Sensor Data Discovery for the Internet of Things

Description

The Web is evolving and incorporating sensors, smartphones, wearables, actuators, and all sort of gadgets or ‘things’. To make things more complicated, these new citizens of the Web produce huge amounts of data, and very fast. Managing such high data volume and velocity is a key requirement for Big Data Systems. A final challenge in this context is related to the high variety of data, as they come from multiple and completely heterogeneous sources.

The purpose of this project is to tackle part of this problem, by designing and implementing a federated sensor metadata system for the Internet of Things. We have developed at LSIR the GSN (Global Sensor Networks) sensor data management platform that takes care of data acquisition and processing in a distributed fashion. However this system needs to expose its metadata (e.g. what type of sensors are available, where they are located, what type of observations, etc.) to the Web in a machine-interpretable way, so that other systems can communicate in an M2M (machine-to-machine) setting. In this way the sensor data of GSN instances can be automatically discovered by data harvesters that crawl the Web.

There are different ways of achieving this goal, and we expect students to actively discuss and propose novel solutions and techniques. The data discovery module would be integrated into the open-source GSN project (http://gsn.epfl.ch/) which has been successfully deployed and used in several projects over the years.

Prerequisites

• Eagerness to engage in technical discussions and research work
• Love for coding
• The main code base is in Java, but we are also happy to use Scala for server-side development, and anything you fancy for Web UI development.

Contacts

In case of any questions, please drop us an email or come to our offices:

• Jean Paul Calbimonte Perez (BC 118)