IoT Analytics for Public Safety Salvatore Longo, Bin Cheng, Martin Bauer, Flavio Cirillo, Ernoe Kovacs NEC Labs Europe, Heidelberg, Germany
Outline Deployment of sensor Smart City and Public Safety OODA paradigm Challenges of IoT Analytics for Public Safety IoT Analytics Techniques and Approachs NEC examples of research Conclusions Page 2 NEC Corporation 2014
Internet Of Things Page 3 NEC Corporation 2014
Internet Of Things Ecosystem Source:http://blog.smarticlelabs.com/internet-of-things-primer Page 4 NEC Corporation 2014
Smart City and Public Safety Globally, by 2020, it is estimated the contribution of different technologies towards the development of safer cities projects to be USD 80-85 billion.[1] [1]http://uk.nec.com/en_GB/en/global/solutions/safety/pdf/Safer_Cities_WP.pdf Page 5 NEC Corporation 2014
OODA Paradigm wotlabs.net Observe Look at the current environment Orient Make theories about the situation (e.g. anomalies) Decide Act Evaluate the theories Make a decision on action Promptly apply the decision Page 6 NEC Corporation 2014
Challenges Dynamic Communication Channel (Observe, Orient, Decide, Act) among sensors, application and actuators need to be dynamic, flexible and scalable Big Data in Real Time (Observe, Orient, Decide) Data highly unstructured, highly heterogeneous, highly dimensional and distributed Control data generation frequency Pre-processing: what? where? Design parallelized real-time data stream processing Orchestrate resources in the cloud and at the edge Page 7 NEC Corporation 2014
Challenges Actionable insights with good accuracy (Orient) Results must be understandable, accurate and well-timed Privacy (Observe) Computer vision and camera systems are too invasive Citizens claim their privacy Page 8 NEC Corporation 2014
IoT Analytics Techniques Sensor Fusion The combining of sensory data or data derived from sensory data such that the resulting information is in some sense better than would be possible when these sources were used individually.[1] Challenges: Potential device error, noise, and flaws in the data gathering process Extraction of meaningful data Ascertaining what will and will not be needed is an exercise left to the experienced design engineer Approach: machine learning [1]http://www.digikey.com/en/articles/techzone/2013/apr/sensor-data-fusion-more-than-just-sensor-integration Page 9 NEC Corporation 2014
NEC example: Crowd Detection NEC approach Fuse and mine sensor data Sampling the area with carefully positioned sensors Measure human activities and correlate them to the density of crowd Privacy preserving sensor Sound, Pressure, CO2, Proximity Page 10 NEC Corporation 2014
NEC example: Mobile Operation Center (MOC) Mobile Operation Center Mobile center (usually located in truck) that enables cooperation between agency Enables dynamic and federated IoT System IoT discovery for presence information IoT Broker for data fetching and subscription Page 11 NEC Corporation 2014
Conclusions Internet of Things is giving us great potential to improve Safety Security Efficiency Paradigm OODA Challenge is to sense critical situations and act in real-time Dynamic communication channel, Big Data in Real Time, Actionable, Privacy Sensor Fusion NEC example: Crowd detection, Mobile Operation Center Future Works at NEC Laboratories Europe: Online machine learning algorithm for sensor fusion Edge computing optimization Page 12 NEC Corporation 2014
NEC brings together and integrates technology and expertise to create the ICT-enabled society of tomorrow. We collaborate closely with partners and customers around the world, orchestrating each project to ensure all its parts are fine-tuned to local needs. Every day, our innovative solutions for society contribute to greater safety, security, efficiency and equality, and enable people to live brighter lives. Page 13 NEC Corporation 2014
Page 14 NEC Corporation 2014