Day 1 12/11/2019
Room #1

Registration 08:30 - 09:00

Opening Remarks 09:00 - 09:30

Keynote Presentation by Prof. Zixiang Xiong 09:30 - 10:30

Fundamental Limits on Energy Consumption in Wireless Network

Break 10:30 - 11:00

Session 1 11:00 - 12:30

Internet of Things (IoT) and Cloud Networks
11:00 - 11:15
A Trust Architecture for Blockchain in IoT

Blockchain is a promising technology for establishing trust in IoT networks, where network nodes do not necessarily trust each other. Cryptographic hash links and distributed consensus mechanisms ensure that the data stored on an immutable blockchain can not be altered or deleted. However, blockchain mechanisms do not guarantee the trustworthiness of data at the origin. We propose a layered architecture for improving the end-to-end trust that can be applied to a diverse range of blockchain-based IoT applications. Our architecture evaluates the trustworthiness of sensor observations at the data layer and adapts block verification at the blockchain layer through the proposed data trust and gateway reputation modules. We present the performance evaluation of the data trust module using a simulated indoor target localization and the gateway reputation module using an end-to-end blockchain implementation, together with a qualitative security analysis for the architecture.
Authors: VOLKAN DEDEOGLU (CSIRO Data61), Raja Jurdak (The Queensland University of Technology), Guntur Putra (UNSW Sydney), Ali Dorri (UNSW Sydney), Salil Kanhere (UNSW Sydney),
Hide Authors & Abstract

Show Authors & Abstract
11:15 - 11:30
Fuzzy Clustering-Based Task Allocation Approach Using Bipartite Graph in Cloud-Fog Environment

Recently, due to the limitations in using cloud computing services for the recent advances IoTs applications, a newly distributed computing architecture is established called cloud-fog paradigm by exploiting the cooperation between fog and cloud entities. Fog nodes are used to reduce monetary cost and transferring latency for cloud resources, while for offloading of large-scale applications cloud servers are used. In this paradigm, The main problem is task allocation which aims to select the optimal nodes among cloud and fog nodes for each task to minimize makespan, monetary and energy costs. In this paper, to solve this problem a new task allocation approach called bipartite graph with fuzzy clustering task allocation approach is proposed and it uses a hybrid DAG for representing independent and dependent tasks. Also, it uses fuzzy clustering and bipartite graph to solve the uncertainty executing problem and find the maximum bipartite matching, respectively. The conducted simulation results show that the proposed approach can achieve a higher performance int terms of makespan, total cost, and cost-makespan tradeoff than existing approaches.
Authors: Asaad Ahmed (Consultant for Vice President of Development, King Abdul-Aziz University, Jeddah, Saudi Arabia (Home Country Address: Department of Mathematics and Computer Science,Faculty Of Science, Al-Azhar University, Cairo, Egypt)), Amin Noaman (Faculty of Computers and Information Technology, King Abdul-Aziz University, Jeddah, Saudi Arabia),
Hide Authors & Abstract

Show Authors & Abstract
11:30 - 11:45
Impact of consensus on appendable-block blockchain for IoT

The Internet of Things (IoT) is transforming our physical world into a complex and dynamic system of connected devices on an unprecedented scale. Recently, blockchain technology has received a lot of attention from the community as a possible solution to overcome security issues in IoT. However, traditional blockchains (e.g., the ones used in Bitcoin and Ethereum) are not well suited to the resource-constrained nature of IoT devices and also with the large volume of information that is expected to be generated from typical IoT deployments. To overcome these issues, researchers have presented lightweight instances of blockchains tailored for IoT. For example, proposing novel data structures based on blocks with appendable data. However, these researchers did not discuss how the consensus algorithm would impact their solutions. In this paper, we improved an appendable-block blockchain framework to support different consensus algorithms through a modular design. We evaluated the performance in different scenarios and studied the impact of varying the number of devices and transactions and employing different consensus algorithms. Even adopting different consensus algorithms, results indicate that the latency to append a new block is less than 161ms (in the more demanding scenario) and the delay for processing a new transaction is less than 7ms, suggesting that our version of the appendable-block blockchain is efficient and scalable, and thus well suited for IoT scenarios.
Authors: Roben Castagna Lunardi (IFRS), Regio Michelin (UNSW), Charles Neu (UNISC), Henry Nunes (PUCRS), Avelino Zorzo (PUCRS), Salil Kanhere (UNSW),
Hide Authors & Abstract

