Modeling of bio-nano communication networks for the human body

• Main Author: Yang, Yiqun
• Other Authors: Yeo Chai Kiat
• Format: Theses and Dissertations
• Language: English
• Published: 2015
• Subjects: DRNTU::Engineering::Computer science and engineering::Computer systems organization::Computer-communication networks
• Online Access: https://hdl.handle.net/10356/65639
• Institution: Nanyang Technological University

Nanonetwork is a new research focus which presents solutions for agricultural, environmental, medical and security fields. Bio-nano hybrid network, which aims to be applied to the human body, is defined as the cooperation of biological units and possible artificial nanomachines. The physical basis of bio-compatible nanonetwork is the intrinsic signal transmission, encoding and decoding in the human body. By studying the communication mechanism, novel nanonetwork could be suggested and further integrated into the standard computer networks. In the meantime, it is also significant for both researchers of experimental biology and computer science to complete a model in cell level, either to provide chances for signal control and recovery in vivo, or to inspire new methods for artificial intelligence. Calcium signaling is a ubiquitous phenomenon in living creatures which takes the responsibility of mediating other messengers and inducing cellular activities. Calcium signaling prevails in astrocyte - a kind of non-electrical cell in the nervous system. Therefore, astrocytes undertake information transmission using calcium waves other than electric pulses. The function of astrocyte in the nervous system has been emphasized in recent years, especially its collaboration with the neurons. Besides, the communication between adjacent astrocytes forms a network with nonidentical intra-network channel efficiency. In addition, given the universal existence of calcium signaling (not exclusive in astrocytes), nano controllers are expected to participate and interfere with the original communication procedure thus leading to signal detection and regulation. This will facilitate disease monitoring and treatment. In this thesis, a conceptual network model is proposed to express the signal transmission from the Peripheral Nervous System to the Central Nervous System. It is divided into four layers: neurons in the Peripheral Nervous System and the spinal cord, neurons in the brain, single astrocyte and groups of astrocytes. The external stimulation is eventually transformed to the states of the grouped astrocytes. The state combination signifies the current prototypical cognitive pattern induced by the stimulation which is converted into conscious episode through associating with the previous patterns. The parameters of each layer are obtained from specific configurations (e.g. the olfactory sensory system in layer 1 in the model) which could be adapted to other sensory systems. In addition, the model is scalable as the metrics employed such as network range, the number of astrocytes and neurons can be altered to emulate the real-time signal processing in the human body. In addition, the author builds the nanonetwork of astrocytes and neurons in the cerebral cortex based on their columnar and laminar arrangement. The complete communication procedure from the thalamic input to cortical Layer 5 output is simulated. More importantly, astrocytes are integrated into the neuronal micro-circuits and its function has been analyzed. As an assumed memory unit, astrocytes process the neuronal activities and generate new responses by coupling the past ones in the local domain. Meanwhile, astrocytes also maintain an interconnected network to deliver information to selected neighbors. The author also envisages a network architecture based on the properties of calcium signaling. Firstly, a detailed explanation of calcium signaling and a more accurate simulation method are provided. Then a two-level network protocol is designed and its performance analyzed. A two-layer protocol stack is established which draws its inspiration from the standard computer network. The physical layer is concerned with the communication channel and entities. The cells are aligned in clusters with dynamic control in certain area. The network layer includes the routing protocols. The layered stack encapsulates the complicated biological process to enable convenient analysis of the message transmission by abstracting the lower layer. Human intervention via nano-controllers at the essential nodes (i.e. central node and gateway nodes) realizes the simple routing according to the routing tables which enhances the message direction control and alleviate the bottleneck effect. Relatively long distance communication is accomplished while the success rate is largely elevated. Given the many source-destination pairs in a typical biological trial, the author is able to model the communication path of a single source-destination pair. The error probability in the proposed ring topology is analyzed by considering the leakage to the uncontrolled clusters to make the results more complete and accurate.