A SVM-based processor for free-space optical communication

In this paper, we propose a novel block-wise detection technique using support vector machine (SVM) to overcome the scintillation noise in free-space optical (FSO) links. Considering that the turbulence fading is of millisecond level and the signalling rates can typically be hundreds of Mbps, the SVM method provides an extremely large data block which is proportional to the bit rates. At the bit rate of 100 Mbps, for example, a block length of 105 bits is reached. Moreover, the calculation of SVM scheme is independent of the channel's turbulence conditions and the received signal models, and requires no knowledge of prior information or heuristic assumptions. Numerical simulations show that the SVM-based detector can mitigate the fading-channel impairments effectively. With the application of the 4-fold cross validation and grid-search, the best SVM parameters can be acquired and adapted to the channel condition dynamically.

[1]  James A. Bucklew,et al.  Support vector machine techniques for nonlinear equalization , 2000, IEEE Trans. Signal Process..

[2]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[3]  Jason D. Schmidt,et al.  Numerical Simulation of Optical Wave Propagation With Examples in MATLAB , 2010 .

[4]  Chih-Jen Lin,et al.  A Practical Guide to Support Vector Classication , 2008 .

[5]  L. Andrews,et al.  Laser Beam Scintillation with Applications , 2001 .

[6]  Anthony Widjaja,et al.  Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond , 2003, IEEE Transactions on Neural Networks.

[7]  M.L.B. Riediger,et al.  Blind detection of on-off keying for free-space optical communications , 2008, 2008 Canadian Conference on Electrical and Computer Engineering.

[8]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[9]  George K. Karagiannidis,et al.  Generalized Maximum-Likelihood Sequence Detection for Photon-Counting Free Space Optical Systems , 2010, IEEE Transactions on Communications.

[10]  Joseph M. Kahn,et al.  Free-space optical communication through atmospheric turbulence channels , 2002, IEEE Trans. Commun..

[11]  Radford M. Neal Pattern Recognition and Machine Learning , 2007, Technometrics.

[12]  Wanyi Gu,et al.  Nonparameter Nonlinear Phase Noise Mitigation by Using M-ary Support Vector Machine for Coherent Optical Systems , 2013, IEEE Photonics Journal.

[13]  Joseph M. Kahn,et al.  Markov chain model in maximum-likelihood sequence detection for free-space optical communication through atmospheric turbulence channels , 2003, IEEE Trans. Commun..