Enhanced clustering and data aggregation approaches using heuristic and machine learning techniques in IoT

Abstract

Internet of Things(IoT) generally provides information about all the objects which are connected to the internet. It controls and manages the functions remotely without any human intervention. It has the capacity of responding to the conditions instantly or through their experiences. Similarly, the machines can gain their experiences from the surrounding specific to the applications and react accordingly without any human intervention. In order to analyze the necessary information from the environment, more number of sensors are deployed throughout the environment. They are growing rapidly in all the fields from the industrial environment to smart home. Sensors are helping to monitor and collect the data from all the real time gadgets which are really dependent on all the kinds of basic needs to the sophisticated environments. This research work mainly focused on improving the efficiency of sensing and network layer of IoT. Sensors are resource constrained devices, therefore it is necessary to find an efficient way to react, to analyze and to forward the sensed data to the base station. Resources such as energy, computing power and storage vary with the different type of sensing devices and communication technologies which are used to connect the real world objects. Physical and medium access control layer of the sensor networks diverge their applications in different spatial and temporal regions. Depending upon the application requirements, transmission coverage range, energy consumption and communication technologies differ independently from low constraint to high resource enrich gadgets, which in turn directly affects the performance and decreases the overall network lifetime of the massive Internet of Things environment. Spatially, identifying and communicating corresponding objects in a massively distributed Internet of Things environment is crucial. newline newline newline