Resource allocation algorithms in cognitive radio networks with outdated CSI

The presentation will summarize the first two outcomes of the JRA and highlight the open issues currently under investigation. The first part concerns a preliminary work about a link resource adaptation (LRA) problem for point-to-point bit-interleaved coded (BIC) OFDM systems with outdated channel state information (CSI) at the transmitter . The LRA strategy is able to select modulation, coding rate and power distribution of a OFDM system with the aim to maximize the expected goodput (EGP) metric. The main contributions consists in the derivation of a novel link performance prediction (LPP) method under outdated CSI. Based on these findings, an effective LRA strategy is then developed. Simulation results corroborate the effectiveness of the proposed method, showing remarkable performance gains w.r.t. the conventional approach which does not care of the outdated CSI. The main results of this study are reported in a paper to be presented at the forthcoming European Wireless 2014 conference. The second activity deals with a centralized resource allocation problem for multicarrier cooperative cognitive radio networks in the presence of imperfect CSI. The problem considers a cooperative network in a cognitive scenario, in which a secondary user (SU) may gain access to the same bandwidth of a primary user (PU) network, if the interference at the PU receivers is kept below a certain interference threshold. The cooperative network considered in this study includes the source node (SN), the destination node (DN) and M relay nodes (RNs). Direct link is not present between the SN and DN and decode-and-forward protocol is used at RNs. In this scenario, the SN transmits several OFDM symbols to the different RNs. Then each RN decodes the messages of a certain subset of subcarriers, then re-encodes these messages and forwards to the DN them, by using on different subcarriers w.r.t. the SN-RN link. The centralized resource allocation problem selects the subcarrier pairing for the SN-RN and the RN-DN links, and a power allocation for each RN with the aim of keeping the interference at each PU receiver below a certain threshold. Finally, the current research activities on the open issues will be briefly summarized. The goal is to derive a distributed resource allocation problem, by resorting to mathematical theory of convex optimization and game theory, for the scenario described above. The activity will also introduce, as performance metric, the goodput and the packet error rate, which both allow to characterize in more a suitable way the actual performance of a packet-oriented communication system. Moreover, the performance of the SN-RN-DN link will be investigated, in the presence of imperfect CSI.