In this paper we use a channel model based on measurements from [l] to test the performance of a coded multicarrier CDMA system over the powerline channel from the secondary transformer coil to the subscriber. We focus on the different coding techniques which show that coding can make the difference under such a hostile medium. The comparison of coding, modulation types and the presencelabsence of interleaving is depicted by BER vs. SNR diagrams and complexity tables.2G.D papadopoulos is with the ATiStorle University Thessaloniki, scattered in the frequency domain. Finally a combining So the MC-CDMA system can be considered as a combination of an OFDM scheme and a CDMA one. The major advantage Of the MC-CDMA scheme in comparison to the OFDM scheme, is that since it can lower the symbol rateThe study 1 1 s been partially supported by a bilateral cooperation program of the General Secretariat for Research and TechnologyInrPere-vl D o n c l l u i a~ . _~_ " in each subcarrier, a longer symbol duration is created, so it 0-7803-7963-2/03/$17.00 02003 IEEE I53

Orthogonal Frequency Division Multiplexing (OFDM) and Multicarrier Code DivisionMultiple Access (MC-CDMA) systems are comparatively evaluated for Power Line Communications (PLC) in a frequency-selective fading environment with additive colored Gaussian noise which is used to model the actual in-home powerline channel. OFDM serves as a benchmark in order to measure the performance of various MC-CDMA systems, since multicarrier modulation systems are considered the best candidate for this kind of channel. Both single-user and multi-user cases are taken into account, making use of the appropriate combiner schemes to take full advantage of each case.System efficiency is enhanced by the application of different coding techniques, a fact which shows that powerful coding can make the difference under such a hostile

Recently, Multi-User Selection Diversity (MUSDiv) for single-carrier systems has been under extensive study on account of the enhancement it provides to system performance with minimum feedback requirements. However, its application to multichannel systems is considered straightforward and thus, it has not been thoroughly examined. In this paper, the performance of MUSDiv is investigated when applied to the spread-spectrum multi-carrier multiple-access system, where the scheduling has to be performed for all the available channels and self interference must also be considered. Specifically, based on the absolute and normalized Signal-to-Noise-Ratio (SNR) scheduling algorithms, two algorithms are presented, modified and optimized, so that they can be applied on a subband instead of a single-channel basis. Moreover, we propose a new scheduling scheme which constitutes a trade-off between the previous schemes, concerning fairness and capacity performance. The new algorithms are related to the symbol Signal-to-Noise-plus-Interference-Ratio (SNIR) instead of SNR and two interference models were devised to this end. Closed-form expressions for the system capacity are extracted for each case, which are compared with simulation results. The research is also extended to the case of non-identical user power profiles among the available subcarriers. The channel state information required to utilize multi-user selection diversity is already necessary for the most common combining schemes, so no more feedback is actually needed.