Show Authors & Abstract
11:45 - 12:00
Pushing Smart Caching to the Edge with BayCache

Caching contents in a small cell base station (SBS) is getting supported more and more widely today due to the Internet traffic growth and the requirement of low access latency. A primary concern and challenging issue with cache-enabled SBSs is how to better utilize network resources to achieve high overall performance. Though existing caching strategies can solve the problem to some extent, they are far from perfect — pure popularity-based ones usually lead to sub-optimal caching performance whereas global coordination ones suffer from extra cost of many control messages. To deal with the issue, we present an adaptive caching scheme based on Bayesian inference, which 1) identifies the traffic features over time for each SBS; 2) synthesizes various features to rank the contents via a Bayesian ranking model; and 3) does cache placement online according to the ranking results. Unlike existing approaches that only highlight some specific factor, our scheme, due to the adoption of a Bayesian approach, can easily support additional features of high impact on caching performance and measure them in a decentralized way within a single SBS. We evaluate our scheme under various circumstances in terms of SBSs density, cache size, content popularity and skewness. The results show that our solution using multiple features exhibits improved performance — it can reduce more than 30% the overall network latency in some cases when compared with solutions that only use a single feature.
Authors: Dongbiao He (Tsinghua University), Jinlei Jiang (Tsinghua University), Guangwen Yang (Tsinghua University), Cedric Westphal (UCSC),
Hide Authors & Abstract

Show Authors & Abstract
12:00 - 12:15
Smart Discovery of Periodic-Frequent Human Routines for Home Automation

In this paper, we present an approach to discover periodic-frequent multi-step human routines in event data from smart devices and sensors deployed at home. Based on the discovered routines, our approach is able to suggest rules to automate the control of different aspects of the home environment. We evaluate our approach through an in the lab study, a study based on synthetic data, and an in-the-wild study. Our results show that our approach exhibits a high recall-precision performance, with a recovery rate of around 90% for most of the cases under investigation.
Authors: Alejandro Sanchez Guinea (Technische Universität Darmstadt), Andrey Boytsov (University of Luxembourg), Ludovic Mouline (University of Luxembourg), Yves Le Traon (University of Luxembourg),
Hide Authors & Abstract

Show Authors & Abstract
12:15 - 12:30
The Impression of Virtual Experience: Mobile Augmented Reality Cloud Solution

Mobile devices have become an enormous platform for augmented reality(AR) technology to meet the need of users to experience the charm of AR technology in anytime and anywhere. However, the development of mobile AR technology is currently in a dilemma. Both extensibility and real-time are often challenging to be taken into account. In our study, we explored cloud-based solutions and designed two systems, S-MARC, and D-MARC. The first system boldly tried the idea of cloud technology supporting mobile AR and explored more possibilities for combining location information. The cloud-based mobile AR design optimization scheme is proposed in our study details. Besides, we invited some users to carry out experimental studies, and their user evaluations showed that the mobile AR cloud solution was implemented in both systems and received excellent feedback in the review. The results show that the proposed two schemes are superior to others in the performance and efficiency, and help to improve the current dilemma of mobile AR and build up users' confidence in the application of mobile AR.
Authors: Ying Zhao, Wenyuan Tao (School of Computer Software,Tianjin University), Chung-Ming Own (School of Computer Software,Tianjin University),
Hide Authors & Abstract

Show Authors & Abstract

Lunch 12:30 - 13:30

Session 2 13:30 - 15:00

Machine Learning for Mobile Networks
13:30 - 13:45
A Deep Spatio-temporal Attention-based Neural Network for Passenger Flow Prediction

