No abstract

a b s t r a c tThis paper investigates the situation where a group of terminals can be simultaneously connected to several base stations, using distinct technologies on non-overlapping frequency bands. We introduce and solve the problem of optimal power allocation (in the sense of the ergodic sum-rate), for terminals in the uplink, using three types of receiver: the optimum receiver, minimum mean square error and matched filters. Key results from random matrix theory allow us to solve the corresponding optimization problems. Simulations validate our approach, and illustrate the performance gain obtained by using several technologies simultaneously, instead of one at a time.

In this contribution, we derive the optimal power allocation under target rate constraints in the case of an uplink multiuser system. Using asymptotic results based on random matrix theory, we provide a unified framework for determining the optimal decoding order when using Successive Interference Cancellation (SIC) receivers, namely the MMSE (Minimum Mean Square Error) and Matched filter SIC. Moreover, for decentralized systems, we show that each user can, under certain conditions, solely determine the power allocation based only on his channel energy and the statistics of the other users. Simulations in the case of finite systems validate the asymptotic claims.

No abstract