Abstract
With the emergence of the Internet of things (IOT), smart cities have changed from concept to reality. Meanwhile, those countless IOT devices are scattered in every corner of the city which generate a mass of sensing big data and IoT services every minute. However, the computing capability of IoT devices is so constrained that IoT devices fail to process computational-intensive services. Mobile edge computing (MEC) is an emergent architecture for reinforcing the computing capabilities of IoT devices to cope with the resource-hungry IoT services. In the MEC system, IoT devices are capable of offloading part of IoT services to the cloudlet for execution. Although offloading can prominently mitigate the computing burden on IoT devices, it may result in enormous transmission cost and consuming resource of cloudlets. Therefore, it poses major challenges to how to achieve trade-offs in terms of time cost, energy cost of IOT devices and resource utilization of cloudlets. In consideration of the challenge, we devise a collaborative computation offloading approach to acquire the above trade-off in the collaboration of IoT device-cloudlet-cloud three ends. Firstly, the balancing strategies of the trade-off are obtained by leveraging the multi-objective evolutionary algorithm based on decomposition (MOEA/D). We then employ a multi-criterion decision-making method that combines the entropy weight (EW) method and technique for order preference by similarity to an ideal solution (TOPSIS), is known as (EW-TOPSIS), to acquire the optimum offloading decision in the acquired balancing strategies. Finally, extensive experiments and theoretical analysis validate the effectiveness of the proposed method.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have