Predicting the passenger flows in a city, especially in a metropolis, can guide traffic dispersion, and help assessing the risks of public safety and improving urban planning. However, it is challenging as passenger flows in a road network may vary with time and space, affected by weather conditions, urban activities, etc. In the paper, we propose a passenger flow prediction approach named Yildun, which constructs an encoder-decoder neural network and captures the spatial and temporal correlations inherent in passenger flows. More specifically, to predict the passenger flows at each and every station, a spatial attention mechanism is presented to adaptively extract inter-station correlations of flows by referring to the previous hidden state of the encoder at each time step. Meanwhile, a temporal attention mechanism is employed to capture time-dependent connections of flows by selecting relevant hidden states of the encoder across all time steps. Further, extra factors, such as POI (Point of Interest) data and day of the week, are fused in the decoder. With this spatio-temporal attention scheme, Yildun not only can make predictions effectively, but also is easily explainable. Extensive experiments are conducted on large-scale real-world data. The experimental results show that Yildun can predict passenger flows with small prediction errors and outperforms five baselines significantly.
Authors: Yanling Cui (Institute of Software, Chinese Academy of Sciences), Beihong Jin (Institute of Software, Chinese Academy of Sciences), Fusang Zhang (Institute of Software, Chinese Academy of Sciences), Xingwu Sun (Meituan-Dianping Group),
Hide Authors & Abstract

Show Authors & Abstract
14:00 - 14:15
How Smart Your Smartphone Is in Lie Detection?

Lying is a (practically) unavoidable component of our day to day interactions with other people, and it includes both oral and textual communications (e.g. text entered via smartphones). Detecting when a person is lying has important applications, especially with the ubiquity of messaging via smart-phones, coupled with rampant increases in (intentional) spread of misinformation today. In this paper, we design a technique to detect whether or not a person’s textual inputs when typed via a smartphone indicate lying. To do so, frst, we judiciously develop a smartphone based survey that guarantees any participant to provide a mix of true and false responses. While the participant is texting out responses to each question, the smartphone measures readings from its inbuilt inertial sensors, and then computes features like shaking, acceleration, tilt angle, typing speed etc. experienced by it. Subsequently, for each participant (47 in total), we glean the true and false responses using our own experiences, and also via informal discussions with each participant. By comparing the responses of each participant, along with the corresponding motion features computed by the smartphone, we implement several machine learning algorithms to detect when a participant is lying, and our accuracy is around 70% in the most stringent leave-one-out evaluation strategy. Later, utilizing fndings of our analysis, we develop an architecture for real-time lie detection using smartphones.
Authors: Md. Mizanur Rahman (Student), Atanu Shome (Bangladesh University of Engineering and Technology), Sriram Chellappan (University of South Florida), A. B. M. Alim Al Islam (Bangladesh University of Engineering and Technology),
Hide Authors & Abstract

Show Authors & Abstract
14:15 - 14:30
Learning Trajectories as Words: A Probabilistic Generative Model for Destination Prediction

Destination prediction is an essential task for many location based services such as recommending sightseeing places and sending targeted advertisements. Most existing techniques utilize the historical trajectories to predict destinations, but they fail to well describe the spatio-temporal characteristics of trajectories and suffer the trajectory sparsity problem, i.e., the available historical trajectories are far from being able to cover all possible trajectories. The temporal sensitivity of historical trajectories highlights the sparsity problem even more. In this paper, we address this problem by building a probabilistic generative model to capture the spatio-temporal features of trajectories. We develop an extended Latent Dirichlet Allocation (LDA) model to characterize the generative mechanism of track points in each trajectory. In this model, trajectory, track point of trajectory and destination are regarded as document, word and response respectively. To address the trajectory sparsity problem, each trajectory is expressed by the distribution of trajectory patterns which are the topics discovered from historical trajectories. Then, the most likely destination is predicted through the trajectory patterns. The experiments performed on a real-world taxi trajectory dataset from Guangzhou confirm the advantage of the probabilistic generative model in destination prediction, achieving remarkable accuracy and strong interpretability.
Authors: Yuhuan Lu (Sun Yat-sen University), Zhaocheng He (Sun Yat-sen University), Liangkui Luo (Sun Yat-sen University),
Hide Authors & Abstract

Show Authors & Abstract
14:30 - 14:45
Mobile Augmented Reality Techniques for Emergency Response

