A hierarchical approach to the classification of digital modulation types in multipath environments

Abstract

A hierarchical approach to the classification of digital modulation types in multipath environments Approved for public release; distribution is unlimited.The report was prepared by: Approved for public release; distribution is unlimited A ABSTRACT (Maximum 200 words)This study presents a hierarchical classification approach to the classification of digital modulation schemes of types [2,4,8]-PSK, [2,4,64,256]-QAM in low SNR levels and multi path propagation channel conditions. A hierarchical tree-based classification approach is selected as it leads to a relatively simple overall scheme with few parameters needed to differentiate between the various modulation types. Back-propagation neural network units are adopted at each tree node because they offer the flexibility needed to cope with varying propagation environments, as is the case in real-world communications. The selection of robust and well-defined higher-order statistics-based class features is considered and a small number of cumulants and moments chosen to differentiate between all various types of modulation types, except for specific M-QAM types. Simulations show that M-QAM types may be so affected by multipath and fading that higher-order statistic parameters become of very limited use. While being part of the hierarchical procedure, the identification of specific M-QAM types is conducted via equalization algorithms. Extensive simulations show overall classification performances to be strongly affected by the amount of multipath distortion and noise in the transmission channels. Results also show a much higher sensitivity of high-order M-QAM types to fading and multi path propagation distortions than other modulation types.

Survey of automatic modulation classification techniques: classical approaches and new trends

“…The design of a FB algorithm first needs some features for data representation and then decision making [100]. Examples of features are the correlation between the in-phase and quadrature signal components [27], the variance of the centered normalized signal amplitude, phase and frequency [28], the variance of the zero-crossing interval [32], [33], the variance of the magnitude of the signal wavelet transform (WT) after peak removal [36]- [38], the phase PDF [44]- [46] and its statistical moments [47]- [49], moments, cumulants, and cyclic cumulants of the signal itself [41]- [43], [53], [54], [58]- [66], etc. The entropy [67], [68], fuzzy logic [69], [70], a moment matrix technique [71], [72] and a constellation shape recovery method [73] were also used for AMC.…”

“…The decision was made based on the minimum absolute value of the difference between the sample estimate and prescribed values of the feature. Reference[43] combined several normalized moments and cumulants for training a NN, to identify FSK, PSK and QAM in multipath environments.…”

The Modulation Recognition Approaches for Software Radio

“…Dai and Wang employed the moments feature for the classification of various modulation types [12]. The same statistical was used by Hatzichristos and Fargues [13]. The researchers Swami and Sadler [14] treated the cumulant feature to classify the signals MPSK, MQAM, MASK, V29, V29c and V32.…”

Kuiper test based modulation level classification under unknown frequency selective channels

“…In this technique, the modulation level is decided based on the closest match between equalized symbols and constellation points of modulation levels. This technique, however, performs poorly at SNRs below 10dB [14].…”