Research On Interference Type Recognition In Dsss Communication Systems

Direct sequence spread spectrum(DSSS)communication systems are widely used in both military and civil field for their appealing characteristics, such as low probability of intercept transmission, confidentiality, anti-multipath fading, easy networking, and so on. As is known to us all, the inherent processing gain of a DSSS communication system will, in many cases, provide the system with a sufficient degree of jamming rejection capability. However, when the interference is powerful enough and exceeds the interference tolerance, anti-jamming measures are needed. Currently, numerous interference rejection techniques have been proposed to suppress the severe jamming. But the problem is that all of the jamming rejection methods are proposed on the assumption that the interference type is known beforehand, and most of the techniques are aimed at a specific type of interference. So, we need to use different jamming suppression method for different interference type to improve the performance of a DSSS system. Therefore, aimed at the characteristic of interference suppression technology, the paper researches on the techniques of interference recognition in DSSS systems, and the main work are summarized as follows:First, the main interferences in DSSS communication systems are researched. In view of the status quo that the interference includes a wide range of styles, and is under the continuous development, we sum up eight major interferences by reading literature and simulation experiments, they are no interference, single-tone interference, narrowband BPSK interference, pulse interference, frequency-hopping interference, broadband comb spectrum interference, linear frequency modulation interference and broadband BPSK interference. For the interference above, the models of the interference signals are constructed, the corresponding mathematical expressions are given, and the characteristics respectively in time domain, frequency domain and time-frequency domain are briefly analyzed.Second, an in-depth study on feature extraction methods of the interference is carried out. On the basis of a detailed analysis of interference signals, and under the sense of from detection to identification, from coarse to fine, level- division design, eight parameters are extracted: Energy Limit Factor, Bandwidth of the Normalized Spectrum, Normalized Spectral Kurtosis, Normalized Spectral Flatness, Maximum-to-mean Ratio of the Signal in Time Domain, Normalized Deviation of Bandwidth of the Entire Signal and the Segmental Signals, Energy Aggregation in FRFT Domain, Difference of Energy Aggregation in FRFT Domain. An integration of interference detection and identification can be achieved by the help of the eight parameters. Then, two interference classifiers are designed. A five-level artificial neural network classifier is realized firstly, the design details and recognition performance are illuminated; Then a five-level decision tree support vector machine classifier is designed, which mends the lack of the ANN classifier; Finally, a comparison is carried out between different classifiers. Simulation results show that both interference classifiers provide a very high recognition rate under a general noise environment and can meet the anti-jamming requirement of a DSSS communication system.Finally, in order to experimentalize expediently, the visual simulation platform of interference classification and rejection in a DSSS system is redacted. Test results demonstrate that the intended function is realized, and the test results are satisfactory.