In an emergency situation, each response agent must act quickly and accurately. The support of a mobile device that can provide an appropriate insight of the surroundings and that allows users to exchange information with the other members of the emergency teams, can prevent harm and even save many lives. This paper presents a mobile application that integrates a georeferenced system with augmented reality techniques, in order to serve the needs of the operatives in emergency situations. The work intends to introduce solutions which optimize the performance with which the user takes advantage of the mobile application, such as the organization of the data flow that is displayed and the augmentation of the surrounding area. User studies were conducted with members of the National Navy. The results were positive although there are still some aspects that should be improved.
Authors: Alexandre Campos (Faculdade de Ciências e Tecnologia, Universidade NOVA de Lisboa), Nuno Correia (NOVA LINCS, Faculdade de Ciências e Tecnologia, Universidade NOVA de Lisboa), Teresa Romão (NOVA-LINCS, Faculdade de Ciências e Tecnologia, Universidade NOVA de Lisboa), Isabel Nunes (UNIDEMI, Faculdade de Ciências e Tecnologia, Universidade NOVA de Lisboa), Mário Simões-Marques (CINAV- Portuguese Navy),
Hide Authors & Abstract

Show Authors & Abstract
14:45 - 15:00
Starling Swarm Algorithm: An Approach to Autonomous Coordination of Intensive Agents

For the Swarm Coordination Problem (SCP), traditional control strategies met some limitations like difficult model parameter adjustment and week coordination effeteness etc. Some PSO and its improvement algorithms did not present good coordination effects, particularly for the applications of intensive agents. Therefore, we specifically designed a novel Starling Swarm Algorithm (SSA) from the latest research findings of coordination mechanism of starling flock, like local perception, self-adaptation, centerless self-organization and self-adaptation. We designed the SSA combining with local perception, safe avoidance, example selection and decision evolution. In the paper, we also presented 9 basic rules that agent group should follow which reflect the essence of group coordination well. To evaluate coordination effect, we defined convergence coefficient to quantify coordination in the paper. Based on it, we carried out our experiments from parameter analysis, efficiency and effectiveness under different situations and presented our results of changes of convergence coefficient and comparison of SSA with classical PSO. In cases of different agent numbers, we obtained the results of number of iterations for implementing task, respectively, which showed that our algorithm was more efficient than PSO. For effectiveness analysis, simulations of effects of group coordination under various conditions of no obstacle, multiple obstacles and intruder were also presented in the paper.
Authors: Rong Xie,
Hide Authors & Abstract

Show Authors & Abstract

Break 15:00 - 15:30

Session 3 15:30 - 17:30

Security and Privacy
15:30 - 15:45
{P}Net: Privacy-preserving Personalization of AI-based Models by Anonymous Inter-person Similarity Networks

Emerging proactive applications need user data to power their underlying AI algorithms. However, both the training and inference tasks are typically performed in the provider's cloud, leading to multiple privacy issues to the data subject. Current privacy-preserving concepts are neither practicable for AI algorithms nor promise mutual benefits for both parties. To address these issues, we propose {P}Net---a novel two-level approach for privacy-preserving personalization, which exploits `divisible' AI algorithms and anonymous inter-person similarity measurements. In short, {P}Net splits the training task into a cloud-based general model learning process and `local' personalization steps in which the general `black-box' model is subsequently adapted to individuals (level 1). Based on anonymously-contributed model modifications (patches) resulting from the first level, {P}Net further allows new users to request a community model---representing the general model personalized by crowd-sourced patches from other similar users (level 2). Our experiments show that {P}Net outperforms existing techniques especially when only a few user data is labeled. With {P}Net, new users can now directly benefit from personalized applications in a practical and privacy-preserving way with reduced labeling effort.
Authors: Christian Meurisch (TU Darmstadt), Sebastian Kauschke (TU Darmstadt), Tim Grube (TU Darmstadt), Bekir Bayrak (TU Darmstadt), Max Mühlhäuser (TU Darmstadt),
Hide Authors & Abstract

Show Authors & Abstract
15:45 - 16:00
Detecting Malicious Applications using System Services Request Behavior

