A lot of work has been focused on desig-ning and analyzing various cooperative diversity pro-tocols for wireless relay networks. To provide a uni-fied queuing analytic framework, we fonmlate an em-bedded Markov chain,...A lot of work has been focused on desig-ning and analyzing various cooperative diversity pro-tocols for wireless relay networks. To provide a uni-fied queuing analytic framework, we fonmlate an em-bedded Markov chain, which rams out to be a Quasi-Birth-and-Death （QBD） process. Using the Matrix-Ce-ometric method, we can analyze the average delay in a unified way. Theoretical analysis is validated by simu-lation results. We show that the delay performances of Amplify-and-Forward or Decode-and-Forwaxd （AF/ DF） and incremental AF/DF schemes can be analyzed in the unified way. Thus, we can always choose the best cooperative diversity scheme in different scenari-os for delay minimization.展开更多
e consider a two-station tandem queue with no intermediate buffer. Jobs at the first station may be blocked when the following station is occupied by another job. The objective is to control the arrival and departure ...e consider a two-station tandem queue with no intermediate buffer. Jobs at the first station may be blocked when the following station is occupied by another job. The objective is to control the arrival and departure processes, subject to some capacity limits, so that an expected discounted profit function is maximized. We prove that the optimal control policy is of a threshold type, and the characterization of the threshold is provided.展开更多
基金This work was supported by the National Basic Research Program of China under Crant No. 2012CB316001 the National Science Foundation of China under Crants No. （：13832008, No. 03902001.
文摘A lot of work has been focused on desig-ning and analyzing various cooperative diversity pro-tocols for wireless relay networks. To provide a uni-fied queuing analytic framework, we fonmlate an em-bedded Markov chain, which rams out to be a Quasi-Birth-and-Death （QBD） process. Using the Matrix-Ce-ometric method, we can analyze the average delay in a unified way. Theoretical analysis is validated by simu-lation results. We show that the delay performances of Amplify-and-Forward or Decode-and-Forwaxd （AF/ DF） and incremental AF/DF schemes can be analyzed in the unified way. Thus, we can always choose the best cooperative diversity scheme in different scenari-os for delay minimization.
文摘e consider a two-station tandem queue with no intermediate buffer. Jobs at the first station may be blocked when the following station is occupied by another job. The objective is to control the arrival and departure processes, subject to some capacity limits, so that an expected discounted profit function is maximized. We prove that the optimal control policy is of a threshold type, and the characterization of the threshold is provided.
