Modeling and performance analysis of diffusion-based molecular communication system for nanonetworks

Abstract

Advances in biological science and nanotechnology make it possible to manufacture biological materials, such as small-scale functional devices, which are capable of interacting with biological molecules and cells. In existing literature, the biological devices had been studied, such as purified protein molecules, genetically engineered cells, artificial protocells, and bio-silicon hybrid devices. Since the bio-nanomachines have their own characteristics, such as its small size, bio compatibility, and high energy efficiency. The bio-nanomachines can cooperate with existing biological systems for biomedical engineering, environmental control, Information and Communication Technology (ICT) and military applications. Among the possible solutions of communication between engineered bio-nanomachines, which were suggested in existing literature, the molecular communication paradigm is expected to be most promising solution. The molecular communication was inspired from the natural cell communication in biology, where information molecules are synthesized, released and received through biochemical process. Molecular communication has been considered as one of the most promising technologies for nanonetworks, which is used for interaction among the nano-devices.

In this thesis, we considered the diffusion-based molecular communication system. The nano-devices which are most important components of nanonetworks, has limited properties, such as size, computational capabilities or power capacities. Although it is difficult to implement a synchronous system in nano-devices processing simple tasks, only synchronous systems were assumed for easy analysis in all the existing literature. Thus, in this thesis, we proposed the frame of asynchronous diffusion based molecular communication system with the timing modulation. Also, we proposed the asynchronous symbol detection algorithm using the variance of the arrival time of a large number of information molecules arriving at the receiver. The performance of proposed symbol detection algorithm was analyzed in terms of error probability, and it is shown that there is good agreement between analysis and simulation results.

On the other hand, the propagation time of information molecules in the diffusion-based molecular communication is extremely long compared with that of conventional electro-magnetic wave communication. The inter-symbol interference (ISI) problem is not avoidable in diffusion based molecular communication. Thus, it is necessary to address the ISI problem to achieve the high data transmission rate. Here, we proposed a new concept of diffusion based molecular communication system with biocatalysts, i.e., enzymes and enzyme inhibitors. In the existing literature, the enzymes were used to reduce the ISI by decomposing the information molecules. Here, we adopt the enzyme inhibitor to near the receiver, and then the information molecule in near the receiver may not be decomposed. As a result, the arrival probability of information molecules at the receiver was increased and the error probability of proposed system was better than that of the system that uses the enzymes and that of the system that does not use enzymes.

In addition, we proposed a diffusion-based molecular communication system with the sequential catalytic process. The information molecules are changed into various forms, because of the various types of enzymes present in the medium. The receiver detects the various types of information molecules, thus, the arrival probability of information molecules can be enhanced within the symbol time. Also, since the information molecules are completely decomposed after a certain time, ISI might be reduced. Moreover, we derived the probability that the information molecules arrive at the receiver and SIR performance in the proposed system. Finally, we confirmed how the arrival probability and SIR performance affect to the error probability, and prove the outstanding performances of the proposed system, using a simple detection algorithm.