Research

Ad-hoc Crowd Computing

Nowadays, the increasing popularity of smart devices – such as smartphones and tablets – is changing our daily life. Two noteworthy changes are: the increment of mobile traffic, due to unlimited connectivity of smart devices using diverse wireless network technologies; and the popularization of social network service (SNS) – also called social media – which results in the creation of social games, social commerce applications, among others. In these new environments, smart devices are a valuable tool to help users to achieve their goals. Since they are omnipresent, they can learn from users’ action and understand their intentions and provide proper services to support these actions.

 

Spontaneous Service Provision Framework

Future environments combined with "cyber world" and "physical things"

Future environments combined with “cyber world” and “physical things”

 

We define Active Surroundings as a pervasive environment where entities (devices or services) actively response to user actions or help users to perform their jobs without intrusion to the users. An active surrounding is inherently composed of heterogeneous devices. Our middleware infrastructure for Active Surroundings aims to support community-centered pervasive environment where system minimizes user distraction or does not intrude users in the environment. The middleware infrastructure is composed of five components such as environment sensing, context management, dynamic adaptive framework, adaptive service discovery, and dynamic adaptive network, to enable users to be free from intrusion by system or devices. Services need to be reconfigured in ubiquitous computing environments because what a person wants and the resources required to accomplish this changes dynamically. A person’s needs change all the time, and not only do people move around a great number of places in their daily lives, but even the environment in a single place can change greatly in a short amount of time. Service reconfiguration considers these changes and provides users with high-quality service by composing the diverse computing services that change dynamically.

 

Urban Computing Research Center: http://ucrc.kaist.ac.kr/

 

Semantic Virtualization for IoT Edge Cloud

 

overview
It is well expected that a collection of smart objects such as IoT devices dynamically form an edge cloud allowing acquiring, storing, communicating, and processing of information done at the edge of the network. Edge cloud workers (i.e. IoT devices) usually have service specific characteristics while legacy cloud workers are more computing oriented. However, existing cloudlet architectures continue to take their allocation policy centered around computation resource virtualization without considering such characteristics of edge IoT clouds. In order to provide service specific QoS driven virtualization of an edge IoT cloud, which we call semantic virtualization, a new approach is required.