• Journal of Communications and Networks
• /
• v.18 no.4
• /
• pp.589-599
• /
• 2016

Most existing resource management problem models arise from the original desire of allocating resources in either a user-centric or network-centric manner. The difference between their objectives is obvious: user-centric methods attempt to optimize the utility of individual users, whereas network-centric models intend to optimize the collective utilities of the entire network. In this paper, from the above two aspects, we analyze the robust power control problem in device-to-device (D2D) communication underlaying cellular networks, where two types of channel uncertainty set (e.g., ellipsoidal and column-wise) are considered. In the user-centric method, we formulate the problem into the form of a Stackelberg game, where the energy efficiency (EE) of each user is the ingredient of utility function. In order to protect the cellular user equipment's (CUE) uplink transmission, we introduce a price based cost function into the objectives of D2D user equipment (DUE). The existence and uniqueness of the game with the influence of channel uncertainty and price are discussed. In the network-centric method, we aim to maximize the collective EE of CUEs and DUEs. We show that by the appropriate mathematical transformation, the network-centric D2D power control problem has the identical local solution to that of a special case of the user-centric problem, where price plays a key role. Numerical results show the performance of the robust power control algorithms in the user-centric and network-centric models.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.10
• /
• pp.2457-2464
• /
• 2015

Distributed Networked Control Systems (NCSs) through wireless networks have a tremendous potential to improve the efficiency of various control systems. In this paper, we define the State Update Interval (SUI) as the elapsed time between successful state vector reports derived from the NCSs. A simple expression of the SUI is derived to characterize the key interactions between the control and communication layers. This performance measure is used to formulate a novel optimization problem where the objective function is the probability to meet the SUI constraint and the decision parameter is the channel access probability. We prove the existence and uniqueness of the optimal channel access probability of the optimization problem. Furthermore, the optimal channel access probability for NCSs is lower than the channel access probability to maximize the throughput. Numerical results indicate that the improvement of the success probability to meet the SUI constraint using the optimal channel access probability increases as the number of nodes increases with respect to that using the channel access probability to maximize the throughput.