Kaifei Chen

PANDAA: physical arrangement detection of networked devices through ambient-sound awareness

Future ubiquitous home environments can contain 10s or 100s of devices. Ubiquitous services running on these devices (i.e. localizing users, routing, security algorithms) will commonly require an accurate location of each device. In order to obtain these locations, existing techniques require either a manual survey, active sound sources, or estimation using wireless radios. These techniques, however, need additional hardware capabilities and are intrusive to the user. Non-intrusive, automatic localization of ubiquitous computing devices in the home has the potential to greatly facilitate device deployments. This paper presents the PANDAA system, a zero-configuration spatial localization system for networked devices based on ambient sound sensing. After initial placement of the devices, ambient sounds, such as human speech, music, foot- steps, finger snaps, hand claps, or coughs and sneezes, are used to autonomously resolve the spatial relative arrangement of devices using trigonometric bounds and successive approximation. Using only time difference of arrival measurements as a bound for successive estimations, PANDAA is able to achieve an average of 0.17 meter accuracy for device location in the meeting room deployment.

Leveraging Physical Locality to Integrate Smart Appliances in Non-Residential Buildings with Ultrasound and Bluetooth Low Energy

Smart appliances and sensors have become widely available. We are deploying them in our homes to manage the level of comfort, energy consumption or security. While such smart appliances are becoming an integral part of modern home automation systems, their integration into non-residential buildings is problematic. Indeed, smart appliance vendors rely on the assumption that the Local Area Network (LAN) guarantees locality and a single unit of use/administration. This assumption is not met in non-residential buildings, where the LAN infrastructure might cover one or several buildings, and where several organizations or functional units are co-located. Worse, directly coupling smart appliances to the Internet opens up a range of security issues as device owners have very little control over the way their smart appliances interact with external services. In order to address these problems, we propose a solution that couples the use and management of smart appliances with physical locality. Put differently, we propose that smart appliances can be accessed via smartphones, but only from the room they are located in. Our solution combines opportunistic connectivity through local Bluetooth Low Energy (BLE) with an ultrasound-based method for room level isolation. We describe and evaluate a prototype system, deployed in 25 offices and 2 common spaces of an office building. This work opens up intriguing avenues for new research focused on the representation and utilization of physical locality for decentralized building management.

Demo: Creating interactive virtual zones in physical space with magnetic-induction

In this demonstration, we present the architecture, implementation, and applications of LiveSynergy --- a system that provides reliable proximity sensing and open interactive abstractions for physical spaces and objects, to enable rich interactions between humans and their environment.

Demo Abstract: Human-in-the-loop BMS Point Matching and Metadata Labeling with Babel

The inconsistent metadata in Building Management Systems (BMS) hinders the deployment of cyber-physical applications in non-residential buildings. In this demonstration we present Babel, a continuous, human-in-the-loop and crowdsourced approach to the creation and maintenance of BMS metadata. Occupants provide physical and digital input in form of actuations (e.g., the switching of a light) and readings (e.g., the reading of the room temperature of a thermostat) to Babel. Babel then matches this input to digital points in the BMS based on value equality. We have implemented a prototype of our system in a non-residential building over the BACnet protocol. While our approach can not solve all metadata problems, this demonstration illustrates that it is able to match many relevant points in a fast and precise manner.

RushNet: practical traffic prioritization for saturated wireless sensor networks

Network traffic prioritization is gaining attention in the WSN community, as more and more features are being integrated into sensor networks. Real-world deployment experience suggests that WSN brings new challenges to existing problems, such as resource constraints, low data-rate radios, and diverse application scenarios. We present the RushNet framework that prioritizes two common traffic patterns in multi-hop sensor networks: low-priority (LP) traffic that is large-volume but delay-tolerant, and high-priority (HP) traffic that is sporadic but latency-sensitive. RushNet achieves schedule-free and coordination-free delivery differentiations with the following features. First, RushNet works with most data collection protocols to deliver LP traffic. Second, RushNet leverages transmission power difference and radio capture effect to implement on-demand HP packet delivery with low overhead. Third, RushNet proposes a retrodiction technique to help nodes minimize the overhead of recovering LP packet loss due to concurrent HP traffic. We evaluate RushNet performance with micro-benchmarks and a crowdsourced office comfort monitoring deployment. The deployment results suggest RushNet can achieve a throughput close to network capacity, and deliver 98% of the HP packets with a latency of less than four seconds.

