Automatic digital modulation recognition using artificial neural network and genetic algorithm

@article{citeulike:1267277, abstract = {Automatic recognition of digital modulation signals has seen increasing demand nowadays. The use of artificial neural networks for this purpose has been popular since the late 1990s. Here, we include a variety of modulation types for recognition, e.g. QAM16, V29, V32, QAM64 through the addition of a newly proposed statistical feature set. Two training algorithms for multi-layer perceptron (MLP) recogniser, namely Backpropagation with Momentum and Adaptive Learning Rate is investigated, while resilient backpropagation (RPROP) is proposed for this problem, are employed in this work. In particular, the RPROP algorithm is applied for the first time in this area. In conjunction with these algorithms, we use a separate data set as validation set during training cycle to improve generalisation. Genetic algorithm (GA) based feature selection is used to select the best feature subset from the combined statistical and spectral feature set. RPROP MLP recogniser achieves about 99\% recognition performance on most SNR values with only six features selected using GA.}, author = {Wong, M. L. D. and Nandi, A. K. }, booktitle = {Special Section on Independent Component Analysis and Beyond}, citeulike-article-id = {1267277}, journal = {Signal Processing}, keywords = {classification, communication, engineering, modulation}, month = {February}, number = {2}, pages = {351--365}, posted-at = {2007-04-30 06:16:47}, priority = {0}, title = {Automatic digital modulation recognition using artificial neural network and genetic algorithm}, url = {http://www.sciencedirect.com/science/article/B6V18-4B0NP1J-2/2/1f81452eb5755b41c25dab4fab482d7b}, volume = {84}, year = {2004} }

TY - JOUR ID - citeulike:1267277 L3 - citeulike-article-id:1267277 TI - Automatic digital modulation recognition using artificial neural network and genetic algorithm T2 - Special Section on Independent Component Analysis and Beyond JF - Signal Processing VL - 84 IS - 2 SP - 351 EP - 365 N2 - Automatic recognition of digital modulation signals has seen increasing demand nowadays. The use of artificial neural networks for this purpose has been popular since the late 1990s. Here, we include a variety of modulation types for recognition, e.g. QAM16, V29, V32, QAM64 through the addition of a newly proposed statistical feature set. Two training algorithms for multi-layer perceptron (MLP) recogniser, namely Backpropagation with Momentum and Adaptive Learning Rate is investigated, while resilient backpropagation (RPROP) is proposed for this problem, are employed in this work. In particular, the RPROP algorithm is applied for the first time in this area. In conjunction with these algorithms, we use a separate data set as validation set during training cycle to improve generalisation. Genetic algorithm (GA) based feature selection is used to select the best feature subset from the combined statistical and spectral feature set. RPROP MLP recogniser achieves about 99% recognition performance on most SNR values with only six features selected using GA. KW - classification KW - communication KW - engineering KW - modulation AU - Wong, MLD AU - Nandi, AK PY - 2004/02// UR - http://www.sciencedirect.com/science/article/B6V18-4B0NP1J-2/2/1f81452eb5755b41c25dab4fab482d7b ER -