Optimal Random Access Strategies for Trigger-Based Multiple-Packet Reception Channels

This paper focuses on trigger-based (TB) random access (RA) strategies for a multiple-packet reception channel with channel capability
$M$
(
$M$
-MPR channel), where up to
$M$
packets can be received simultaneously, while more than
$M$
concurrent packet transmissions result in collisions and are considered lost. We model the contention for the TB MPR framework and derive the optimal RA strategies that maximize two metrics:
i)
the normalized saturation throughput, and
ii)
the number of stations successfully occupying the MPR channel within each access round. We generalize the
$p$
-persistent carrier sense multiple access (CSMA) by enabling it to explore both the MPR dimension and the time dimension to adapt the access probabilities. We also propose suboptimal strategies to reduce the complexity, customized for the considered TB framework. Comprehensive performance evaluations and comparisons with respect to a wide range of system parameters and metrics are provided.