You are here

Human-Driven Edge Computing and Communication

Feature Topic


Aims and Scope

The vision of Edge Computing considers that tasks are not exclusively allocated on centralized Cloud platforms, but are distributed towards the edge of the network (as in the Internet-of-Things and Fog Computing paradigms), and transferred closer to the business operations via the Content Delivery Networks. The traditional gateway becomes a set-top-box machine, with additional computation and storage capabilities, where micro tasks can be offloaded first, instead of directly to the Cloud. Mobile Edge Computing can also be a more suitable approach to extract knowledge also from privacy sensitive data, which are not to be transferred to third party entities (global cloud operators) for processing. The proliferation of the networking connectivity and the progressive miniaturization of the computing devices have paved the way to the sensor networks and their success in the automation of the several monitoring & control applications. Such networks are built in an ad hoc manner and deployed in an unsupervised manner, without an a-priori design. The consequent availability of long-range communication means at certain nodes of those networks has enabled the possibility of the Internet connection of the sensor network, to make use of cloud-based services.

The new challenge addressed by this Feature Topic (FT) is how to put users in the loop so that they can retake control of their information. The massive proliferation of personal computing devices is opening new human-centered designs that blur the boundaries between man and machine.

In addition, Edge services are also used to exchange the data collected and processed within the context of the IoT towards external services and/or to visualize them through traditional browser by the users. Now, the frontier for the research on the data management is related to the so-called Edge Computation and Communication, consisting of an architecture of one or more collaborative multitude of computing nodes that are placed between the sensor networks and the cloud-based services. Such a mediating level is responsible for carrying out a substantial amount of data storage and processing to reduce the retrieval time and have more control over the data with respect to the Cloud-based services and to consume less resources and energy to reduce the workload. The interdependencies among those three different levels of storage and computing within an IoT solution are complex and determining at which data should be collocated and elaborated is demanding but not simple to handle. Such a complex situation is further exacerbated if we consider to achieve Quality-of-Service goals such as reliability, availability, security, mobility and energy efficiency, without compromising the correct behavior of the system and the service duration of the devices batteries. Moreover, the interconnection between the sensor networks and the upper level is not simple to be supported, in fact, falls within those situations where traditional Internet architectures fail to provide it effectively. This is because the sensor networks are deployed on hostile and challenging environments implying intermittent connectivity, a heterogeneous mix of nodes, frequent nodal churn, and widely varying network conditions.

The analysis of human activity and their interactions with physical and digital artefacts will also be extremely useful for closing the control loop of adaptive distributed systems. This may open a new research playground for distributed systems that adapt to user behaviours in different contexts, moving more and more to the network edge through devices such as the 5th Generation mobile networks or 5th Generation wireless systems. The second aspect of the frontier of the current research is therefore related to the application of challenging networking solutions to support the Fog Communication and Computation in the Internet of Things.

The aim of this FT is to solicit novel contributions to the current debate on realizing the Edge Computing perspective to the Cloud platforms and Internet of Things by focusing on the human-driven resource management, challenging networking aspects and communication issues, by also seeking practical experiences in using these intelligent solutions in concrete use cases.

Topics of interest include, but are not limited to:

  • Novel models and architectures of Edge-centric computing
  • Fog-to-Cloud integration and protocols
  • Communication protocols and issues
  • Crowdsensing and crowdsourcing information
  • Human-driven design and implementation of edge computing
  • Novel socially-informed architectures
  • Delay-tolerant networks, opportunistic communication and computing
  • Reliability and availability, mobility and connectivity in edge-centric computing
  • User-guided management of Fog systems and services
  • Resource management and provision
  • Data harvesting and analytics in challenged networking
  • Information centric and content-centric networking
  • Distributed storage services
  • Heterogeneity of edge systems
  • Energy-efficient communication and computation
  • Security and privacy, attacks and resiliency
  • Secure and sensitivity-aware applications
  • Novel safe methods for including humans in the data-analysis loop
  • QoS-aware communication protocols
  • Daily use applications and programming models
  • Test and simulation tools for evaluating challenged systems
  • Modelling and simulations


Florin Pop
University Politehnica of Bucharest, Romania

Aniello Castiglione
University of Salerno, Italy

Jiannong Cao
The Hong Kong Polytechnic University, Hong Kong

Giovanni Motta
Google Inc., USA

Yang Yanjiang
Huawei Singapore Research Centre, Singapore

Wanlei Zhou
Deakin University, Australia