
Cognitive two‐way relaying using orthogonal channelisation: spectrum partition, subchannel permutation, and power allocation
Author(s) -
Li Yong,
Wang Tingting,
Lu Na
Publication year - 2016
Publication title -
iet communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.355
H-Index - 62
eISSN - 1751-8636
pISSN - 1751-8628
DOI - 10.1049/iet-com.2015.0364
Subject(s) - cognitive radio , relay , computer science , computer network , mathematical optimization , transmitter power output , partition (number theory) , channel (broadcasting) , throughput , transmission (telecommunications) , power (physics) , mathematics , telecommunications , wireless , physics , transmitter , quantum mechanics , combinatorics
In cognitive radio networks, primary users can share some licensed spectrums with secondary users in exchange for cooperation. In this study, an overlay scenario is investigated where a secondary node provides two‐way relaying service to a pair of primary users and thus is allowed to transmit its own messages by using some licensed spectrums. Orthogonal channelisation is assumed for the two‐way relay channel to form a number of parallel subchannels. Hence, the licensed spectrums can be partitioned between primary and secondary signals transmitted by the relay in the broadcast phase. Moreover, subchannels in the two phases of two‐way relaying can be permutated to further improve performance. The authors seek to maximise achievable data rate of secondary transmission subject to data rate requirements of the two primary users under individual power constraints. The joint spectrum partition, subchannel permutation, and power allocation problem is formulated into a mixed combinatorial programming problem, which is then transformed into a solvable form and finally yield an asymptotically optimal solution through dual decomposition method. They further prove that the solution to the relaxed problem is exactly the same as that to the original problem. Numerical results demonstrate performance gain of the proposed algorithm.