الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 156 |  |  |
| | | from | 156 | | |
Abstract
With the rapid growth of the citizen services to exchange the data (voice, text, video…etc.) with anyone, anytime and anyplace with any terminals (devices, sensors, actuators). M2M communications are becoming an intelligent technology to fulfill the people requirements without human being, in addition, to utilize the resources in our daily life. Internet of Things (IoT) technology is coming as an extension and an evolution of the M2M applications. IoT technology coexists with the present cellular communication networks for developing an emerging 5G technology. Billions of IoT big data result from the rapid growth of IoT services upon human beneficiaries, especially in the smart city use case, which exhibits several and different IoT data traffic characteristics and then traffic data complexity. Accordingly, it is difficult to carry and handle IoT data traffic services over Human type communication. The traffic modeling concept of different IoT data characteristics is envisioned in this thesis to fulfill this problem. We propose a novel ON/OFF traffic modeling technique to recover IoT data characteristics of diverse IoT smart city services. In a novel model, the characteristics of IoT smart city use cases are formatted and summarized into five major traffic patterns. To prove the concept, IoT smart city architecture, smart home case study, and realistic smart home networks have been designed and implemented as a case study of massive IoT smart city tracks. The experimental measures present several traffic profiles that are generated from a pilot according to be modeled into theoretical models by Easy-fit statistical tool. A proposed five traffic patterns of the novel ON/OFF Traffic modeling technique are concluded and verified in the experimental results.
The thesis proposes the IoT traffic aggregation study as a major developer technique for improving the performance of the next generation IoT networks. The proposed traffic aggregation is going toward solving the communication link problems and optimizing the IoT network performance in the physical layer. The traffic aggregation approach could exploit the aggregation concept in both space and time to abstract and reduce redundancy raw data, recurring packet headers, the number of packets, and collision probability on one side. Besides that, traffic aggregation can increase the throughput and PDR on the other side.
The thesis installs various IoT network experiments to test the effect of the traffic aggregation approach on the traffic profile shaping and performance metrics of IoT networks. It is proven that the proposed traffic aggregation techniques have the potential impact on the IoT network performance from many perspectives: throughput, collision probability, traffic congestion, and recurring overheads.
Finally, the internet of things could be adapted by modern technologies like machine learning and the block chain in the present time or the near future to reduce an expected latency or delay occurred by processing the Big data (Zettabytes) in the application layer, and maintain the privacy & the security among IoT networks, respectively.