Price-based resource allocation for revenue maximization with cooperative communication

In recent years, cooperative communication becomes a promising technology to improve the spatial diversity for the future mobile network. Under this communication paradigm, both relay assignment and power allocation will greatly impact the network performance. However, since each device may be selfish, a significant challenge is to make the joint resource allocation robust to node selfishness while pursuing the maximum system revenue in the commercial network. This paper mainly studies the price-based joint relay-power allocation problem which aims to maximize the revenue (JRP-RM) while satisfying the bandwidth requirement with cooperative communication. We first analyze the power allocation under the bandwidth and selfishness constraints for different communication modes. According to the analytical results, this paper presents a polynomial-time algorithm, namely MRPA, to solve the JRP-RM problem. We prove its correctness and show that it can achieve an approximation ratio of 1-ε\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$1 - \varepsilon$$\end{document}, where ε\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon$$\end{document} is an arbitrarily positive constant. Moreover, we study the JRP-RM problem under a special case where the cost and payment are both linear functions of the transmission power. The simulation results show that MRPA can increase the system revenue about 34.2 and 18.9 % compared with the OPRA Yang et al. (2011 IEEE International Conference on Communications (ICC). IEEE, pp 1–6, 2011) and JRPA Xu et al. (Wirel Netw pp 2209–2219, 2010) algorithms while satisfying the selfishness constraint in most situations.