Open AccessPerformance Analysis of Automatic Modulation Classification using Time Frequency Transforms under Non-Ideal Channel Conditions
Author(s) -
M.Venkata Subbarao*,
P. Samundiswary
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.l3171.1081219
Subject(s) - additive white gaussian noise , modulation (music) , computer science , channel (broadcasting) , fading , algorithm , metric (unit) , ideal (ethics) , electronic engineering , time–frequency analysis , frequency modulation , pattern recognition (psychology) , mathematics , speech recognition , artificial intelligence , telecommunications , acoustics , engineering , radio frequency , physics , radar , operations management , philosophy , epistemology
Classification of different analog and digital modulation classes using Time-Frequency Transforms (TFTs) through MST and MFSWT under ideal channel conditions is presented in this paper. It also deals with performance analysis of proposed Modified S- Transform (MST) and Modified Frequency Slice Wavelet Transform (MFSWT) based Automatic Modulation Classification (AMC) methods under different channel conditions such as Gaussian and fading channels. The performance of the proposed TFT based AMC methods under AWGN (with SNR values varied from -10 dB to 20 dB) and fading channels is examined through simulation. Moreover, the performance of the proposed TFT based AMC is compared with that of the existing techniques in terms of performance metric namely classification accuracy which is also discussed in this paper.