Efficient Resource Allocation Algorithm for Underwater Wireless Sensor Networks Based on Improved Stochastic Gradient Descent Method
Meiqin Tang, Changjing Ren and Yalin Xin
Due to the poor wireless communication performance of underwater wireless sensor networks (UWSNs) nodes in underwater environment, acoustic transmission is usually used for communication. Resource allocation efficiency and lifetime of nodes are very important in UWSNs. In this paper, multi-homing technology is used in UWSNs. Considering the actual network rate constraint, power constraint and energy return constraint, the multi-homing technology is adopted for UWSNs. Then, an optimization model of network resource allocation is established to maximize the throughput of communication system under the mechanism of energy borrowing and energy recovery. Based on the advantages of gradient descent method, such as simplicity, fast convergence speed and reliable effect, an improved stochastic gradient descent algorithm is proposed. In each iteration process of the algorithm, it is not necessary to traverse all the data, only a random sample is selected to calculate the gradient, and the weight vector is updated iteratively. The momentum factor ensures the optimal step size and greatly reduces the computational complexity of the algorithm. The convergence and numerical simulation results show that the algorithm can effectively reduce node energy consumption and improve the throughput of UWSNs.
Keywords: Underwater Wireless Sensor Networks (UWSNs), Improved Stochastic Gradient Descent Method, Resource Allocation