Title: Programming devices and services in P
Abstract: P is a programming framework for design, implementation, and validation of event-driven asynchronous systems. The P language incorporates deep modeling and specification techniques into asynchronous programming. It allows the programmer to systematically test and debug their applications before deployment, thus preventing Heisenbugs that are extremely difficult to find and fix later.
P is used in Microsoft products. The USB drivers shipped by Microsoft (Windows 8.1 onwards) have been written in P; these drivers run on hundreds of millions of devices. The design of P has also been implemented independently by engineers in Microsoft Office and Azure for components being written by their team. In Microsoft Azure, there are ongoing projects that are using P to implement services. Finally, researchers at Microsoft and UC Berkeley are exploiting P to build a reliable software stack for autonomous robots.
My talk will provide an overview of the key ideas behind P and conclude with a discussion of open research problems.
Bio: Shaz Qadeer is a Principal Researcher at Microsoft. His research interest lie at the intersection of (in)formal methods, program verification, programming languages, and software engineering. Currently, he spends most of his time on the P and P#, available open-source at https://github.com/p-org.
Title: An Experimental Study on Energy Efficiency in the Industrial Internet of Things
Abstract: The Industrial Internet of Things (IIoT) is the use of Internet of Things (IoT) technologies in manufacturing. The driving philosophy behind the IIoT is that smart machines are better than humans at accurately, consistently capturing and communicating data. In manufacturing specifically, IIoT holds great potential for quality control, sustainable and green practices, supply chain traceability and overall supply chain efficiency. IIoT systems could suffer from high and unbalanced energy consumption due to the nature of the network deployment. Such behavior is undesirable as it not only increases the carbon footprint of the plant, but also makes the planned maintenance of IoT devices for battery replacement a huge challenge.
In this work, we propose a heuristic and opportunistic link selection algorithm, HOLA, which not only reduces the overall energy consumption of the IIoT network but also balances it across the network. HOLA achieves the energy efficiency by opportunistically off-loading IoT device data to smart devices being carried by the workforce in the factory settings. Further, these smart devices with multiple radio links such as Bluetooth, Wi-Fi, and 3G/4G LTE heuristically determine the best link to
transmit the data to the Cloud based on the quality and energy cost of the link. Our experimental and simulation studies validate that HOLA can improve the energy efficiency of IIoT systems by reducing the overall energy consumption and balancing it across the network.
Dr. Rajeev Shorey is the Principal Scientist at the TCS Innovation Lab, Cincinnati, USA and Bangalore, India. Dr. Shorey received his Ph.D and MS in Electrical Communication Engineering from the Indian Institute of Science (IISc), Bangalore, India in 1997 and 1991 respectively. He received his B.E degree in Computer Science and Engineering from IISc, Bangalore in 1987. Dr. Shorey’s career spans several reputed research labs – General Motors (GM) India Science Laboratory (ISL), IBM India Research Laboratory and SASKEN Technologies. He was an adjunct faculty in the Computer Science Dept at IIT, Delhi from 1998 to 2005. He was a faculty in the Computer Science Dept at the National University of Singapore from 2003 to 2004, while on leave from IBM Research Labs in New Delhi.
Title: Jana Care: Building cost-effective and massively scalable solutions for diabetes
Abstract: The combination of obesity and diabetes is a rapidly growing epidemic all across the world. It is estimated that there are 400 million diabetics globally. Jana Care, which operates out of its Bangalore and Boston offices, has as its goal to fight the global spread of diabetes using cost-effective and scalable technologies.
The problem of diabetes is two-fold: only half of diabetics are diagnosed, while only a tenth of them meet their diabetes therapy goals. Jana Care addresses the first problem with their Aina Device – a phone connected blood diagnostic device that can perform among others HbA1c and blood glucose tests using only fingerstick samples, at a fraction of the cost of traditional laboratory equipment.
Secondly, in order to help diabetics effectively manage their condition after they are diagnosed, Jana Care has developed the Habits Program – a mobile application that drives lifestyle improvement and better diabetes management by using interactive educational modules as well as personalized coaching support based on the patients’ data. The Habits Program is based on the landmark Diabetes Prevention Program and Look Ahead trials.
This talk will cover Jana Care’s journey in developing products with a global market in mind.
