Direct Sequence Spread Spectrum Communications: Applications to Multiple Access and Jamming Resistance.

Abstract

The performance of binary asynchronous direct sequence spread spectrum multiple access communication systems is analyzed in this thesis in terms of the average error probability with and without the presence of white Gaussian noise. Previous methods of evaluating the average error probability are reviewed. Then classical methods, such as the series expansion method and the Gauss quadrature rule, are used to evaluate the average error probability for a larger number of users than has previously been possible. In these methods only the moments of the multiple access noise are required, rather than the probability density function, which is too complicated to evaluate. The computational complexity and the accuracy of each method are examined. In general these methods approximate the error probability with finite sums. If more terms are included, the accuracy will be improved. The price to pay is the computational complexity and in one case the risk of numerical instability. Average error probability is also evaluated for the case that the chip waveform is an arbitrarily shaped, time-limited, finite energy waveform. This assumption affects the power spectral density of the spread signal and results in improved average error probability in certain cases. Finally, the performance of direct sequence spread spectrum modulation when corrupted by multiple continous wave jamming signals and white Gaussian noise is examined. The average error probability is also evaluated for the single tone continuous wave jamming case. It is found that the jamming signal affects performance in proportion to a parameter that depends on the codeword chosen for the direct sequence spread spectrum signal and on the difference between the frequency of the jamming signal and the spread spectrum carrier frequency.