pageHeader("Automatic Modulation Scheme Recognition");
Automatic Modulation Scheme Recognition
The advent of realisable software radio allows the implementation of creative transceiver designs, which can dynamically adapt to the communications channel and user applications. One such transceiver design that demonstrates how the flexibility of a software radio may be exploited is a software radio that automatically determines the modulation scheme used in an unknown signal.
An intercepted unknown signal may be either modulated
using an analogue modulation scheme e.g. AM,FM... or using a digital modulation scheme e.g. M-aryPSK,FSK...
Each particular modulation scheme has measurable characteristics such as the frequency domain profile and
underlying signal structure. For a software radio implementation employing real-time modulation scheme
recognition, the techniques must have a processing time overhead that still allows
the software radio to maintain its real-time objectives.
A real-time AM/FM modulation scheme recognition technique has been implemented on a software radio using general-purpose
processors. In addition, a technique designed for discriminating between BPSK,QPSK,8-PSK,QAM and 8-QAM using a
moment-based modified signal space representation metric has been developed.
A diagram illustrating the performance of the analogue modulation scheme recognition technique is shown below.
In this example, we consider the case where the intercepted signal is modulated using either AM or FM, and
the channel SNR is varied from 60dB to 2dB (AWGN channel) only.
Following on from this work, we examine cases where the signal is a Continuous Phase Modulated (CPM) signal and consider an Intersymbol Interference (ISI) channel. One specific case is to attempt to classify a GMSK signal using an ISI channel subjected to AWGN, random phase noise, Doppler and multipath propagation fading effects. Pulse spreading over adjacent transmitted symbols drastically affects phase based modulation scheme classifiers. The nature of the research outlined here is to examine the effects of adjacent pulse interference.
Consider two cases where the transmitted pulse affects 3 and 4 adjacent pulses (channel SNR = 100dB):
The Signal Space diagram on the left illustrates how pulse spreading over 3 adjacent transmitted pulses results in received signal point variations. For the case where the transmitted pulse 'smears' 4 adjacent pulses, the phase variations result in the example shown in the second diagram (on right).
Shown below is a graph of the High Order Statistical Moments vs Channel SNR. Three cases of GMSK are shown where the
transmitted pulse affects 2,3 and 4 adjacent symbols. The channel is subjected to AWGN, random phase noise and Doppler fading and the SNR is varied from 100dB
A point to note is that the StatisticalMoment measurements for the 3 adjacent pulse smearing is lower
in the high SNR range (>20dB range)
Onto OFDM! »
For further information about these schemes and performance analysis of the digital modulation recognition
techniques, please refer to the publications below.
Nolan,K.E., Doyle,L., Mackenzie,P. and O'Mahony,D., Modulation Scheme Classification for 4G Software Radio Wireless Networks, to appear in Proceedings of the IASTED International Conference on Signal Processing, Pattern Recognition, and Applications (SPPRA 2002), June 25-28, 2002, Crete, Greece.
Nolan,K.E., Doyle,L., O'Mahony,D. and Mackenzie, P., Signal Space based Adaptive Modulation for Software Radio, in Proceedings of the IEEE Wireless Communications and Networking Conference WCNC 2002, March 17-21 2002, Orlando, Florida.
Nolan, K.,Doyle, L., O'Mahony, D. & Mackenzie, P., Modulation Scheme Recognition Techniques for Software Radio on a General Purpose Processor Platform, in Proceedings of the First Joint IEI/IEE Symposium on Telecommunications Systems Research, Dublin, November 27th, 2001
Send a question or comment to Keith Nolan