Rayman Preet Singh

On hourly home peak load prediction

The Ontario electrical grid is sized to meet peak electricity load. A reduction in peak load would allow deferring large infrastructural costs of additional power plants, thereby lowering generation cost and electricity prices. Proposed solutions for peak load reduction include demand response and storage. Both these solutions require accurate prediction of a homes peak and mean load. Existing work has focused only on mean load prediction. We find that these methods exhibit high error when predicting peak load. Moreover, a homes historic peak load and occupancy is a better predictor of peak load than observable physical characteristics such as temperature and season. We explore the use of Seasonal Auto Regressive Moving Average (SARMA) for peak load prediction and find that it has 30% lower root mean square error than best known prior methods.

A cloud-based consumer-centric architecture for energy data analytics

With the advent of utility-owned smart meters and smart appliances, the amount of data generated and collected about consumer energy consumption has rapidly increased. Energy usage data is of immense practical use for consumers for audits, analytics, and automation. Currently, utility companies collect, use, share, and discard usage data at their discretion, with no input from consumers. In many cases, consumers do not even have access to their own data. Moreover, consumers do not have the ability to extract actionable intelligence from their usage data using analytic algorithms of their own choosing: at best they are limited to the analysis chosen for them by their utility. We address these issues by designing and implementing a cloud-based architecture that provides consumers with fast access and fine-grained control over their usage data, as well as the ability to analyse this data with algorithms of their choosing, including third party applications that analyse that data in a privacy preserving fashion. We explain why a cloud-based solution is required, describe our prototype implementation, and report on some example applications we have implemented that demonstrate personal data ownership, control, and analytics.

Lab of things: a platform for conducting studies with connected devices in multiple homes

Researchers who develop new home technologies using connected devices often want to conduct large-scale field studies in homes to evaluate their technology, but conducting such studies today is extremely challenging. Inspired by the success of PlanetLab, which enabled development and evaluation of global network services, we are developing a shared infrastructure for home environments, called Lab of Things. Our goal is to substantially lower the barrier to developing and evaluating new technologies for the home environment.

Information dissemination in VANETs using zone based forwarding

Several VANET-based safety applications require information dissemination to all vehicles within a certain area for a certain time. Among the existing information dissemination protocols for VANETs, only Stored Geocast provides retention of information within a pre-determined area of the road for a duration of time. However, the overhead incurred by it is large. In this paper, we propose and evaluate ZBF, a zone based forwarding scheme for information dissemination which provides both a spatial and temporal retention of information and incurs less overhead than Stored Geocast.

Traffic congestion estimation in vANETs and its application to information dissemination

Traffic congestion estimation in vehicular ad hoc networks can have many interesting applications. In this paper, we propose two measures of traffic congestion around a vehicle suited for different applications. A scheme for a vehicle to estimate congestion around it is proposed and evaluated through simulations. We also show an application of the estimated congestion for information dissemination to achieve high coverage while sending much less number of redundant messages compared to a flooding based scheme.

IP address multiplexing for VEEs

The number of publicly accessible virtual execution environments (VEEs) has been growing steadily in the past few years. To be accessible by clients, such VEEs need either a public IPv4 or a public IPv6 address. However, the pool of available public IPv4 addresses is nearly depleted and the low rate of adoption of IPv6 precludes its use. Therefore, what is needed is a way to share precious IPv4 public addresses among a large pool of VEEs. Our insight is that if an IP address is assigned at the time of a client DNS request for the VEEs name, it is possible to share a single public IP address amongst a set of VEEs whose workloads are not network intensive, such as those hosting personal servers or performing data analytics. We investigate several approaches to multiplexing a pool of global IP addresses among a large number of VEEs, and design a system that overcomes the limitations of current approaches. We perform a qualitative and quantitative comparison of these solutions. We find that upon receiving a DNS request from a client, our solution has a latency as low as 1 ms to allocate a public IP address to a VEE, while keeping the size of the required IP address pool close to the minimum possible.

TussleOS: managing privacy versus functionality trade-offs on IoT devices

Networked sensors and actuators are increasingly permeating our computing devices, and provide a variety of functions for Internet of Things (IoT) devices and applications. However, this sensor data can also be used by applications to extract private information about users. Applications and users are thus in a tussle over access to private data. Tussles occur in operating systems when stakeholders with competing interests try to access shared resources such as sensor data, CPU time, or network bandwidth. Unfortunately, existing operating systems lack a principled approach for identifying, tracking, and resolving such tussles. Moreover, users typically have little control over how tussles are resolved. Controls for sensor data tussles, for example, often fail to address trade-offs between functionality and privacy. Therefore, we propose a framework to explicitly recognize and manage tussles. Using sensor data as an example resource, we investigate the design of mechanisms for detecting and resolving privacy tussles in a cyber-physical system, enabling privacy and functionality to be negotiated between users and applications. In doing so, we identify shortcomings of existing research and present directions for future work.

HomeLab: a platform for conducting experiments with connected devices in the home

The downward spiral in the cost of connected devices and sensors (e.g., cameras, motion sensors, remote controlled light switches) has generated a vast amount of interest towards using them in the home environments. Companies and researchers are developing technologies that employ these devices in a diverse range of ways. These include improving energy efficiency, increasing comfort and convenience through automation, implementing security and monitoring, and providing in-home healthcare. However, conducting experimental work in this domain is extremely challenging today. Evaluating the effectiveness of research prototypes typically requires some form of deployment in real homes. This task is riddled with not only social and legal constraints, but also logistical and technical hurdles. Examples include recruiting participants, hardware and software setup in the home, training participants and residents who typically possess varying levels of technical expertise, and diverse security and privacy concerns. Because of these challenges, individual research groups rarely manage to deploy their prototypes on more than a dozen or so homes concentrated in their geographic area. Such deployments tend to lack the scale and diversity that is needed to confidently answer the research hypothesis. Our goal is to lower the barrier towards deploying experimental technology in a large number of geographically distributed homes.