Investigation of Energy Optimization for Spectrum Sensing in Distributed Cooperative IoT Network Using Deep Learning Techniques
M. Pavithra, Rajendrane Rajmohan, Tamilarasan Ananth Kumar, S. Usharani, P. Manju Bala · 2022
The growing popularity of the Internet of Things has led to large datasets from multiple diverse devices. This has caused an increase in the use of energy by these connected devices. Smart buildings are specific types of buildings with several microcontrollers and sensor systems, acquiring a large amount of data at a redundant stage. As a result, cloud solutions may not be as useful as possible due to delays by the cloud solution. This needs a massive number of computers at the cloud service end to manage the data generated by these sensors, which does not fit the green computing criteria. Deep learning has been a great success in recent years in a variety of smart IoT applications. This chapter investigates the application of deep learning methods to Internet of Things (IoT) applications focusing on energy optimization mechanisms. The increasing use of wireless technology and the user demands results in the need for wireless communication systems that are larger and more complex than ever. We believe that deep learning methods will play a significant role in solving next-generation wireless communications issues. This article discusses cognitive radio, an enticing advanced technology to enhance spectrum efficiency and develop deep learning-based techniques to enhance cognitive radars’ spectrum sensing abilities. The study looks at the energy efficiency of distributed cooperative sensing being formulated as a combinatorial optimization problem. From the foundation laid by our previous work, we develop a deep learning framework that incorporates concurrent neural networks and reinforcement learning to improve the overall system's energy efficiency. Simulated studies using different network sizes demonstrates the effectiveness of our proposed approach.