SENSYS 2018
SESSION
Paper Name Author(s)
E-Eye: Hidden Electronics Recognition Through MmWave Nonlinear Effects

the use of e-devices for malicious attacks has increased. the authors propose to develop a low-cost and practical hidden e-device recognition technique to enable efficient screenings for threats of hidden electronic devices in daily life. E-Eye comprises a low-cost (i.e., under $100), portable (i.e., 11.8cm by 4.5cm by 1.8cm) and lightweight (i.e., 45.5g) 24GHz mmWave probe and a smartphone-based e-device recognizer.To validate the E-Eye performance, we conduct experiments with 46 commodity electronic devices under 39 distinct categories.Results show that E-Eye can recognize hidden electronic devices in parcels/boxes with an accuracy of more than 99% and has an equal error rate (EER) approaching 0.44% under a controlled lab setup.

Zhengxiong Li, Zhuolin Yang, Chen Song, Changzhi Li, Zhengyu Peng, Wenyao Xu
MARVEL: Enabling Mobile Augmented Reality With Low Energy And Low Latency

This paper presents MARVEL, a mobile augmented reality (MAR) system which provides a notation display service with imperceptible latency and low energy consumption on regular mobile devices.

Kaifei Chen, Tong Li, Hyung-Sin Kim, David Culler, Randy Katz
Exploiting WiFi Guard Band For Safguarded ZigBee

Cross-technology interference (CTI) from dense and prevalent wireless has become a primary threat to low-power IoT.This paper presents G-Bee, a CTI avoidance technique that uniquely places ZigBee packet on the guard band of ongoing WiFi trafic. Another exclusive feature of G-Bee is spectrum-synchronized low duty cycling - by utilizing guard bands of periodic WiFi beacons, active slots are efectively synchronized to spectrum availability (i.e., guard band) for significant delay improvement.Extensive evaluations on our prototype system demonstrates G-Bee PRR over 95% where legacy ZigBee drops to below 15% under significant interference with hundreds WiFi users and reduction of low duty cycle delay by 87.5%, all of which are achieved with a light computational overhead of 0.3%.

Yoon Chae, Shuai Wang, Song Min Kim
Continuous Low-Power Ammonia Monitoring Using Long Short-Term Memory Neural Networks

this paper studies accurate and continuous ammonia monitoring. the authors propose a new ammonia monitoring approach that is low-power, automatic, accurate, and wireless. The prediction model is built on long short-term memory (LSTM) neural networks.We built 38 prototype sensors and a home-grown gas flow system.In a 3-month in-lab testing period, we conducted extensive experiments and collected 13,770 measurements.Our model accurately predicts the equilibrium state resistance value, with an average error rate of 0.12%.

Zhenhua Jia, Xinmeng Lyu, Wuyang Zhang, Richard P. Martin, Richard E. Howard, Yanyong Zhang
InK: Reactive Kernel For Tiny Batteryless Sensors

Tiny energy harvesting battery-free devices promise maintenance free operation for decades, providing swarm scale intelligence in applications from healthcare to building monitoring.These devices operate intermittently because of unpredictable, dynamic energy harvesting environments, failing when energy is scarce.Despite this dynamic operation, current programming models are static; they ignore the event-driven and time-sensitive nature of sensing applications, focusing only on preserving forward progress while maintaining performance.This paper proposes InK; the first reactive kernel that provides a novel way to program these tiny energy harvesting devices that focuses on their main application of event-driven sensing.InK brings an event-driven paradigm shift for batteryless applications, introducing building blocks and abstractions that enable reacting to changes in available energy and variations in sensing data, alongside task scheduling, while maintaining a consistent memory and sense of time.

Kasım Sinan Yıldırım, Amjad Yousef Majid, Dimitris Patoukas, Koen Schaper, Przemysław Pawełczak, Josiah David Hester
Automatic Unusual Driving Event Identification For Dependable Self-Driving

This paper introduces techniques to automatically detect driving corner cases from dashcam video and inertial sensors. We evaluate the system based on more than 120 hours real road driving data.It shows 82% accuracy improvement versus strawman solutions for sudden reaction detection and above 71% accuracy for rare visual views identification.

Hongyu Li, Hairong Wang, Luyang Liu, Marco Gruteser
System Architecture Directions For Post-SoC/32-bit Networked Sensors

The emergence of low-power 32-bit Systems-on-Chip (SoCs) presents an opportunity to re-examine design points and trade-ofs at all levels of the system architecture of networked sensors.the authors develop a post-SoC/32-bit design point called Hamilton, showing that using integrated components enables a ∼$7 core and shifts hardware modularity to design time. We design a system architecture, based on a tickless multithreading operating system, with cooperative/adaptive clocking, advanced sensor abstraction, and preemptive packet processing.

