Abstract:
Due to the recent advancements in micro-electromechanical systems (MEMS) in the last a few
decades, we have witnessed mushroom growth of wireless sensor networks (WSN) and their ap plications. With the fast pace of life, the Internet of Things (IoT) is the dire need of today’s life.
Everything in this world is interconnected in this way that, dependency is increasing day by day.
With the advancements in the field of artificial intelligence and IoT systems, life is becoming
more easy. IoT systems with the blend of advance machine learning and artificial intelligence
algorithms can provide much more efficient solutions for human race.Using cheap internet en abled devices like sensors, actuators and microprocessors, one can design efficient solution for
day to day life. The interconnection of these things resulted in more complex problems like
security, the life of the network, and reliability. This system will bring a lot of ease for human
beings to the table but on the other hand, is a real risk to the human race. Sending minimal data
over the internet can help to increase the wireless sensor network life also mitigate the risk of
being compromised.
Through modifications in the architecture of sensor networks and introducing advance load
balancing techniques, iFogsim is used to simulate the wireless sensor network life and improve ments are recorded. First order radio model assisted to calculate the estimated wireless sensor
network life along with the network usage stats and processing time. Previously IoT systems
were used without any machine learning algorithms which caused a lot of hindrance during
the execution of their applications. Future predictions are important aspect in internet enabled
world which can assist human beings at its best. We have seen a cumulative increase in the per formance of about 20 percent with respect to power using optimized architecture and machine
learning. This research will focus on providing machine learning assistance to smart car parking
IoT system which can further help to predict the sensor network life. Hence predicting sensor
network life can improve efficiency and prevent nodes from dying.
