[DPl17] Energy-Efficiency Based Resource Allocation Framework for Cognitive Radio Networks with FBMC/OFDM
Revue Internationale avec comité de lecture :
Journal IEEE Transactions on Vehicular Technology,
vol. 66(6),
pp. 4997-5013,
2017
motcle:
Résumé:
In this paper, we study resource allocation for a
multi-carrier-based cognitive radio network. More specifically,
we investigate the problem of secondary users’ energy-efficiency
(EE) maximization problem under secondary total power and
primary interference constraints. Firstly, assuming cooperation
among the secondary base stations (BSs), a centralized approach
is considered to solve the EE optimization problem for the
CR network where the primary and secondary users are using
either orthogonal frequency-division multiplexing (OFDM) or
filter bank based multi-carrier (FBMC) modulations. We propose
an alternating-based approach to solve the joint powersubcarrier
allocation problem. More importantly, in the first
place, subcarriers are allocated using a heuristic method for
a given feasible power allocation. Then, we conservatively approximate
the non-convex power control problem and propose
a joint Successive Convex Approximation-Dinkelbach Algorithm
(SCADA) to efficiently obtain a solution to the non-convex power
control problem. The proposed algorithm is shown to converge to
a solution that coincides with the stationary point of the original
non-convex power control subproblem. Moreover, we propose
a dual decomposition-based decentralized version of SCADA.
Secondly, under the assumption of no cooperation among the
secondary BSs, we propose a fully distributed power control
algorithm from the perspective of game theory. The proposed
algorithm is shown to converge to a Nash-equilibrium (NE)
point. Moreover, we propose a sufficient condition that guarantees
uniqueness of the achieved NE. Extensive simulation analyses are
further provided to highlight the advantages and demonstrate the
efficiency of our proposed schemes.