In a conventional cellular network underlaid with femtocells, the authors employ interference alignment (IA) to restrict cross‐layer co‐channel interference (X‐interference). In their scenario, locally obtained channel state information is allowed to be conveyed between heterogeneous nodes to construct a decentralised IA‐based scheme. Furthermore, they provide closed‐form formulas for the upper bounds of X‐interference in both uplink and downlink directions and characterise how they asymptotically approach zero for big enough channel dimensions used by individual nodes. Simulation results demonstrate that X‐interference is always bounded by the above‐mentioned limits and prove accuracy of their analytical formulas.

Cellular networks are of particular interest when communication takes place. The main barrier of employing cellular networks to reach broad access everywhere is interference. Although interference alignment has by theory raised hopes to leap this last hurdle, however, no operator will show interest for practical implementation of the complicated algorithms proposed so far, especially considering the need for global channel knowledge. Besides, there is no measure to appraise the performance of existing algorithms. In this study, the authors study blind cellular networks to characterise their achievable degrees of freedom. As it is very hard – if not impossible – to obtain its exact value, they provide related upper and lower bounds for this special scenario when power control mechanism is applied in the network. They also propose constructive proof for the lower bound and offer their seminal blind resource allocation algorithm in which they utilise concepts of both frequency alignment and fractional frequency reuse. Analytical upper bounds are also proposed and proved to be tight with respect to simulated results.

In this paper, novel routing tree construction algorithms for emerging multi-channel WiMax networks are presented. Both centralized and distributed schemes are explored. Simulation results show the improvement achieved compared to similar networks with a mesh topology.

Two iterative cross layer schemes are introduced for throughput maximization in multi channel WMN's where each link has a single designated channel. For both schemes, a two phase algorithm is developed where each phase is composed of routing and channel assignment steps. Results extracted from the channel assignment carried out in the first phase are fed back into phase 2 and appropriately utilized to improve aggregate network throughput.