Achieving High Energy Efficiency and Physical-Layer Security in AF Relaying

Authors:
Publisher website
Google Scholar
Abstract
For transmitting data in a secret and energy-efficient manner in collaborative amplify-and-forward relay networks, the secure energy efficiency (EE) defined as the secret bits transferred with unit energy is maximized to satisfy each node power constraint and target secrecy rate requirement, based on physical security framework. The secure EE is maximized by joint source and relay power allocation, which is a nonconvex optimization problem. To cope with this difficulty, a solution scheme and corresponding algorithms are developed by jointly applying fractional programming, exact penalty, alternate search, and difference of convex functions programming. The key idea of the scheme is to convert the primal problem into simple subproblems step by step, such that related methods are adopted. It is verified that, compared with secrecy rate maximization, the proposed scheme improves the secure EE significantly yet with a certain loss of the secrecy rate due to the tradeoff between secure EE and secrecy rate. Furthermore, the proposed scheme achieves higher secure EE and secrecy rate than total transmission power minimization does, while with a certain increase of power consumption. These results indicate that a reasonable balance among secure EE, secrecy rate, and power consumption can be reached by the proposed scheme.
Funding Information
  • National Basic Research Program of China (973 Program) (2012CB316000, 2013CB336600)
  • NSFC Young Investigator (61401249)
  • NSFC Excellent Young Investigator (61322111)
  • NSFC Innovative Research Groups (61321061)
  • MoE Specialized Research Fund for the Doctoral Program of Higher Education (SRFDP) (20130002120001)
  • MoE New Century Talent (NCET-12-0302)
  • National Innovative Talents Promotion (2013RA2151)
  • Beijing Nova (Z121101002512051)
  • U.S. NSF (CCF-1456921, CNS-1443917, ECCS-1405121, ECCS 1547201, NSFC 61428101)

This publication has 34 references indexed in Scilit: