Water filling in cellar: the optimal power allocation policy with channel and buffer state information

In multiuser wireless communication systems, dynamic allocation of transmit power is an important means to deal with the time-varying nature both at physical layer and at network layer. Optimal power allocation with perfect channel and buffer state information is studied in this paper. We first build up the cross-layer model by integrating that of physical layer and network layer and then formulate the optimization problem on power allocation. We prove that the optimal solution to power allocation problem can be regarded as an extension of the traditional water-filling (TWF) technique, which is called "water-filling in cellar" (WFIC) policy. The corresponding dynamic programming algorithm is presented. Finally, numerical experiments are employed to illustrate the advantage of our proposed policy.