Adaptive resource management for cognitive power line communication system

The authors investigate the adaptive resource allocation problem for cognitive power line communication system. The dynamic optimisation model is employed to maximise the energy-aware network utility, which is defined as the difference of the weighted network throughput and the power consumption. Then, the authors decompose the stochastic optimisation into a traffic management subproblem and a joint subchannel and power allocation subproblem. Since the joint subchannel and power allocation subproblem is a mixed integer non-linear optimisation, a continuity relaxation and Lagrange dual method is proposed to obtain the optimal solution of the mixed integer non-linear optimisation subproblem. After that, the authors develop an adaptive power line communication resource allocation algorithm to cope with the power line communication network dynamics only according to the current network state information. Finally, we derive a trade-off mechanism between network utility and network delay, in which the network delay is proportional to the control parameter V and the network utility is proportional to the parameter 1/V$1/V$. Simulation results not only show the effectiveness but also verify the theoretical analysis of our proposed algorithm.