Bounds on the Capacity of ASK Molecular Communication Channels with ISI

TL;DR

This paper derives bounds on the capacity of ASK molecular communication channels considering inter-symbol interference caused by transit time randomness, providing insights into channel limits and optimal input distributions.

Contribution

It introduces upper and lower bounds for ASK molecular channels with ISI and adapts the Blahut-Arimoto algorithm to optimize input distributions.

Findings

Bounds are close over wide parameters

Modified BAA effectively finds optimal input distributions

Provides new insights into channel capacity limits

Abstract

There are now several works on the use of the additive inverse Gaussian noise (AIGN) model for the random transit time in molecular communication~(MC) channels. The randomness invariably causes inter-symbol interference (ISI) in MC, an issue largely ignored or simplified. In this paper we derive an upper bound and two lower bounds for MC based on amplitude shift keying (ASK) in presence of ISI. The Blahut-Arimoto algorithm~(BAA) is modified to find the input distribution of transmitted symbols to maximize the lower bounds. Our results show that over wide parameter values the bounds are close.