Low-complexity iterative receiver for multi-user OFDM systems in time-varying MIMO channels

Abstract

High rates at high velocity is the ambitious goal of future wireless communications. increasing demand for communications at vehicular speeds in turn requires high processing at the base station, to estimate the fast changing channel and decode large amount of data. Real-time constraints or energy consumption limit the amount of affordable signal processing. Multiple antennas at each transceiver, multi-user detection, frequency multiplexing are many possibilities to improve performance that also enhance the complexity of a system.<br />Estimating fast time-varying channels and treating this large quantity of information in an optimum manner becomes unfeasible.<br />Methods that perform well for simple systems are not always applicable when it comes to high dimensional systems and can not necessarily be implemented in real time. Therefore, sub-optimal methods and algorithms permitting reduction of computational complexity while keeping sustainable quality of signal are widely exploited.<br />This thesis investigates an iterative receiver for an multi-carrier (MC) code division multiple access (CDMA) system based on orthogonal frequency division multiplexing (OFDM). The aim is to reduce computational complexity within this iterative receiver while avoiding performance degradation.<br />Linear detection using Krylov subspace methods as well as non-linear detection using sphere decoders are investigated to design various receivers. Their respective complexity and performance are discussed.<br />Due to a series of approximations, there will always be need for defining a trade-off between the complexity and the performance degradation that are acceptable. Determining the trade-off will obviously depend on the system considered, its parameters and the available hardware. This thesis aims at presenting a fair comparison to facilitate such a discussion.