Search

A minimum-output-energy-based carrier tracking scheme is proposed for MC-CDMA systems. By utilizing an exponentially weighted empirical correlation whose weights are dictated by an adaptive forgetting factor, the proposed non-data-aided gradient-descent algorithm improves the tracking speed, while maintaining a small steady-state tracking error. Simulation results show that it takes only around 160 iterations to converge in response to an abrupt carrier frequency change of 10% of the intercarrier spacing.

The windows for achieving good multiple-access performance of multicarrier code-division multiple-access (MC-CDMA) systems employing minimum mean-squared error (MMSE) receivers are studied. The optimization objective, mean-squared error (MSE) minimization, is examined in the time domain. In a simplified two-user system, the optimization problem is shown to be equivalent to the problem of selecting the chip waveform in an asynchronous DS-CDMA system employing a matched filter. The rectangular window is shown to be optimum, while regarded as an inferior window in OFDM systems due to high spectral sidelobes and slow spectral rolloff.