ADAPTIVE RESOURCE ALLOCATION AND CAPACITY COMPARISON OF OFDMA AND MC-CDMA SCHEMES BASED ON IMPERFECT POWER-LINE CSI

The challenge of imperfect channel state information (CSI) has been generally ignored in the literature for resource allocation in power-line communication (PLC) systems. However, power-line channels have a time varying nature, frequency selective attenuation, and inevitable impulsive noise due to the random switching of power appliances. Thus, it is unrealistic to expect to have perfect CSI. In this paper, the potential maximum erg odic capacity of the orthogonal frequency division multiple access (OFDMA) scheme and that of a proposed improved multicarrier code division multiple access (MC-CDMA) scheme are investigated and compared for imperfect power-line CSI. The resource allocation with fairness constraint is formulated as a convex optimization framework, and optimal user selection and power allocation algorithms are proposed for both schemes. The simulation results and analyses show that higher capacity gain can be achieved by modelling the statistical characteristics of the imperfect CSI and solving the expectation of CSI correlation function. Furthermore, the capacity loss of the proposed improved MC-CDMA scheme when compared with that of OFDMA is significantly reduced by the relaxation of fairness constraint, and the loss can be ignored in some cases.