LAUSR

Recent advances in scheduling transmissions for underwater acoustic networks utilize and exploit long propagation latency of acoustic waves for achieving throughput gain. These techniques utilize the propagation delay information of the considered network geometry and schedule transmissions in time slots. Time-slotted transmissions are such that most of the interference overlaps with the transmission slots and the receiving slots are interference free. Moreover, exploiting propagation delays lead to multiple transmissions per time slot, thereby resulting in higher throughput. However, the packet duration of each transmission in the time slot is assumed to be fixed. The packet duration, however, provides a degree of freedom that, if utilized, results in strategies that can be adopted to achieve throughput closer to the established upper bound. Therefore, we consider the problem of finding unslotted transmission schedules allowing unequal packet duration. Given the propagation delay between nodes in the network and packet traffic demands, we formulate an optimization problem for minimizing the fractional idle time in a frame (or period) of the schedule as a mixed-integer linear fractional problem (MILFP). We compare our results to the recent advancements that exploit large propagation delays and result in time-slotted and unslotted schedules with fixed packet duration. We also present schedules computed for various network geometries with arbitrary packet traffic demands.