On the Performance of Network-Coded Cooperative OFDMA Systems with Subcarrier Allocation

In a recently proposed network-coded cooperation scheme, denoted as NCC-OFDMA, the authors adopt the random bipartite graph theory to allocate subcarriers and achieve optimal frequency diversity, in a scenario where $M$ users transmit independent information to a common destination with the help of $R$ relays, in a frequency-selective channel with $L$ coherence blocks. However, the analysis of the aforementioned scheme assumes that the subcarrier allocation algorithm is capable of providing full frequency diversity to all the links in the network, which generally does not hold in practice since the allocation is optimal only from the perspective of a single destination. Therefore, in this letter, we reevaluate the performance of NCC-OFDMA in a more realistic scenario where only the messages addressed to the destination benefit from the full frequency diversity provided by the subcarrier allocation algorithm. Our analysis, which assumes Nakagami-m fading, shows that the diversity order of NCC-OFDMA is considerably smaller, being actually m(R+L) instead of mL(R+1).