Effective capacity optimization for multiuser diversity systems with adaptive transmission

The effective capacity (EC) for a wireless system expresses the maximum arrival source rate that the system can reliably transmit over the wireless channel while fulfilling a probabilistic delay constraint. This paper presents an EC optimization for a multiuser diversity system exploiting proportional opportunistic scheduling at the medium access control layer in conjunction with adaptive modulation and coding (AMC) at the physical layer. We consider the downlink of a time-slotted multiuser system with two different types of channel state information (CSI) at the transmitter, that is, (i) full CSI, whereby each user feeds back its normalized signal-to-noise ratio to the transmitter and (ii) quantized CSI, where each user feeds back the AMC mode for transmission. For each case, we first derive the EC function of the queue service process for the individual users at the medium access control layer. We then design the AMC scheme with power control aiming at maximizing the said function subject to a target packet-error rate constraint. Both independent and identically distributed and time-correlated fading channels are considered. The results illustrate the superior performance of the proposed schemes, when compared with the previous systems such as opportunistic scheduling with single-rate transmission and round-robin scheduling with AMC. Copyright (c) 2012 John Wiley & Sons, Ltd.