Cooperative networks

Previously, the Wireless Communications Team have conducted an EPSRC funded project that investigated space-time coding and spatial multiplexing techniques to improve the throughput and capacity of broadband FWA systems. However, although useful gains in performance have been achieved, further improvements are likely to be more modest owing to fundamental aspects of the FWA channel. Specifically, FWA systems are in general characterized by LOS or near LOS propagation, yielding slow time variation, mild frequency selectivity and correlated propagation paths. Therefore it is difficult to exploit time, frequency and space diversity in an effective manner in order to improve system performance. To address this problem, the use of node cooperation diversity is being investigated in collaboration with Newcastle University in a project funded by the EPSRC.

In node cooperation diversity, nodes assist one another in their transmission to the destination node. Consequently, diversity in cooperative systems is associated with the existence of multiple paths to the destination node that fade independently. Therefore, node cooperation diversity is one of the key concepts to enable further improvements in the performance of FWA systems. A number of cooperation schemes have also been proposed in the literature. These include amplify-and-forward (AF), decode-and-forward (DF), adaptive AF and DF and coded cooperation.

Note that diversity in cooperative systems is associated with the existence of two or more paths to the access point (AP) that fade independently, i.e., the direct path and the relayed paths. However, a number of open problems remain concerning cooperative communication schemes, for example:

  • To investigate and apply information theoretic approaches to the determination of capacity bounds.
  • To investigate and implement centralised and distributed power control or power allocation schemes.
  • To investigate and implement node clustering techniques.
  • To design coded cooperation schemes.
  • To investigate and implement Multi-hop and hybrid networks.