Power and Bandwidth Allocation in Cooperative Dirty Paper Coding
作者单位：Dept.of Electrical EngineeringStanford UniversityStanfordCA 94305 USA Dept.of Electrical and Computer EngineeringUniversity of MinnesotaMinneapolisMN 55455 USA Dept.of Electrical EngineeringUniversity of Southern CaliforniaLos AngelesCA 90089 USA
会议名称：《2008 IEEE International Conference on Communications(ICC 2008)》
会议日期：2008年
学科分类：07[理学] 08[工学] 070104[理学-应用数学] 081101[工学-控制理论与控制工程] 0701[理学-数学] 0811[工学-控制科学与工程]
基 金：supported by the US Army under MURI award W911NF-05-1-0246 the ONR under award N00014-05-1-0168 DARPA’s ITMANET program under grant 1105741-1-TFIND a grant from Intel the NSF ITR under grant CCF-0313392
关 键 词：Channel allocation Transmitters USA Councils Fading Upper bound Ad hoc networks Channel state information System performance Bandwidth Communications Society
摘 要：The cooperative dirty paper coding （DPC） rate region is investigated in a two-transmitter two-receiver network with full channel state information available at all terminals. The transmitters cooperate by first exchanging messages over an orthogonal cooperation channel,then they mimic a broadcast channel （BC） and jointly perform DPC to send to the two independent receivers. The allocation of network power and bandwidth between the data and the cooperation channel is studied to characterize the cooperative DPC rate region. First,the optimal sum power allocation for a multiple access channel （MAC） is presented. Then through an application of the MACBC capacity duality,the cooperative DPC rate region is evaluated under different bandwidth allocation assumptions. Cooperative DPC outperforms non-cooperative time-division （TD） only when the cooperation channel is strong,since the joint-encoding capacity gain is negated by the overhead of message exchanges in a weak cooperation channel. Moreover,the cooperative capacity advantage over TD is more pronounced at the maximum sum rate point than when the rate vector is skewed toward one of the users.