Subcarrier-allocation in downlink multicarrier DS-CDMA systems

In LTE/LTE-A orthogonal frequency-division multiple-access (OFDMA) systems, the number of subcarriers is usually very high, upto 2048, and the number of simultaneously supported users may also be very high. These characteristics of OFDMA systems not only result in the severe peak-to-average power ratio (PAPR) problem, but also prevent schedulers from employing the optimum or near-optimum subcarrier-allocation algorithms, due to their relatively high-complexity when the number of subcarriers and the number of users are high. Multicarrier direct-sequence code-division multiple-access (MC DS-CDMA) is one of high-flexibility communication schemes. Owing to the DS spreading, the number of subcarriers in MC DS-CDMA can be significantly lower than that in OFDMA, and can be configurated according to the frequency-selectivity of wireless channels, so that each subcarrier experiences independent fading. Therefore, the relatively high-complexity optimum or near-optimum subcarrier-allocation algorithms may be employed by MC DS-CDMA for achieving the best possible performance. In this contribution, we study the subcarrier-allocation issue in MC DS-CDMA. Specifically, three representative subcarrier algorithms, namely the greedy algorithm, worst subcarrier avoiding (WSA) algorithm and the optimum Hungarian algorithm, are introduced to and studied associated with the MC DS-CDMA. Furthermore, according to the characteristics of the MC DS-CDMA, a subcarrier-allocation algorithm, named as the parallel Hungarian algorithm, is proposed. The achievable performance of these subcarrier-allocation algorithms are compared in terms of complexity and bit error rate (BER).