quasi-likelihood ratio test. IEEE Trans. Signal Process 65:6381–6396
Bell M, Kochman Y (2017) on composite binary hypothesis testing with training data. In: Fifty-fifth annual Allerton conference, pp 1026–1033
Naderpour M, Ghobadzadeh A, Tadaion A, Gazor S (2015) Generalized wald test for binary composite hypothesis test. IEEE Signal Process Lett 22:2239–2243
Ma J, Xie J, Gan L (2018) Compressive detection of unknown-parameters signals without signal reconstruction. Signal Process 142:114–118
Audhkhasi K, Osoba O, Kosko B (2016) Noise-enhanced convolutional neural networks. Neural Netw (Elsevier) 78:15–23
Chen H, Varshney PK, Kay SM (2007) Theory of stochastic resonance effects in signal detection: part I fixed detectors. IEEE Trans Signal Process 55(7):3172–3184
Franzke B, Kosko B (2011) Using noise to speed up Morkov chain Monte Carlo estimation. Procedia Comput Sci 53:113–120
Audhkhasi K, Osoba O, Kosko B (2013) Noise benefits in backpropagation and deep bidirectional pre-training. IJCNN, pp 1–8
Kay SM (2008) Fundamentals of Statistical signal processing, detection theory, vol 2. Prentice Hall PTR, Upper Saddle River
Guo G, Mandal M, Jing Y (2012) A robust detector of known signal in non-Gaussian noise using threshold system. J Signal Process 92:2676–2688
Urkowitz H (1967) Energy detection of unknown deterministic signals. In: Proceedings of the IEEE, pp 523–531
Han J, Liu H, Sun Q, Huang N (2015) Reconstruction of pulse noisy images via stochastic resonance. Nat Sci Rep. Article number: 10616
Wiesenfield K, Moss F (1995) Stochastic resonance and the benefits of noise: from ice ages to crayfish and squids. Nature 373:33–36
Sun Q, Liu H, Huang N, Wang Z, Han J, Li S (2015) Non-linear restoration of pulse and high noisy images via stochastic resonance. Nat Sci Rep. Article number: 16183
Jha RK, Biswas PK, Chatterji BN (2012) Contrast enhancement of dark images using stochastic resonance. Image Process IET 6:230–237
Weissman I (2017) Sum of squares of uniform random variables. Stat Prob Lett 129:147–154
Shuhei I, Fabio D, Koh H (2018) Noise-modulated neural networks as an application of stochastic resonance. Neurocomputing 277:29–37
Acknowledgements
We thank Digital India Corporation under Ministry of Electronics & Information Technology, Government of India (Grant No. U72900MH2001NPL133410) for the financial support. We also express our deep gratitude to the anonymous reviewers for their highly professional recommendations, instructions, and motivation which contributed significantly to improve the quality of this paper.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Kumar, S., Kumar, A. & Jha, R.K. A Novel Noise-Enhanced Back-Propagation Technique for Weak Signal Detection in Neyman–Pearson Framework. Neural Process Lett 50, 2389–2406 (2019). https://doi.org/10.1007/s11063-019-10013-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11063-019-10013-z