Some current projects
Computation Offloading Using NFC
Near Field Communication (NFC) enables devices with a distance of less than 10 cm to exchange small amounts of data. The majority of recent Android devices already implement this functionality. Although the bandwidth of NFC is typically about 50-340 times smaller than Bluetooth and Wi-Fi, NFC’s short range and operational characteristics provide several advantages compared to Bluetooth and
Wi-Fi. In this project we propose and implement the first known Near Field Communication (NFC)-based computation offloading framework. We overcome the limitations of the default protocol; removing the need for constant user interaction, the one–way communication restraint, and the limit on low data size transfer.
A Real-Time Recommendation System for Mobile Augmented Reality Ecosystems
In this project, we propose a real-time recommendation system for Mobile Augmented Reality Ecosystems. We detect and present the most suitable virtual objects on the mobile user's screen. The selection of the proper virtual objects depends on the mobile users' context. We consider the user's context as a set of variables that can be either drawn directly by user's device or can be inferred by it or can be collected in collaboration with other mobile devices.
Interactive Visual Privacy Control
Cameras of mobile and wearable devices enable a number of applications such as augmented reality, continuous sensing, and life-logging systems. However, being recorded by unauthorized or unnoticed cameras have raised people's concerns about visual privacy. In this project, we propose a novel interactive
method to control visual privacy. We allow individuals to interact with cameras using static tags and more flexible hand gestures for privacy control. By delivering privacy control preferences via visual indicators, devices will automatically perform control operations according to privacy indicators detected and control rules.
A New Lane Change Aid System with the Help of Augmented Reality Technology
In this project, we propose a new lane change aid system with the help of augmented reality technology. By using the graphic processing technique instead of radar, our system should be able to detect the vehicle position and velocity in real time. With the help of advanced lane-changing models and vehicle detection technology, the system will provide the driver with augmented information on an augmented reality device, such as displaying a symbol or a maneuver recommendation on an augmented reality device. The system will enable the safer lane-changing experience for the driver and reduce the risk of LCE collision to a minimum level.
Ubii, Ubiquitous interface and interaction
Ubii is an integrated system that put the work of augmented reality and ubiquitous computing together. In the system smart devices are connected together over the network. We develop a hand gesture detection algorithm that track the user gestures continuously. The end result is a new human-computer paradigm that allows user to interact with devices with simple hand gestures. We develop Ubii on Google Glass, which users see the interface through an augmented reality interface. Currently users can transfer files between computers freely, print files remotely and control remote devices over a distance. Ubii presents a more convenient and intuitive natural user interface to the users. Ubii attains a seamless interaction between the physical and digital worlds.
Behavioural economics of Device to Device ecosystems
In this project we consider the collaboration between mobile devices in multiple application types. One application example is video compression offloading with the assistance of nearby devices using Wifi-direct. Furthermore, we analyse the incentives for all the participating mobile users and we propose a currency for computation offloading. Also, by examining the incentives of mobile users, we focus on their behaviour when deciding to help others and in their motives. Using this information we implement a reputation system in order to evaluate the contribution of each mobile user to the general good and treat her in an analogous way. Moreover, we research the characteristics of cryptocurrencies, such as Bitcoin, and we investigate whether and how they should be modified in order to be able to operate on top of mobile ad hoc networks.
Augmented Reality (AR) is the next big thing for mobile devices (e.g. Smartphones and AR glasses), but it requires a lot of resources. Sound/image processing is expensive and requires large knowledge bases - they cannot be done locally on a device. The objective of the CloudRider project is to develop the mobile cloud computing technology and the in-cloud big data processing algorithms to enable an AR ecosystem. CloudRider proposes a bottom up approach and aims to cover a wide landscape of system level research, identify the important research challenges in each area, and secure the AR business by producing key patents in each important aspect.
Vehicular Data Center
With sophisticated array of sensing and computation capacities, as well as growing Internet presence and unlimited power supply, the present day vehicle models become perfect candidates to act as the nodes of “data center on wheel”.
Compare to moving counterparts, parked vehicles show great advantages in storage, communication and manageability. The vehicle groups on our roadways, streets and parking lots are recognized as an abundant and underutilized computational resource that can be leveraged for the purpose of providing networking, computing or social services.
It is also expected that, once adopted and championed by municipalities and third-party infrastructure providers, parked clouds will redefine the way in which pervasive computing and its myriad applications is thought of.
Research in SDN-enabled network
This project aims to investigate in the SDN-enabled network resource management solutions in Wide Area Network (WAN) environments. Particularly, the research will focus on traffic management issues related to system architecture, management framework, and algorithm design for SDN-based Traffic Engineering (TE) to improve inter data center network performance.
UniCar: Ridesharing Recommendation System Using Human Mobility Clustering and Prediction
This project has the motivations both in the environment and the society: car usage reduction is necessary because of severe environmental problems, and human mobility is predicable by mining the rich social interactions and properties. Related systems including ridesharing recommender and taxi dispatching. For the travelers, the system can provide ridesharing candidates recommendation and waiting points recommendation; for the taxi drivers, the system can provide passenger recommendation.
MADNet: (On mobile data offloading for metropolitan area networks)
ThinkAir: Intelligentce in the Air (On using cloud for performance and energy efficiency improvement on mobile devices)
LENS: On LEveraging (anti)-social Network against Spam)
GlobalSense: (On capturing and modelling global city mobility and dynamic using online public webcam)
A Million People Project: (Planet-scale Mobility Measurement)