Linear successive interference cancellation in DS-CDMA systems

Successive interference cancellation techniques have the potential to reduce the interference and therefore increase the capacity of cellular radio systems. In this work, we analyze a linear DS-CDMA receiver utilizing soft feedback interference cancellation. To be able to meet diverse QoS requirements, we introduce a power control scheme that takes into account heterogeneous service levels as well as the impact of imperfect interference cancellation. We address the capacity resulting from this receiver and give its user capacity region on closed-form. We show that our results are general and provide a framework for capacity comparison for many proposed linear successive cancellation receiver models. Several receiver modifications are included, partial interference cancellation; when only part of the signal is canceled, and limited cancellation; when not every signal is canceled. We derive the minimum power solution of a single-cell, from which convergent multicellular distributed power control algorithms that iteratively update the powers are suggested. The numerical results show that there are large gains in user capacity from interference cancellation when the E-b/I-0 requirements are high. However, regardless of the required targets the maximum capacity is always achieved through partial successive interference cancellation. It is also found that the power levels are sensitive to the cancellation efficiency and worse performance than single-user detection can occur. Copyright (C) 2003 John Wiley Sons, Ltd.