Direct Sequence Spread Spectrum Communication Signal Detection

Direct-Sequence Spread Spectrum (DS-SS) system uses a high speed Pseudo-Random (PN) sequence to modulate a modulated signal to extend the bandwidth of the signal as well as to lower down the power spectrum density. It is a advanced communication system which can get many excellent properties such as Multiple Access, Low Probability of Interception and Ami-Jamming, and so widely used.We need to detect this kind of non-cooperative weak communication signals in reconnaissance. Detection in reconnaissance is quite different from that in cooperative communication systems, in which all the parameters about the signal are unknown. This increases the difficulty of detection. Traditional detection techniques such as analyzing the spectrum of the observations and looking for abnormalities cannot effectively to detect the kind of weak signals. We have to find new approaches to the problem that are the main task of this dissertation.The early approach is to accumulate the energy in the band of interest called Energy Detection, Energy Detector or Radiometer. This detector has very good detection performance in stationary background, but has bad performance in the non-stationary background in which the DS-SS technique is often applied and it cannot get the parameters of the signal except the coarse band where the signal lies. Crosscorrelation Radiometer is the extent of the radiometer to overcome the drawbacks of the radiometer. Feature Detection or Feature Detector often uses nonlinear transform of the signal to produce the line spectrum at the specific value that has relation with the parameter of the signal and detects the line spectrum to determine the presence of the signal of interest. Usual detectors include rate-line detector that detects the line spectrum at the rate of the chips and carrier detector that detects the line spectrum at the harmonics of carrier frequency. The method of producing rate line is delay and multiply; the method of producing carrier harmonics is quadratic or four-order nonlinearity. Feature detectors can detect the presence of signal, at the same time gets the signal parameters and overcomes the fluctuation of power of the background noise and suppress the jamming. Communication signal should bemodeled as cyclostationary. The straightforward examples are the feature detectors such as rate line detector, carrier detector. The cyclic spectrum analyzer detects the signal directly on the plane of frequency-cyclic frequency by detecting the features of the signal. This approach can get the parameters of the signal and suppress the stationary noise completely in theory and suppress the narrow-band jamming if the jamming signal has different cyclic frequency from that of the signal of interest. It is a good detecting technique especially for search-type detection.The dissertation analyses the principals of the detectors mentioned above and evaluates the performance and proposes some improvement techniques to improve the performance. Computer simulations and performance comparisons are presented.In general, Energy Detector works best in the stationary environment, but is sensitive to noise fluctuation and jamming and cannot get signal parameters. Rate line detector and carrier detector works well in nonstationary environment and can get the parameters of the signal, so is very practical. Cyclic spectrum analyzer is the method approaching likely-hood detection. It can not only get the parameters of the signal but also suppress the noise and jamming and works best among the feature detectors.