Automatic classification of QAM signals by neural networks

Abstract

In this paper, automatic classification of QAM signals including 64-state QAM and 256state QAM is discussed Three layer neural networks whose input data is the histogram distribution of instantaneous amplitude at symbol points is used for the classification. The evaluations of classdication performance are canid out for both cases in which the synchronization of symbol timing is assured at the receiver and not assured Good classification results are obtained by the computer simulations at SNR2lodB. The influence of the number of symbol points which are used for the calculation of histogram is also dmussed

Clustering based distribution fitting algorithm for Automatic Modulation Recognition

“…[3] proposed an algorithm in identification of digital modulation types (QAM, PSK and FSK) using the wavelet transform. Meanwhile, the authors in [4,15] proposed to classify different modulation models within the same modulation type. [4] proposed the algorithm which is used to distinguish the models between BPSK and QPSK, while the authors in [15] applied the neural network algorithm on the classification of 16, 64 and 256-QAM.…”

“…Meanwhile, the authors in [4,15] proposed to classify different modulation models within the same modulation type. [4] proposed the algorithm which is used to distinguish the models between BPSK and QPSK, while the authors in [15] applied the neural network algorithm on the classification of 16, 64 and 256-QAM. In [2], the authors proposed an algorithm for multi-models classification between M-PSK, M-FSK, M-ASK.…”

“…As each modulation type has its own set of basis functions such as shape and number of constellation signals, collection of these basis functions can [4,6,7,10] in the past. In recent years, pattern recognition methods are dominant especially with the use of neural networks [8,15]. There are several different modulation recognition approaches as mentioned in [8] like quasi-log-likelihood to differentiate BPSK and QPSK, log-likelihood function to classify between Offset-QPSK (OQPSK) / BPSK / QPSK, and sequential probability ratio test in the context of hypothesis testing to classify among several QAM signals, etc.…”

Development and Comparative Study of Effects of Training Algorithms on Performance of Artificial Neural Network Based Analog and Digital Automatic Modulation Recognition

“…In addition, in Taira [13], the histogram distribution of instantaneous amplitude at symbol points was used for the automatic classification of digitally quadrature amplitude modulation (QAM) signals including 64-state QAM and 256-state QAM. The author obtained good classification results by computer simulations at SNR greater or equal to 10 dB.…”

“…The first family of AMR classifier developed are those that were designed to recognize only analog modulated signals [2,[5][6][7][8][9]. The classifier in the second family are those that are developed to recognise only digital modulated signals [10][11][12][13][14][15] while the third family of the AMR classifier are those that are designed to recognise or classify some joint analog and digital modulated signals [16][17][18]. These few selected algorithms: [2,[5][6][7][8][9], [10][11][12][13][14][15] and [16][17][18] for first, second and third families of AMR are briefly reviewed in Section 2.…”