Adaptive Aggregation of IoT Data Packets in Near-Edge Areas in Wide Area Networks

Our everyday lives will increasingly revolve around the Internet of Things. A lot of little data packets flow in networks as a result of IoT devices frequently contacting required servers and other devices via the Internet. This causes high overhead, inefficient network performance, and a growth in the volume and quantity of packets in network nodes. Numerous packet aggregation strategies have been created or researched in order to address these issues. A few of them deal with IoT packet aggregation in wide area networks, but they don't go into great detail or examine various aggregation strategies. The packet aggregation technique proposed in this paper can choose from a range of useful alternatives that are appropriate for wide-area near-edge software-defined networks. The suggested packet aggregation mechanism avoids the increase in latency brought on by an increase of application layer message types by carrying out aggregation at a lower level than the application layer, utilizing a unique aggregation header, and employing certain suitable techniques, so it has a lower end-to-end delay than related works while maintaining the same level of packet reduction. The proposed mechanism can classify and aggregate packets based on their type, along with features like ToS and algorithms like K-Means, and is capable of adapting to the packets conditions. The suggested mechanism's effectiveness and superiority over previous works of a similar kind are demonstrated by the simulation results for various modes, which take into account the network's traffic from Internet of Things-based smart cities. Not only have simulations been conducted, but essential conversations and comparisons have also been undertaken in this regard.