Digital chaotic coding and modulation in CDMA

Conventional CDMA system uses PN sequences to spread user information. The number of shift registers used to generate them limit the lengths of these sequences. Good correlation properties can be achieved by Walsh codes and time-shifted version of PN sequences. The ability to predict future sequence is nevertheless possible, though difficult. Transmission therefore is not completely secured. Several other codes (Gold codes, Kasami etc) perform equally well as PN sequences, however it has been reported that these do not improve the security aspect of the system. Hence there is a need for a highly secure code that can be regenerated at the receiver and that can mimic PN sequences for spreading. In this paper, chaotic codes are investigated and their performance compared with the conventional PN sequences in an AWGN and Raleigh fading channel. The other type of spreading code which is the Walsh code uses its orthogonal property for user separation in the channel. This paper also investigates the behaviour of chaotic code in a multi-user situation. The auto and cross correlation properties of the spreading codes are also investigated. Chaotic codes are dependent on initial conditions and are continuous in nature. The undeterministic property of its amplitude from past sequences yields better security. Common modulation techniques (BPSK, QPSK) require higher SNRs to give similar performance in a multi-user environment. The use of QAM in order to improve BER whilst simultaneously inheriting secure property of chaotic codes is also investigated.