Hyung-Sin Kim, Michael P Andersen, Kaifei Chen, Sam Kumar, William J. Zhao, Kevin Ma, David E. Culler
FastDeepIoT: Towards Understanding And Optimizing Neural Network Execution Time On Mobile And Embedded Devices

the authors propose a novel framework, called FastDeepIoT, that uncovers the non-linear relation between neural network structure and execution time, then exploits that understanding to find network configurations that significantly improve the trade-of between execution time and accuracy on mobile and embedded devices. We evaluate FastDeepIoT using three diferent sensing-related tasks on two mobile devices: Nexus 5 and Galaxy Nexus.

Shuochao Yao, Yiran Zhao, Huajie Shao, Shengzhong Liu, Dongxin Liu, Lu Su, Tarek Abdelzaher
ShieldScatter: Improving IoT Security With Backscatter Assistance

the authors present ShieldScatter, a lightweight system that attaches battery-free backscatter tags to single-antenna devices to shield the system from active attacks.

Zhiqing Luo, Wei Wang, Jun Qu, Tao Jiang, Qian Zhang
CapeVM: A Safe And Fast Virtual Machine For Resource-Constrained Internet-of-Things Devices

This paper presents CapeVM, a sensor node virtual machine aimed at delivering both high performance and a sandboxed execution environment that ensures malicious code cannot corrupt the VM's internal state or perform actions not allowed by the VM.

Niels Reijers, Chi-Sheng Shih
Sentio: Driver-in-the-Loop Forward Collision Warning Using Multisample Reinforcement Learning

Thanks to the adoption of more sensors in the automotive industry, context-aware Advanced Driver Assistance Systems (ADAS) become possible.On one side, a common thread in ADAS applications is to focus entirely on the context of the vehicle and its surrounding vehicles leaving the human (driver) context out of consideration.In this paper, we propose Sentio1; a Reinforcement Learning based algorithm to enhance the Forward Collision Warning (FCW) system leading to Driver-in-the-Loop FCW system.Since the human driving preference is unknown a priori, varies between diferent drivers, and moreover, varies across time for the same driver, the proposed Sentio algorithm needs to take into account all these variabilities which are not handled by the standard reinforcement learning algorithms. Our evaluation, across distracted human drivers, shows a significant enhancement in driver experience-compared to standard FCW systems-reflected by an increase in the driver safety by 94.28%, an improvement in the driving experience by 20.97%, a decrease in the false negatives from 55.90% down to 3.26%, while adding less than 130 ms runtime execution overhead.

Salma Elmalaki, Huey-Ru Tsai, Mani Srivastava
SALMA: UWB-based Single-Anchor Localization System Using Multipath Assistance

In this paper, the authors present SALMA, a novel low-cost UWB-based indoor localization system that makes use of only one anchor and that does neither require prior calibration nor training. We further study the performance of SALMA in the presence of obstructed line-of-sight conditions, moving objects and furniture, as well as in highly dynamic environments with several people moving around, showing that the system can sustain decimeter-level accuracy with a worst-case average error below 34 cm.An experimental evaluation in an ofice environment with clear line-ofsight has shown that 90% of the position estimates obtained using SALMA exhibit less than 20 cm error, with a median below 8 cm.

Bernhard Großwindhager, Michael Rath, Josef Kulmer, Mustafa Bakr, Carlo Alberto Boano, Klaus Witrisal, Kay Römer
Fabric As A Sensor: Towards Unobtrusive Sensing Of Human Behavior With Triboelectric Textiles

existing textile-based sensing techniques rely on tight-fitting garments to obtain sufficient signal to noise, making it uncomfortable to wear and limiting the technology to niche applications. this paper's solution leverages functionalized fabric to measure the triboelectric charges induced by folding and compression of the textile itself. Our design uses a simple-tomanufacture layered architecture that can be incorporated into any conventional, loosely worn textile.

Ali Kiaghadi, Morgan Baima, Jeremy Gummeson, Trisha Andrew, Deepak Ganesan
PassiveZigBee: Enabling ZigBee Transmissions Using WiFi

In IoT networks, to demonstrate further ultralow power consumption, the authors introduce Passive-ZigBee that demonstrates we can transform an existing productive WiFi signal into a ZigBee packet for a CoTS low-power consumption receiver while consuming 1,440 times lower power compared to traditional ZigBee. We built a hardware prototype and implement these devices on a commodity ZigBee, WiFi, and an FPGA platform.

Yan Li, Zicheng Chi, Xin Liu, Ting Zhu
EAR: Exploiting Uncontrollable Ambient RF Signals In Heterogeneous Networks For Gesture Recognition

In this paper, the authors explore how to leverage the ambient wireless trafic that i) generated by uncontrollable IoT devices and ii) sensed by ambient noise floor measurements for human gesture recognition. Specifically, we introduce our system EAR, which can conduct ifne-grained human gesture recognition using coarse-grained measurements (i.e., noise floor) of ambient RF signals generated from uncontrollable signal sources.We conducted extensive evaluations in both residential and academic buildings.Experimental results show that although EAR uses coarse-grained noise floor measurements to sense the uncontrollable signal sources, the signal sources can be distinguished with an accuracy up to 99.76%.