System Architecture Directions for Post-SoC/32-bit Networked Sensors

The emergence of low-power 32-bit Systems-on-Chip (SoCs), which integrate a 32-bit MCU, radio, and flash, presents an opportunity to re-examine design points and trade-offs at all levels of the system architecture of networked sensors. To this end, we develop a post-SoC/32-bit design point called Hamilton, showing that using integrated components enables a ~

SnapLink: Fast and Accurate Vision-Based Appliance Control in Large Commercial Buildings

7 core and shifts hardware modularity to design time. We study the interaction between hardware and embedded operating systems, identifying that (1) post-SoC motes provide lower idle current (5.9 μA) than traditional 16-bit motes, (2) 32-bit MCUs are a major energy consumer (e.g., tick increases idle current >50 times), comparable to radios, and (3) thread-based concurrency is viable, requiring only 8.3 μs of context switch time. We design a system architecture, based on a tickless multithreading operating system, with cooperative/adaptive clocking, advanced sensor abstraction, and preemptive packet processing. Its efficient MCU control improves concurrency with ~30% less energy consumption. Together, these developments set the system architecture for networked sensors in a new direction.

Crowd-sourced BMS point matching and metadata maintenance with Babel

As the number and heterogeneity of appliances in smart buildings increases, identifying and controlling them becomes challenging. Existing methods face various challenges when deployed in large commercial buildings. For example, voice command assistants require users to memorize many control commands. Attaching Bluetooth dongles or QR codes to appliances introduces considerable deployment overhead. In comparison, identifying an appliance by simply pointing a smartphone camera at it and controlling the appliance using a graphical overlay interface is more intuitive. We introduce SnapLink, a responsive and accurate vision-based system for mobile appliance identification and interaction using image localization. Compared to the image retrieval approaches used in previous vision-based appliance control systems, SnapLink exploits 3D models to improve identification accuracy and reduce deployment overhead via quick video captures and a simplified labeling process. We also introduce a feature sub-sampling mechanism to achieve low latency at the scale of a commercial building. To evaluate SnapLink, we collected training videos from 39 rooms to represent the scale of a modern commercial building. It achieves a 94% successful appliance identification rate among 1526 test images of 179 appliances within 120 ms average server processing time. Furthermore, we show that SnapLink is robust to viewing angle and distance differences, illumination changes, as well as daily changes in the environment. We believe the SnapLink use case is not limited to appliance control: it has the potential to enable various new smart building applications.

MARVEL: Enabling Mobile Augmented Reality with Low Energy and Low Latency

Cyber-physical applications, deployed on top of Building Management Systems (BMS), promise energy saving and comfort improvement in non-residential buildings. Such applications are so far mainly deployed as research prototypes. The main roadblock to widespread adoption is the low quality of BMS metadata. There is indeed a mismatch between (i) the anecdotal nature of metadata for legacy BMS - they are usually initialized when the BMS is commissioned and later neglected-, and (ii) the imperious need for consistent and up-to-date metadata for supporting building analytics or personalized control systems. Such applications access sensors and actuators through BMS metadata in form of point labels. The naming of labels is however often inconsistent and incomplete. To tackle this problem, we introduce Babel, a crowd-sourced approach to the creation and maintenance of BMS metadata. In our system, occupants provide physical and digital input in form of actuations (e.g., the turning on/off a light) and readings (e.g., reading room temperature of a thermostat) to Babel. Babel then matches this input to digital points in the BMS based on value equality. We have implemented a prototype of our system in a non-residential building. While our approach can not solve all metadata problems, we show that it is able to match end-user relevant points in a fast and precise manner.

Democratizing authority in the built environment

This paper presents MARVEL, a mobile augmented reality (MAR) system which provides a notation display service with imperceptible latency (<100 ms) and low energy consumption on regular mobile devices. In contrast to conventional MAR systems, which recognize objects using image-based computations performed in the cloud, MARVEL mainly utilizes a mobile devices local inertial sensors for recognizing and tracking multiple objects, while computing local optical flow and offloading images only when necessary. We propose a system architecture which uses local inertial tracking, local optical flow, and visual tracking in the cloud synergistically. On top of that, we investigate how to minimize the overhead for image computation and offloading. We have implemented and deployed a holistic prototype system in a commercial building and evaluate MARVELs performance. The efficient use of a mobile devices capabilities lowers latency and energy consumption without sacrificing accuracy.