Information rate analysis of ASK-based molecular communication systems with feedback

Abstract

In this paper we develop lower and upper bounds on the capacity of an amplitude shift keying (ASK)-based molecular communication (MC) system with feedback. Analyzing the effect of feedback on the performance of MC is motivated by the growing use of feedback in controlling drug delivery. Based on causal knowledge of the number of transmitted and received molecules, the input probability of symbols is adapted so as to maximize directed information in the molecular communication channel. We considered one dimensional channel with drift velocity caused by blood flow. In our system, molecules propagate in a fluid with a drift velocity; the receiver absorbs the molecule unless it is saturated (saturation models a limit on ligand binding). The input is limited by a toxicity constraint of injected molecules. We also study the effects of feedback on the achievable information rates in terms of the time delay in receiving the feedback. We show that, especially for higher values of toxicity constraint and sequence length, feedback, in terms of causal knowledge of the number of delivered molecules, improves performance of ASK-based molecular communication.