Recently, a paper titled “Reliability-Aware and Deadline-Constrained Mobile Service Composition Over Opportunistic Networks” by Peng Qinglan, a doctoral candidate of the School, as the first author as guided by Professor Xia Yunni from the Big Data Intelligence and Service Computer Team of the School of Computer Science, was published by IEEE TRANSACTIONS ONAUTOMATION SCIENCE AND ENGINEERING (type CCF-B). Co-authors of the paper include Professor Zhou Mengchu from New Jersey Institute of Technology, an IEEE-FELLOW, Researcher Luo Xin from Chongqing Institute of Green and Intelligent Technology, CAS, and Professor Pang Shanchen from China University of Petroleum. In recent one or two years, the team led by Xia Yunni presented a number of papers with great influence at three major international conferences in the service calculation area such as (IEEE-ICWS (CCF-B type), ICSOC (CCF-B type) and IEEE SCC (CCF-C type) and other several CCF recommended high-standard conferences. Those papers are mainly about the latest progress the team has made in intelligent service composition, mobile edge computing, information service system reliability prediction, and have drawn extensive attention. Some have even been awarded the IEEE SCC Best Paper of 2018.
In recent years, the world has seen full-speed development of mobile devices (such as smart phones, tablets and wearable devices) and mobile communication technologies. In addition, the number of mobile devices is growing fast and has even exceeded that of the stationary internet hosts. Furthermore, mobile applications and services are also developing and being provided at an amazing speed, and the demand from mobile users is becoming higher and higher. More and more complex applications are implemented on mobile devices. Mobile devices are changing the way people access and process information. However, mobile devices usually have limited energy supply and computing power, so resource intensive service applications are usually uploaded to the remote cloud. However, such upload operation will lead to additional data transmission overhead and weaken the system’s response ability. Opportunistic computing is a computing paradigm based on mobile ad hoc networks. It dynamically builds opportunistic networks depending on the effective range of mobile devices. It can be used to share resources, content, services, applications and computing resources with mobile devices in mobile opportunistic social networks, thus providing a decentralized platform for the execution of distributed computing tasks. It also allows users with mobile devices to form an instant social network, so as to communicate with each other and share data objects. The main difference between opportunity computing and fog computing and edge computing is that the former depends on the end user (participant). The communication in opportunistic network is based on the establishment of direct physical connection between mobile nodes, which makes full use of the resources of each node in the network, so that the system can use low-end devices to provide services, and the cost can be reduced by service composition.
The paper proposes a new method for service selection and composition in opportunistic networks based on reliable sensing and deadline constraints. Mobile users in mobile service opportunistic networks can not only use device-to-device communications, but can also utilize the resources of nearby devices. The team fully considered the mobility of users, that is, the time-varying availability brought about by the mobility of different services during operation, and proposed the Krill-Hid algorithm to solve the optimization problems of the reliability of service and quality of composite services under the constraints of deadline, so that users could experience lower delay and higher reliability when using composite services in a mobile environment. At the same time, through the analysis of algorithm time complexity and optimization of composite service scheduling algorithm, a large number of simulation experiments based on real data sets have been carried out to verify the proposed algorithm. The results showed that the method was better than the traditional method in terms of both reliability and completion time.
The paper proposes a new method for service composition in a mobile environment, which gives priority to deadline constraints and fully considers the reliability of terminal information transmission. It ensures high reliability and low delay of transmission to the maximum extent on the basis of traditional opportunistic network architecture, and can greatly improve the quality of composite services. It is of great significance to the networking of mobile devices, vehicle network interaction, and local area network transmission in remote areas.