Bio: Michal Depa is the co-founder and CTO of Jana Care, a health technology startup that has built a mobile phone connected diagnostic platform for blood tests including HbA1c, glucose and lipids, as well as integrated clinically validated coaching programs for diabetes.
Previously, Michal was a researcher at MIT’s Computer Science and Artificial Intelligence Lab, where he worked on image analysis algorithms for medical imaging applications. He holds a MS degree from MIT in computer science and a BS from McGill University in electrical engineering.
Title: Networking Technologies for the Internet of Things
Abstract: IoT relies on a low power, inexpensive, secure, stable, and sometimes long-range connectivity from the device to the cloud. In this talk I will explain the various networking technologies that are being used for various IoT scenarios, from personal devices, to devices in the home, to building, cities, and long range connectivity. I will also discuss the pros and cons of these technologies, and the open research problems.
Title: FarmBeats: IoT for Agriculture
Abstract: Food requirements are expected to double by 2050 to meet the demands of the world population, but the amount of land fit for agriculture is shrinking. Data-driven techniques, such as precision agriculture, could help meet the increased demand. In this talk, I will present FarmBeats, an agricultural IoT system that uses a combination of unmanned aerial vehicles (UAVs) and wireless sensors to enable data-driven agricultural techniques.
Title: Achieving household energy breakdown at scale
Abstract: Energy breakdown: the process of breaking down a home’s energy consumption into constituent appliances, is considered an important step towards energy saving in the residential setting. Most current energy breakdown techniques exploit IoT hardware to be installed in each home and are thus not scalable. This is specifically true in the Indian context where penetration of smart meters is close to zero. In this paper, I will first discuss how data at different temporal and device resolution can address different analytical approaches to achieve energy efficiency. I will then discuss in detail our recent research work wherein we propose an approach that does not require any additional hardware installation and can be scaled across a large number of homes to provide accurate monthly per-appliance energy consumption feedback. Our approach exploits the already available datasets with per-appliance energy consumption for a small set of homes through direct instrumentation. It intelligently transforms the energy consumption from such small set of homes even to regions with different weather conditions, thus further eliminating the need to instrument homes across all different types of regions. We demonstrate how our approach can easily scale to provide per appliance energy consumption feedback to all homes in India.
Title: Zenatix and the art of IoT Driven Energy Efficiency
Abstract: In this talk I will provide an overview of how Zenatix exploits recent advancements in IoT to provide analytics that can drive significant energy control and reduction for commercial establishments. I will touch upon some of the analytical outcomes that drive energy savings for our customers. I will also discuss how the overall product offering from Zenatix evolved over time to where we are today.
Bio: Dr. Amarjeet Singh received his Ph.D. in Electrical Engineering at University of California, Los Angeles (UCLA) in 2009. As part of his Ph.D. thesis work, he developed robotic sensing systems for monitoring lakes and rivers ecosystems. He collaborated with several biologists and ecologists for multiple real world deployments involving pollution monitoring in rivers and lakes. Amarjeet joined IIIT Delhi as Asst. Professor in 2009. At IIIT Delhi, he has been involved in several IoT and data analytics projects in the domain of energy and healthcare, with grants worth several crores from both industry and government. He has published in diverse venues of repute in the domains of sensor networks, robotics, data analytics, artificial intelligence and environmental science. He is on sabbatical from IIIT Delhi since January 2014 to work on his startup called Zenatix (www.zenatix.com). With Zenatix, Amarjeet aims to scale up his research outcomes in energy analytics for maximal impact, specifically targeting energy savings in commercial buildings. Zenatix has already deployed more than 300 systems and is targeting to deploy more than 1000 by end of 2016. He did his B.Tech. from IIT Delhi in 2002 and MS from UCLA in 2007. He was awarded Outstanding MS student in UCLA School of Engineering in 2007 and Chorafas Foundation Award for research with long term impact.
Title: Ubiquitous Cloud-based Cardiac Diagnostics – Timely, Accurate, Affordable Heart Care
Abstract: Tricog Health has created a cardiac diagnostic service that includes a cloud-connected ECG machine and a cloud-based interpretation system. The ECG machine is clinical-grade and uses the latest in IoT technology for high-quality real-time signal acquisition. The interpretation system uses a combination of machine learning and signal processing, reinforced by human verification for accurate analysis. Our service is designed to be low cost and scalable for wide-spread accessibility and reports in 5 min or less. This talk will discuss the the various sociological and engineering challenges that Tricog has overcome to build the platform that has been used by over 70,000 people and which has helped over 4,000 heart attack patients get early detection and quick treatment.
