Quantum Node Clustered Cuttlefish Optimization Based Deep Belief Network for Secured Data Aggregation in UWSN

Abstract

Underwater Wireless Sensor Network (UWSN) technology is used in different underwater monitoring and exploratenapplications. The data aggregation is the process of gathering the datain UWSN to attain better outcomes. A Quantum Node Clustered Cuttlefish Optimization based Deep Belief Network (QNCCO-DBN) Technique is introduced for secured data aggregation in UWSN with lesser packet droprate and lesser end-to-end delay. QNCCO-DBN Technique comprises two processes, namely clustering and secured data aggregation in UWSN. Deep Belief Network in QNCCO-DBN Technique perform four layers, namely one input layer, two hidden layer and one output layer in UWSN. Initially in QNCCO-DBN Technique, underwater sensor nodes are considered as the input at input layer for performing secured data aggregation. Input layer transmits the number of underwater sensor nodes to the hidden layer 1. In that layer, Quantum k-means Node Clustering is carried out to group the underwater sensor nodes into different clusters depending on the energy level. After that, the underwater sensor node with higher energy level is considered as the cluster head for every cluster. Then, the cluster head aggregates the data from remaining underwater sensor nodes and optimal cluster headis selected using Improved Multi-Criterion Cuttlefish Optimization in QNCCO-DBN Technique for transmitting the data packets to the base station. The proposed technique initializes the populations of the cluster head (i.e Cuttlefish). The fitness of each cluster head is calculated based on distance and trust. Then, aggregated data sent to the optimal cluster head in UWSN. By this way, secured data aggregation is performed in underwater sensor network. We administer considerable simulations to measure the performance of our proposed method and compare it with other two routing algorithms on NS2 platform. Experimental evaluation is carried out in QNCCO-DBN Technique on factors such as energy consumption, packet droprate, end-to-end delay and data confidentiality rate with respect to number of underwater sensor nodes and number of data packets.