Efficient Neuro-Fuzzy Based Energy-aware Relay Selection in IoT-enabled SDWSN

The Internet of Things comprises wireless sensor devices (nodes) that work together to create a dynamic network without central management or continuous assistance. Due to their high mobility, sensor nodes cause periodic topological changes in the network that cause link failures, frequently forcing nodes to rediscover new routes for efficient data transmission in the IoT. This process consumes more energy, which makes the network’s lifetime shorter. This brings attention to energy management and network lifetime issues. A single artificial intelligence technique is insufficient to solve these issues. A relay selection is one way to reduce node energy while routing the data in an IoT network. The proposed work aims to develop an efficient energy-aware relay node selection during the routing process using an adaptive neuro-fuzzy model (ANFIS). The proposed work utilizes a centralized controller architecture called software-defined network which minimizes the overhead of sensor nodes by managing the topology control and routing decisions through intelligent algorithms. This paper presents an energy-aware relay selection technique (ERST) using an ANFIS to optimize the overall energy usage and improve the span of the network. The relay node is selected based on the remaining energy, signal strength, mobility, and the expected transmission ratio of the nodes, which is given to the fuzzy inference system to make intelligent decisions based on the fuzzy rules and neural network used to fine-tune the fuzzy system to select the optimal relay node. The proposed work is evaluated using MATLAB and NS3 simulators. The obtained results of the suggested work outperform the previous protocols by minimizing 5% of end-to-end delay and 4% of energy usage and maximizing 8% of average throughput, packet delivery, and overall network lifetime. The proposed ERST achieves efficiency, reliability, and scalability in IoT.