Title: Enabling Participatory Urbanism for Real-time Air Quality Monitoring using a Fast Time-Series IoT architecture
Abstract: India has the dubious distinction of having 10 out of the 20 most polluted cities in the world. The number of government air quality monitoring stations in these cities, baring New Delhi (which has 12), are between 1 or 2 stations for a population of 1 to 2 million people. This is woefully inadequate. A new wave of Low-cost Particulate Matter sensors (PM2.5 & PM10 being the most harmful of the air pollutants) enabled with IoT technologies is empowering citizens & communities to participate in their own governance & policy making. IndiaSpend has built India’s first Independent, Real-time Air Quality Monitoring Network using indigenously built Low-cost IoT devices. The project was launched in December 2015 and is currently active in 6 cities of India (with 10 more cities coming online in the near future). The talk will cover lessons learned from the IoT architecture perspective and the current Fast Time-Series based APIs developed which has enabled a significant enhancement in the accuracy & scalability of the air quality data analysis.
Title: Planetary Scale Swarm Sensing, Planning and Control for Weather Prediction
Abstract: Weather forecasting is a canonical predictive challenge that relies on extensive data gathering operations. We explore new directions with forecasting weather as a data-intensive challenge that involves large-scale sensing of the required information via planning and control of a swarm of aerial vehicles. First, we will demonstrate how commercial aircraft can be used to sense the current weather conditions at a continental scale and help us create Bayesian deep-hybrid predictive model for weather forecasts. Beyond making predictions, these probabilistic models can provide the guidance of sensing with value-of-information analyses, where we consider uncertainties and needs of sets of routes and maximize information value in light of the costs of acquiring data from a swarm of sensors. The methods can be used to select ideal subsets of locations to sense and also to evaluate the value of trajectories of flights for sensing. Finally, we will discuss how to carry out such large sensing missions using novel algorithms for robot planning under uncertainty.
Bio: Ashish Kapoor is a senior researcher at Microsoft Research, Redmond. His recent research focuses on machine learning with applications to controls and planning of aerial vehicles. In the past he has worked in many different areas that include quantum machine learning, computer vision, affective computing and human-computer-interaction. Ashish received a PhD at the MIT Media Lab in 2006 and prior to that graduated from Indian Institute of Technology, Delhi.
Title: Connected Vehicles: Opportunities in Fleet Sensing, Safety, and Orchestration
Abstract: Connected vehicles create plentiful opportunities to improve comfort, efficiency, and safety through fleet sensing, safety messaging, and orchestration. This talk will discuss current trends in vehicle connectivity and present examples of traffic, location, and parking sensing techniques that are enabled by this connectivity. It will then review the effort to enable direct communications between vehicles through Dedicated Short Range Communications and the associated networking challenges, particularly in the area of congestion control. Including other vulnerable road users, such as pedestrians, creates additional challenges and is likely to require context sensing on pedestrians’ devices. One example is detecting stepping from sidewalks into streets. I will discuss inertial and camera-based sensing techniques to improve such detection and how learning from crowds could help mobile devices interact synergistically with future connected and automated vehicles to improve safety.
Title: Location Privacy for Connected Vehicles
Abstract: This talk will provide an overview of the location privacy problem in the context of connected vehicles. Starting with a tutorial on fundamental information and location privacy concepts, the talk will review privacy-enhancing techniques and their applications in connected vehicles. These privacy-enhancing techniques include depersonalizing application-layer location data through spatial cloaking and path cloaking techniques. They also include disposable identifiers at lower layers of the wireless stack. I will also describe how we incorporated such ideas, as a privacy-by-design case study, in a smartphone-based automotive traffic monitoring system that has been trialled with hundreds of users in the Bay Area, and reflect on industry adoption. In the latter part of this talk, I will discuss recent results on the identifiability of persons from in-vehicle data.
Title: Experiments in Indoor Localization and Vehicle Classification
Abstract: Sensing, Computing and Communication are essential components of an IoT system. The talk will cover two experiments in indoor localization for harsh environments and vehicle classification for road traffic monitoring. Details about the system specifications, approach to designing the system under constraints, theoretical modelling, and implementation of algorithms on a resource-constrained platform will be presented.