Human-Enabled Edge Computing: Exploiting the Crowd as a Dynamic Extension of Mobile Edge Computing

The MEC vision leverages the availability of powerful and low-cost middleboxes, statically deployed at suitable edges of the network and acting as local proxies for the centralized cloud backbone; this potentially enables, among other things, better scalability and better reactivity in the interaction with mobile nodes via local control decisions and actuation. MEC has already been proposed as an enabler for several Internet of Things and cyber-physical systems application scenarios, and also mutual benefits due to the integration of MEC and mobile crowdsensing (MCS). The article originally proposes human-driven edge computing (HEC) as a new model to ease the provisioning and to extend the coverage of traditional MEC solutions. From a methodological perspective, we show how it is possible to exploit MCS i) to support the effective deployment of fixed MEC (FMEC) proxies and ii) to further extend their coverage through the introduction of impromptu and human-enabled mobile MEC (M2EC) proxies. In addition, we describe how we have implemented these novel concepts in the MCS ParticipAct platform through the integration of the MEC Elijah platform in the ParticipAct living lab, an ongoing MCS real-world experiment that involved about 170 students at the University of Bologna for more than two years. Reported experimental results quantitatively show the effectiveness of the proposed techniques in elastically scaling the load at edge nodes according to runtime provisioning needs.