CAIN: An energy-aware and intelligent increasing coverage area routing protocol for future 6G networks

Abstract

More devices are being introduced to the cellular network every day, and it is time to think about how the next generation will deal with them. The 6G will bring new ways to think about how to share information and will not rely just on communication with a fixed Base Station (BS). A post-disaster scenario may damage the network infrastructure; the devices need to create an ad-hoc network to communicate in this situation. The ad-hoc network must achieve some of the 6G’s requirements, such as communication frequency, data rate, and latency, to communicate with the 6G networks when available. The increase in the frequency used to achieve such data rate has the downside of decreasing the signal range, i.e., for two nodes to communicate, they must be closer to each other. Academia is developing alternatives to increase the network coverage area, allowing devices to further the ad-hoc Access Point (AP) to overcome this drawback. Thus, with this work, we propose CAIN, an 6G routing protocol that uses Artificial Intelligence (AI) techniques to increase the network coverage area. Depending on the node’s connectivity to an AP, i.e., if the node is connected to an AP or not, CAIN uses different messages to make nodes further the AP to communicate with it, increasing the network coverage area. Hence, if there is no connected node to an AP, it chooses a neighbor with a connection to one to forward the message. Also, if there is no connected neighbor to an AP, the node chooses a neighbor to flood the message until it reaches an AP. To improve its performance, we enhance CAIN with AI techniques, such as Reinforcement Learning (RL), Federated Learning (FL), and Deep Neural Network (DNN). These techniques allow the node to make a better neighbor choice and increase the network coverage area even more. To the best of our knowledge, this is the first work that proposes a routing protocol for the 6G scenario that uses RL techniques with FL and DNN to increase the network coverage. With these characteristics, CAIN makes devices further the APs to communicate with it using less energy and storage resources, producing less delay compared to other protocols.