Zicheng Chi, Yao Yao, Tiantian Xie, Xin Liu, Zhichuan Huang, Wei Wang, Ting Zhu
UbiTap: Leveraging Acoustic Dispersion For Ubiquitous Touch Interface On Solid Surfaces

this paper studies ubiquitous computing interfaces.the authors propose UbiTap, an input method that turns solid surfaces into a touch input space, through the use of sound.

Hyosu Kim, Anish Byanjankar, Yunxin Liu, Yuanchao Shu, Insik Shin
Efficient Many-to-All Broadcasting In Dynamic Wireless Mesh Networks

This paper presents Mixer, a many-to-all broadcast primitive for dynamic wireless mesh networks. Mixer integrates random linear network coding (RLNC) with synchronous transmissions and approaches the order-optimal scaling in the number of messages to be exchanged.

Carsten Herrmann, Fabian Mager, Marco Zimmerling
EXIMIUS: A Measurement Framework For Explicit And Implicit Urban Traffic Sensing

the existing tra c sensing approaches can be classi ed into two categories, i.e., explicit and implicit sensing. In this paper, the authors design a measurement framework called EXIMIUS for a large-scale data-driven study to investigate the strengths and weaknesses of these two sensing approaches In this paper, we design a measurement framework called EXIMIUS for a large-scale data-driven study to investigate the strengths and weaknesses of these two sensing approaches by using two particular systems for tra c sensing as concrete examples, i.e., a vehicular system as a crowdsourcing-based explicit sensing and a cellular system as an infrastructure-based implicit sensing.In our investigation, we utilize TB-level data from two systems: (i) vehicle GPS data from 3 thousand private cars and 2 thousand commercial vehicles, (ii) cellular signaling data from 3 million cellphone users, from the Chinese city Hefei.

Zhou Qin, Zhihan Fang, Yunhuai Liu, Chang Tan, Wei Chang, Desheng Zhang
3D Localization For Sub-Centimeter Sized Devices

The vision of tracking small IoT devices runs into the reality of localization technologies.the authors present the first localization system that consumes microwatts of power at a mobile device and can be localized across multiple rooms in settings like homes and hospitals. We build sub-centimeter sized prototypes which consume 93 µ W and could last five to ten years on button cell batteries.We achieved ranges of up to 60 m away from the AP and accuracies of 2, 12, 50 and 145 cm at 1, 5, 30 and 60 m respectively.To demonstrate the potential of our design, we deploy it in two real-world scenarios: five homes in a metropolitan area and the surgery wing of a hospital in patient pre-op and post-op rooms as well as storage facilities.

Rajalakshmi Nandakumar, Vikram Iyer, Shyamnath Gollakota
Accurate 3D Localization For 60 GHz Networks

In this paper, the authors focus on millimeter-wave wireless networks

Ioannis Pefkianakis, Kyu-Han Kim
Fall-Curves: A Novel Primitive For IoT Fault Detection And Isolation

this paper studies detection of faulty servers in IoT. it presents a novel primitive called the Fall-curve - a sensor's voltage response when the power is turned of - that can be used to characterize sensor faults.

Tusher Chakraborty, Akshay Nambi, Ranveer Chandra, Rahul Sharma, Manohar Swaminathan, Jonathan Appavoo, Zerina Kapetanovic
CapBand: Battery-free Successive Capacitance Sensing Wristband For Hand Gesture Recognition

the authors present CapBand, a battery-free hand gesture recognition wearable in the form of a wristband. We present CapBand, a battery-free hand gesture recognition wearable in the form of a wristband.We present successive capacitance sensing, an ultra-low power sensing technique, to capture small skin deformations due to muscle and tendon movements on the user’s wrist, which corresponds to speci￿c groups of wrist muscles representing the gestures being performed.We build a wrist muscles-to-gesture model, based on which we develop a hand gesture classi￿cation method using both motion and static features.We prototype CapBand with a custom-designed capacitance sensor array on two ￿exible circuits driven by a custom-built electronic board, a heterogeneous material-made, deformable silicone band, and a custom-built energy harvesting and management module.Evaluations on 20 subjects show 95.0% accuracy of gesture recognition when recognizing 15 di￿erent hand gestures and 95.3% accuracy of on-wrist localization.

Hoang Truong, Shuo Zhang, Ufuk Muncuk, Phuc Nguyen, Nam Bui, Anh Nguyen, Qin Lv, Kaushik Chowdhury, Thang Dinh, Tam Vu
Hidebehind: Enjoy Voice Input With Voiceprint Unclonability And Anonymity

To address privacy issues in speech recognition, the authors propose to add an intermediary between users and the cloud, named VoiceMask, to anonymize speech data before sending it to the cloud for speech recognition.

Jianwei Qian, Haohua Du, Jiahui Hou, Linlin Chen, Taeho Jung, Xiang-Yang Li