Data transmission scheme based on node model training and time division multiple access with IoT in opportunistic social networks

With the advent of 5G technology, the Internet of Things has shown a rapid development trend. A large number of IoT communication devices have formed a community-like network within a certain range. These devices can be viewed as network nodes in the opportunistic social network. In traditional opportunistic routing protocols, data broadcasting strategies are often used for data transmission. Under the requirement of fast transmission of a large amount of data, it is easy to cause data redundancy and node death, resulting in high data transmission delay and high energy consumption. At the same time, node devices in opportunistic networks often need to undertake some complex learning and computing tasks, which leads to an increase in transmission delay and also leads to data privacy and security issues. To this end, we propose a data transmission strategy (MTDR) based on node model training and time division multiple access. A horizontal federated learning model architecture is established between the source node and other user nodes participating in computing, allowing edge nodes to use data locally to complete computing tasks, thus avoiding the high latency of data transmission for complex tasks. This effectively improves network bandwidth pressure and improves data privacy. The node performs uplink data transmission based on the time division multiple access mode, which improves the data transmission performance. Experimental results show that our method could significantly reduce system delay and energy consumption. It can also increase the data transmission rate compared with other opportunistic routing algorithms.