Design And Implementation Of Energy Optimized Routing Protocol For Underwater Wireless Sensor Networks

Underwater wireless sensor network is the key to open the underwater world,and it has been widely used in the fields of marine pollution monitoring,resource exploration,submarine geological disaster prevention and national defense.It consists of a group of sensor nodes with acoustic transmitters,and data packets are forwarded from the bottom of the sea to ships or transit stations on the water surface in a multi-hop manner.Routing protocols determine the forwarding behavior of packets and are the core of underwater networks.Nodes in underwater networks are usually powered by batteries with limited energy,so how to improve the energy efficiency of underwater networks and prolong the network lifetime is a key issue that must be considered in the design of routing protocols.Classical underwater routing protocols usually consider a single factor when forwarding packets.For example,VBF and HHVBF protocols only consider the location factor,which leads to frequent forwarding of some sensor nodes in the network and depletion of energy in advance,thus shortening the survival time of the network.In addition,they do not consider the future path when selecting the next hop node,so they cannot choose the global optimal path.In order to solve the above problems,this paper proposes an underwater network routing algorithm based on reinforcement learning and fuzzy inference system.Markov decision model is used to model the packet forwarding process,and the forwarding node is selected from the global consideration through model-based learning.A reward function based on a fuzzy inference system is constructed.When calculating the reward value,multiple factors such as node link quality,remaining energy and node location are considered comprehensively,which can reduce the number of retransmissions and balance the energy consumption between nodes,thereby reducing energy consumption and extending Network lifetime.In this paper,experiments are carried out on the Aqua-Sim-NG simulation platform and compared with the typical underwater routing protocol VBF.The experimental results show that the routing algorithm proposed in this paper has a significant improvement in the total network energy consumption and network lifetime and other performance indicators.In order to better study the routing algorithm based on fuzzy inference system so that it can be quickly applied to the actual scene,this paper designs and implements the relevant hardware module of fuzzy inference system to calculate the direct reward value by using FPGA,and verifies it on the PYNQ-Z1 development board.