Title: Time series analytics for addressing the problems of too much sensor data

Presenter: A/Prof. Rachel Cardell-Oliver, CRC for Water Sensitive Cities and The University of Western Australia

Abstract:

Wireless sensor networks are becoming mainstream for real-world applications from large scale environmental networks to personal networks.  New research and commercial technologies including internet of things devices, energy harvesting, low power wireless and cloud services, offer many new challenges for researchers. A recurring problem in the design of sensor network applications is that of too much data.  For sensor nodes, recent advances such as energy harvesting place strict limits on the amount of data that can be transmitted. Time series analytics need to create representations of sensor data streams that are as accurate as possible while minimising the energy needed to transmit them.  On the end-user side, the challenge is to maximise the value of the information provided to users from their sensor data.  Time series analytics need to discover interesting patterns that summarise the data and to present these results to users in an understandable way.  This talk will summarise recent work in these areas with examples from applications projects in smart metering, public transport use, thermal performance of buildings, occupancy sensing and environmental monitoring.

BIOGRAPHY:

Rachel Cardell-Oliver is a Computer Scientist with the CRC for Water Sensitive Cities (http://watersensitivecities.org.au/) and the University of Western Australia.  She designs intelligent sensor systems that integrate data measurement using sensors; data collection with wireless communication systems; and data mining to make sense of the collected data.  Her research has contributed to the end-to-end reliability of sensor network systems through theoretical models, building and designing field applications, practical field experiments, and designing new protocols and analytics. She has designed and deployed sensor networks in many application domains, focusing on the use of technology to address environmental challenges such as reducing household water consumption, understanding water use by native Australian plants, understanding public transport use, and measuring the performance of rammed earth for sustainable housing in outback Australia.  She holds an MSc degree in distributed systems from the University of Western Australia and a PhD degree in protocol verification from the Computer Laboratory, University of Cambridge, UK.