Opportunistic routing for mobile edge computing: A community detected and task priority aware approach

Abstract

As the 5G era develops quickly, there has been a significant growth in the amount of network data. Concurrently, traditional routing algorithms are encountering growing challenges. Traditional routing algorithms typically rely on maintaining stable end-to-end paths to facilitate smooth data transmission. However, in wireless multi-hop networks, the mobility of nodes and the dynamic nature of network topology make maintaining stable end-to-end paths exceedingly difficult, if not impossible. Consequently, data transmission efficiency decreases while network overhead increases. To address these challenges, opportunistic routing has emerged. Opportunistic routing is self-organizing networks that rely on chance encounters between nodes to transmit messages, and the store-and-carry forwarding technique is used for data transmission between nodes. Typically, most traditional opportunistic routing algorithms divide nodes into communities. However, such approaches often overlook node characteristics, potentially leading to disparate nodes being assigned to separate communities despite positive correlations. Thus, accurately dividing communities remains a significant challenge. To address this, we propose the Efficient Community Detection Data Transmitting (ECDDT) algorithm. During community detection, ECDDT combines social attributes and network topology using mobile edge computing. Subsequently, to expedite task execution, we schedule these tasks in a reasonable manner. Through simulation experiments across four diverse real datasets, we evaluate the algorithm’s efficacy. Our findings demonstrate favorable performance in delivery rate and routing overhead.