Molecular communication via diffusion with spherical receiver & transmitter and trapezoidal container

摘要

Molecular communication is a new paradigm that works on the principle of the discharge of molecules as carriers of information. The channel of molecular communication is determined by the signal that is received. The designs of the receiver, the modulation, coding, and capacity rely on the models of the received signal. Therefore, it is difficult to model the number of the received modules (until the time t) using an analytical approach. Indeed, communication through molecular diffusion is ubiquitous. In particular, the present, closed form of the impulse response of the diffusion channel usually perceives the transmitter as a source point. This research attempts to develop molecular communication via diffusion through a spherical transmitter and receiver and trapezoidal container. The study has been conducted in two parts. These parts include dataset generation using a probability distribution, and the modelling of the received signal - using a multitude of machine learning (ML) techniques. In this study the code is executed using Python language. In addition, the study utilizes deep neural networks, support vector regression, logistic regression, and ridge regression to model the received signal for a spherical transmitter before comparing the performance of all the models. The results of the study have been compared in terms of the models' accuracy to determine the best performing approach. The proposed techniques offer promising results for modelling the received molecules number until time. The proposed technique models the received signal efficiently for longer distances. (C) 2020 Elsevier B.V. All rights reserved.