Editorial for Special Issue on Flexible Cloud and Edge for Internet‐of‐Things
Deze Zeng,Ruidong Li,Zhi Zhou,Ruiting Zhou,Rami Langar,Md Zakirul Alam Bhuiyan
DOI: https://doi.org/10.1002/ett.4266
IF: 3.6
2021-01-01
Transactions on Emerging Telecommunications Technologies
Abstract:With the Internet of Things (IoT), billions of devices will be interconnected to empower the future intelligent world and life, which is supported by upcoming and emerging technologies, such as 5G, network function virtualization, and artificial intelligence. Along with the communication and computation technology evolution, it is observed that the service platform should also evolve to explore low latency and high throughput capability with the support of edge computing. Due to the comparatively long distance and the resulted high latency between the centralized cloud and IoT devices, edge computing, by pushing the cloud services to the network edge, has become a promising platform to provision computation and storage resources as edge services in the proximity of the IoT devices. Thus, edge computing can well facilitate the IoT devices. While, although with lower latency, the resources on the edge computing environments are limited in comparison to cloud computing. In this case, other than treating edge computing as an alternative to cloud computing, it would be better to treat it as a complementary. Cloud, edge, and IoT devices shall be systematically organized and scheduled to provide efficient and secure IoT services. It is envisioned that there will be a large number of different IoT applications with diverse needs in the near future. Meanwhile, there will be a large number of geo-distributed edge servers scattered in the proximity of the IoT devices. This raises a natural question on how the edge resources shall be managed to cater for the diverse needs of IoT applications, together with the cooperation of cloud. Targeting at this goal, there are many remaining challenges. For example, from the perspective of resource allocation and task scheduling, it is expected that the task can be freely executed on IoT device, edge servers, or the cloud. Obviously, the resources on different devices are highly heterogeneous. Efficient algorithms are expected to address such issues from different aspects. From the perspective of security and privacy, the data generated by IoT devices need to be shuffled and processed across different domains (eg, different cloud servers, different edge servers) with different security and privacy level guarantee. How to guarantee the security and privacy of the sensitive IoT data is also a critical issue toward flexible cloudand edge-empowered IoT application development. Regarding the above challenges, we are motivated to organize such a special issue to collect the state-of-the-art concept and methodology to tackle these challenges. This special issue contains nine accepted papers from researchers relating to cloud computing, edge computing, and IoT, spanning from edge-cloud collaboration, task scheduling, memory management to energy management, and application placement, and so on. In the paper “Incentive Mechanism Design for Edge-Cloud Collaboration in Mobile Crowd Sensing” authored by Zhuo Li et al, a Pricing-based Incentive Mechanism for Edge-Cloud collaboration (PIM-EC) is proposed for mobile crowd-sensing. To achieve the goal of high energy efficiency in edge computing, the authors design a two-stage game, which includes the bargaining game between users and edge-cloud servers, as well as a data trading game among different edge-cloud servers deployed in different regions. For the two-stage game, a finite-period dynamic bargaining algorithm and an optimal auction mechanism are proposed. The data optimal price thus can be obtained using the augmented Lagrange multiplier method. The experimental results show that in PIM-EC the social welfare has been significantly improved, compared with the existing solutions. In the paper “Q-learning based Flexible Task Scheduling in a Global View for Internet-of-Things” authored by Junxiao Ge et al, the problem of effective task scheduling with resource-limited sensor nodes is studied. Unlike existing Q-learning-based task scheduling schemes, which only focus on the node angle and lead to poor performance of the whole network, a Q-learning-based flexible task scheduling with global view (QFTS-GV) scheme is proposed to improve task scheduling success rate, reduce delay, and extend the lifetime for IoT devices. The key idea of QFTS-GV is to define the state set, action set, and rewards function in a global view so as to establish a task scheduling policy with distinguishing reward. In the paper “Internet of things-fog computing-based framework for smart disaster management” authored by Abdullah Aljumah et al, a novel technique of stochastic game net (SGN) is proposed to minimize the tangible and intangible infrastructural losses due to disasters. Specifically, IoT technology is used for disaster management and control in a