Using a genetic-based algorithm to solve the scheduling optimization problem for long-range molecular communications in nanonetworks

Inspired by nature, nanonetworks can expand the capabilities of nanomachines with regards to execution of more complex tasks, extension of workspace and coverage. Various communication techniques such as acoustic, nanomechanical, electromagnetic and molecular communications are proposed for nanonetworks but molecular communication is the most promising method. Molecular communication has three categories in terms of distance between nanomachines: short range, medium range, and long range. In this paper, axon-based molecular communication as one of wired long range molecular communication options in nanonetwork is studied. Since axon-based molecular communication uses neural network (body area network (BAN)) as an underlying structure to connect nanomachines, neuron and neural network are described. This paper uses a neural time division multiple access (NTDMA) scheduling to allow nanomachines to send their information through a shared neural network to the destination nanomachine (sink) without any collision and interference. In a NTDMA protocol, nanomachines activate neurons, one after the other, in their own time slots. Since determination of these time slots is a multi-objective optimization problem, thus a genetic-based algorithm (GBA) is proposed to solve it. Simulation results verify that GBA obtains the best feasible solutions and outperforms some existing methods.