This paper investigates the throughput of a Framed Slotted Aloha (FSA) random access scheme, called Segmented FSA (SFSA), in which each packet is encoded and subdivided into segments before the transmission, and where Interference Cancellation (IC) is adopted to improve the success probability in case of collision between segments. The overall system performance is estimated by considering a decoding criterion based on the average information rate that is experienced during each communication. This criterion, which enables to properly consider the capture effect in the presence of interference, noise, and fading, is compared to the classic packet erasure channel model, with the aim of identifying the reliability of the two decoding approaches in different network scenarios. The final purpose of the presented study is to discuss the benefits of the interference diversity mechanism that is triggered by the combination of segmentation and IC in a capture channel.
