BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Data mining at the edge of the Internet of Things - Dr James Bruse
 y\, Reader in Pervasive Computing Director\, Cogent Labs\, Coventry Univer
 sity
DTSTART:20150219T140000Z
DTEND:20150219T150000Z
UID:TALK57924@talks.cam.ac.uk
CONTACT:Philip Woodall
DESCRIPTION:The talk examines the benefits of edge mining - data mining th
 at takes place on the wireless\, battery-powered\, and smart\, networked\,
  sensing devices that sit at the edge points of the Internet of Things. Th
 rough local data reduction and transformation\, edge mining can quantifiab
 ly reduce the number of packets that must be sent\, reducing energy usage\
 , and remote storage requirements. In addition\, edge mining has the poten
 tial to reduce the risk in personal privacy through embedding of informati
 on requirements at the sensing point\, limiting inappropriate use. The ben
 efits of edge mining are examined with respect to two specific algorithms:
  linear Spanish Inquisition Protocol (L-SIP) and Bare Necessities (BN). In
  general\, the benefits provided by edge mining are related to the predict
 ability of data streams and availability of precise information requiremen
 ts\; results show that L-SIP typically reduces packet transmission by arou
 nd 95% (20-fold) while BN reduces packet transmission by 99.98% (5000-fold
 ). Context for the above is provided through a buildings’ monitoring (en
 ergy and environment) case study. Results from deployed systems are presen
 ted and design guidelines are emerged. Sound designs and minimization of s
 oftware overheads can lead up to a 10-fold battery life extension for L-SI
 P\, in this application. Concepts presented are also extrapolated to human
  physiological/movement monitoring\, forging the path towards “forever
 ” wearable body sensor networks.
LOCATION:Seminar room 1\, Institute for Manufacturing\, Cambridge
END:VEVENT
END:VCALENDAR