| Code division multiple access (CDMA) technology is a strong candidate for the evolving wireless communications systems. In CDMA systems the users share the same frequency band, thus a good care must be taken to limit mutual interference. In principle, orthogonal codes would be the simplest solution. However, orthogonality between the users is destroyed by multipath propagation.That leads to the correlation between the user symbols. This is why the multiple-access interference (MAI) exists. Multiuser detection (MUD) is a technique which tries to exploit the structure of interference to be able to suppress it. Optimal MUD, however, is computationally exhausting, and requires several system parameters to be known. As a consequence, many suboptimal multiuser receivers and adaptive multiple access interference (MAI) suppression techniques have been studied extensively during the past ten years. These techniques only utilize second order statistics. However, the assumption of independence is also realistic, which can be utilized in a statistical technique called Independent Component Analysis (ICA).Recently, ICA has attracted a lot of interest both in communication and statistical signal processing communities. It is a statistical model where the observed data is expressed as a linear transformation of latent variables that are non-Gaussian and mutually independent. It does not need any prior knowledge except that the latent variables are statistically independent. Several good algorithms utilizing higher-order statistics either directly or indirectly via suitable nonlinearities are now available for solving the basic linear ICA problem. There exist many motivating reasons to use the means of ICA in the reception of a CDMA system:(1) ICA provides a near-far resistant receiver, being able to resist strong interferences. Resistance is achieved by ICA quite naturally, since ICA only requires the source signals to be statistically independent, but their strengths are allowed to differ.(2) With ICA we should expect some robustness against erroneous parameter estimation. The propagation delay and the state of the channel should be estimated prior to actual symbol estimation. For subspace-type receivers also the estimationfor the model order should be available. The estimation of these parameters will always include some measurement errors, which degrade the accuracy of symbol estimation. ICA, on the other hand, does not need that precise knowledge of the system parameters, since the estimation is based purely on the high order statistical properties of the signal.(3) An ICA block can be used as an add-on feature to be attached to any existing receiver structure. This makes it possible to consider hybrid receiver structures, in which the ICA block could be intelligently activated only when it is expected to improve performance.In this paper, we first focus on the direct-sequence CDMA system, namely the DS-CDMA communication system. After studying the data model of the receiving signal of the system, we find that it satisfy the condition of ICA framework. Therefore we propose a method that uses a Bias Removal Independent Component Analysis (BR-ICA) algorithm to directly exploit the user information. This method can remove the bias of coefficients for ICA when the measured signals are contaminated by large additive noise in the communication systems. Simulation results show that this method provides good interference suppression performance and low BER.With a WCDMA communication application in mind, it is not so suitable to apply ICA on its own, since the known spreading codes can help identify the source signals. Rather, it would be meaningful to consider ICA as an additional element, attached to some exiting receiver structure. The existing receiver would perform the task of user identification, after which ICA would offer additional interference suppression capability (since also the independence of the source signals is utilized) and mitigate the performance drop due to erroneous timing and channel estimation.That is just another aim we try to achieve. In this paper, the output of MMSE detector is used to initialize the ICA iterations. And then we use ICA algorithm to iterative until the request is satisfied. The method takes the advantage of not only the character of source statistical independence but also the superiority of ICA by itself. The simulation results show that the performance of the improved detector is significantly enhanced compared with the traditional one, which could compensate for any performance loss caused by erroneous propagation delay or channel estimation and then make the bit-error-rate degrade.The simulation results demonstrate that using ICA singly and combining it with traditional multiuser detectors can both exploit the independence of the users'symbol. The proposed algorithm can achieve considerable performance gain and suppress the MAI effectively. |
