Abstract
D.Ing. (Electrical Engineering)
A wireless sensor network usually has a sensor to monitor physical and environmental conditions. The popularity of the wireless network has led to the need to update networks with a reliable and efficient communication mechanism and to ensure speed and security. In this research an energy-efficient network is proposed by reducing the data size by maximising the network lifetime, enhancing the security pattern for safe transmission of packets over a network, an adaptive pre-processing technique for streaming through wireless network and an efficient Time Division Multiple Access (TDMA) scheduling method for data collection minimisation using multiple sinks over a wireless sensor network.
In the research, the network lifetime has been maximised using the data compression mechanism with the Huffman code, LEACH and Dijkstra algorithm which show efficient nodes for data transmission. During transmission, a major problem occurs in terms of security mainly in the military or other high security fields where many attackers try to hack the data. Thus, the research performed on the network security set out to generate a security pattern to safeguard data that is transmitted over a network. Some of the other common problems caused in a network during packet transmission are noise, redundancy and missing values. To solve these problems, the work performed introduced an adaptive pre-processing technique using Principal Component Analysis (PCA) and Hyperbolic Hopfield Neural Network (HHNN) to make streaming data efficient. This process provides higher efficiency by increasing the prediction accuracy. The other major problem with the WSN is data collection and scheduling in multiple sink environments. To overcome the problem, the study used TDMA scheduling to preempt the delay in data collection and reduce high energy consumption while the pocket-driven trajectories (PDT) algorithm reduces the iteration.
The results show the success of the research in the simulation. The simulation results indicate high-performance efficiency in terms of reducing delay, increasing network lifetime, increasing security, reducing energy consumption and reducing data loss during data transmission in the wireless sensor network. Thus, the proposed research ended successfully by giving better results than other existing methods that are used to solve certain problems handled by wireless sensor networks.