Broadcast Coded Slotted ALOHA: A Finite Frame Length Analysis

TL;DR

This paper introduces B-CSA, a novel uncoordinated broadcast MAC protocol that leverages coded slotted ALOHA with finite frame analysis, demonstrating superior performance over CSMA in vehicular networks.

Contribution

It provides a finite frame length analysis of B-CSA, including stopping set analysis and an approximation of error floor performance, highlighting its advantages in vehicular communication.

Findings

B-CSA outperforms CSMA in supporting more users at the same reliability.

The analytical approximation accurately predicts B-CSA's error floor performance.

B-CSA exhibits a double unequal error protection phenomenon due to half-duplex operation.

Abstract

We propose an uncoordinated medium access control (MAC) protocol, called all-to-all broadcast coded slotted ALOHA (B-CSA) for reliable all-to-all broadcast with strict latency constraints. In B-CSA, each user acts as both transmitter and receiver in a half-duplex mode. The half-duplex mode gives rise to a double unequal error protection (DUEP) phenomenon: the more a user repeats its packet, the higher the probability that this packet is decoded by other users, but the lower the probability for this user to decode packets from others. We analyze the performance of B-CSA over the packet erasure channel for a finite frame length. In particular, we provide a general analysis of stopping sets for B-CSA and derive an analytical approximation of the performance in the error floor (EF) region, which captures the DUEP feature of B-CSA. Simulation results reveal that the proposed approximation predicts very well the performance of B-CSA in the EF region. Finally, we consider the application of B-CSA to vehicular communications and compare its performance with that of carrier sense multiple access (CSMA), the current MAC protocol in vehicular networks. The results show that B-CSA is able to support a much larger number of users than CSMA with the same reliability.