Research On Clustering Algorithms Of Ad-hoc Network Consisting Of Static Nodes Using Underwater Wireless Optical Communication

The research goal of this thesis is to lay the foundation for the future development of a high-bandwidth,high-throughput marine big data system.After comprehensively considering the networking difficulties and the cost of underwater wireless optical communication system,we intend to use the low-cost,mobile-restricted nodes(ie,static nodes)without maneuverability for networking.It would form a dynamic,small autonomous system(ie,mobile ad-hoc network)that nodes cooperate with each other and would include multi-hop routing in complex,changeable and shallow sea environment,while we also consider the limited capacity and limited energy of static nodes,changes of network topology and other factors undersea.The simulation work is mainly carried out on Matlab platform.The specific contents would include: combining the real environment of the ocean to simulate node movement model,background noise model,absorption and scattering model of light,etc.;an efficient clustering routing algorithm which would combine with an improved iterative self-organizing data analysis technique algorithm,cluster head election mechanism and link quality evaluation mechanism.The primary task of studying underwater wireless optical communication is to study the attenuation of light underwater.In this environment,the substances that have a serious impact on the transmission performance of light are particles and dissolved substances in the ocean.Both scattering and absorption effects play a vital role in light attenuation.Scattering includes Rayleigh scattering caused by small molecule substances,such as water molecules and metal compounds dissolved in seawater,and Mie scattering introduced by many large molecules and insoluble substances.The effect of absorption on light is reflected in the energy loss when the photon collides with the object,and the light distribution changes accordingly.In the process of information transmission,these effects usually cause a significant increase in the bit error rate in communication system.Therefore,it is very important to evaluate the quality of the underwater communication link,which can pave the way for improving the performance of the underwater wireless optical communication system.Underwater wireless optical communication sensor network is a dynamic sensor network composed of a large number of sensor nodes with limited communicationcapabilities,limited power,and limited capacity.They are usually difficult to be replaced undersea.Therefore,extending the life cycle of the network and improving its energy utilization are the key performance indicators of underwater wireless sensor networks.Therefore,we plan to use machine learning to design an energy-efficient clustering routing algorithm suitable for three-dimensional network underwater.The cluster head nodes are selected in the network,mainly based on their communication load,energy consumption,location and other information.The non-cluster head nodes perform clustering on demand to save their energy as much as possible,and avoid premature death of nodes,to extend the life cycle of the network,and improve the energy utilization rate of the network.In addition,since the communication link in this dynamic environment is unstable,a delay-tolerant state would always exist in the network.Therefore,designing an evaluation mechanism of link quality is also beneficial to improve the routing and reduce the energy consumption of the network.