Channel-coded Collision Resolution by Exploiting Symbol Misalignment

TL;DR

This paper introduces a novel decoding scheme that leverages symbol misalignment and channel coding to improve collision resolution in random-access networks, significantly reducing failure rates.

Contribution

It proposes an integrated symbol-level decoding approach that combines collision resolution with channel coding, outperforming separate processing methods.

Findings

Outperforms existing collision resolution schemes

Reduces failure rate in packet decoding

Enhances success probability in collided packet scenarios

Abstract

In random-access networks, such as the IEEE 802.11 network, different users may transmit their packets simultaneously, resulting in packet collisions. Traditionally, the collided packets are simply discarded. To improve performance, advanced signal processing techniques can be applied to extract the individual packets from the collided signals. Prior work of ours has shown that the symbol misalignment among the collided packets can be exploited to improve the likelihood of successfully extracting the individual packets. However, the failure rate is still unacceptably high. This paper investigates how channel coding can be used to reduce the failure rate. We propose and investigate a decoding scheme that incorporates the exploitation of the aforementioned symbol misalignment into the channel decoding process. This is a fine-grained integration at the symbol level. In particular, collision resolution and channel decoding are applied in an integrated manner. Simulation results indicate that our method outperforms other schemes, including the straightforward method in which collision resolution and channel coding are applied separately.