Font Size: a A A

Algorithms For Modulation Recognition Of Digital Communication Signals

Posted on:2002-06-10Degree:DoctorType:Dissertation
Country:ChinaCandidate:W D ChenFull Text:PDF
GTID:1118360062975200Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
Communication signals travel in space with different modulation formats. In many applications, it is required to monitor the activities of these signals, identify their characteristics, even to intercept the signal information content. For instance, Civilian authorities may wish to monitor their transmissions in order to implement signal confirmation, interference identification and spectrum management. Especially in military applications, communication intelligence system, as one of the Electronic Support Measures (ESM) in Electronic Warfare (EW), is used to monitor electromagnetic spectra activities, implement threat detection and warning, and help to select jamming strategy and to intercept useful military intelligence.Modulation format is one of the most important characteristics used to distinguish communication signals. Knowing the modulation format and modulation parameters of a communication signal is the first step of correct demodulation. Given a received communication signal, the objective of modulation recognition is to decide the modulation format and estimate the modulation parameters of the communication signal without any priori knowledge about the signal information content.Based on the analysis of invariant features in cumulant domain of communication signals, the classification of communication signals with MPSK, MASK and MQAM modulation formats is investigated this dissertation. The main works can be summarized as follows:1. A new sub-optimum likelihood ratio classification algorithm is proposed at the assumption that the communication signals be received in "Higher Signal to Noise Ratio (HSNR)" environment with engineering practice significance. The new algorithm integrates the benefit of the average likelihood ratio algorithm with higher correct classification probability and that of the standard general likelihood ratio algorithm with low computation complexity, meanwhile, overcomes the invalidation problem of standard general likelihood ratio algorithm, when there are including relations between constellation sets of communication signals to be recognized. Through theoretically analyzing the HSNR condition for our algorithm to be applied, we relate the correct classification probability and bit error ratio of demodulation. To the best of our knowledge, this is the first time that modulation classification algorithm considers quantificationally the relations between signal environment of modulation classification and that of signal demodulation. The theoretical analysis and computer simulations reveal that, with HSNR conditionssatisfied and enough data received, the correct classification probability of our sub-optimum algorithm can achieve about 100%.2. A new cumulant based algorithm of estimating Signal to Noise Ratio (SNR) and reference phase is proposed for likelihood ratio classification of MPSK signals, which is blind to unknown phase order of MPSK signals. The asymptotic statistic distribution characteristics of cumulant function estimation of MPSK signals are given. The performance of our parameter estimation algorithm applied to MASK and MQAM signals with known level order are preliminarily investigated through computer simulation.3. The new algorithms for classification of MPSK, MASK and MQAM signals using cumulant invariants are proposed in Gaussian noise and ideal communication channel environment. The new classification features are blind to unknown SNR and reference phase. The asymptotic performances of our algorithms are verified through theoretical analysis and/or computer simulations. According to the respective properties of different signal sub-set, we give recursive order-reducing classification algorithm. So that, beyond certain SNR and with enough received data, theoretically, our classification algorithms can recognize digital communication signals with any modulation order.4. Extending cumulant invariants based signal classification algorithms to Gaussian multipath channel environment, we proposed recognition algorithm...
Keywords/Search Tags:Modulation recognition, Likelihood ratio classification, Pattern recognition, Invariant classification feature, Higher order cumulant, Cyclic cumulant
PDF Full Text Request
Related items