Abstract :
Generalized selection multiuser diversity (GSMuD) is a new scheduling scheme which provides a near-optimal low-complexity solution to parallel access multiuser scheduling. In this scheme, a total of L users awaiting transmissions are ranked by their signal-to-noise ratios (SNRs) and the Nc (1 les Nc les L) users with the largest absolute (or normalized) SNRs are selected for parallel channel access. In this paper, we analyze the second order statistics (including the level crossing rate and average fade duration) and channel access performance (including the average channel access rate, the average access time, and the average waiting time) of the proposed GSMuD scheme, taking into account different types of generalized fading channels. Simulation results verify the derived analytical formulas. Numerical results show that compared to the selective multiuser diversity (SMuD), the GSMuD significantly improves the channel access rate and access time, and reduces the access waiting time. The presented results will be useful for the cross-layer design of multiuser parallel scheduling systems.
Keywords :
diversity reception; fading channels; multi-access systems; multiuser channels; scheduling; statistical analysis; channel access statistics; generalized fading channels; generalized selection multiuser diversity; near-optimal low-complexity solution; parallel access multiuser scheduling; Channel capacity; Cross layer design; Delay; Fading; Multiaccess communication; Peer to peer computing; Performance analysis; Processor scheduling; Statistics; Ultra wideband technology;