Widespread growth in Android malware stimulates security researchers to propose different methods for analyzing and detecting malicious behaviors in applications. Nevertheless, current solutions are ill-suited to extract the fine-grained behavior of Android applications accurately and efficiently. In this paper, we propose ServiceMonitor, a lightweight host-based detection system that dynamically detects malicious applications directly on mobile devices. ServiceMonitor reconstructs the fine-grained behavior of applications based on their interaction with system services (i.e., SMS manager, camera, wifi networking, etc). ServiceMonitor monitors the way applications request system services in order to build a statistical Markov chain model to represent what and how system services are used. Afterward, we use this Markov chain as a feature vector to classify the application behavior into either malicious or benign using the Random Forests classification algorithm. We evaluated ServiceMonitor using a dataset of 8034 malware and 10024 benign applications and obtaining 96.7% of accuracy rate and negligible overhead and performance penalty.
Authors: Majid Salehi (KU Leuven), Morteza Amini (Sharif University of Technology), Bruno Crispo (KU Leuven, University of Trento),
Hide Authors & Abstract

Show Authors & Abstract
16:00 - 16:15
HEliOS: Huffman Coding Based Lightweight Encryption Scheme for Data Transmission

Demands for fast data sharing among the smart devices tend to increase in a great pace now-a-days. Such an increase possess challenges towards ensuring essential security for online shared data while maintaining the resource usage at a reasonable level. Existing research studies attempt to leverage compression based encryption for enabling such secure and fast data transmission replacing the traditional resource-heavy encryption schemes. Current compression-based encryption methods mainly focus on error insensitive digital data formats and prone to be vulnerable to different attacks. Therefore, in this paper, we propose and implement a new Huffman compression based Encryption scheme using lightweight dynamic Order Statistic tree (HEliOS) for digital data transmission. The core idea of HEliOS involves around finding a secure encoding method based on a novel notion of Huffman coding, which compresses the given digital data using a small sized secret intelligence. HEliOS does this in such a way that, without the possession of the secret intelligence, an attacker will not be able to decode the encoded compressed data. Hence, by encrypting only the small-sized intelligence, we can secure the whole compressed data. We demonstrate the level of security of HEliOS through theoretical analysis. Our extensive real experimental evaluation for downloading and uploading digital data to and from a personal cloud storage Dropbox server confirms efficacy and lightweight nature of HEliOS.
Authors: Mazharul Islam (Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology, Dhaka, Bangladesh), Novia Naurin (Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology, Dhaka, Bangladesh), Mohammad Kaykobad (Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology, Dhaka, Bangladesh), Sriram Chellappan (Department of Computer Science and Engineering, University of South Florida, Tampa, FL, USA), A. B. M. Alim Al Islam (Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology, Dhaka, Bangladesh),
Hide Authors & Abstract

Show Authors & Abstract
16:15 - 16:30
Jive: Spatially-Constrained Encryption Key Sharing using Visible Light Communication

This paper investigates a novel encryption key sharing mechanism using the emerging wireless technology Visible Light Communication (VLC). Based on the idea of transmitting data by modulating light, we are able to (1) share a secret key within a constrained physical space and (2) communicate encrypted information among co-located mobile devices using the shared key. We present the demonstration JIVE (Joint Integration of VLC and Encryption), a framework to support secret key sharing over Visible Light Communication. In defining JIVE, we tackle challenges related to data encoding, message synchronization, and environmental noise to build a reliable, low complexity system using off-the shelf hardware. Our system is capable of sending encryption keys at speeds of more than 750bps using ultra short, high speed light pulses imperceptible to the human eye. Additionally, we have developed an application for Android that interfaces with the VLC device through serial communication so that applications running on mobile devices can subsequently use the keys to encrypt application data. Experimental results illustrate the high accuracy of our system across a variety of different variables. Finally, we position our system for use by a variety of applications that require a high-level of data security among physically co-located devices.
Authors: Jayanth Shenoy (The University of Texas at Austin), Aditya Tyagi (The University of Texas at Austin), Meha Halabe (The University of Texas at Austin), Christine Julien (The University of Texas at Austin),
Hide Authors & Abstract

Show Authors & Abstract
16:30 - 16:45
MultiLock: Biometric-Based Graded Authentication for Mobile Devices

While traditionally smartphones have relied on methods such as a passcode or pattern-based authentication, biometric authentication techniques are gaining popularity. However current biometric methods are heavily dependent on various environmental factors. For example, face authentication methods depend on lighting conditions, camera shake and picture framing, while fingerprint scanning relies on finger placement. All of these variables can result in these systems becoming time-consuming for the user to use. To remedy these problems, we propose MultiLock, a passive, graded authentication system, which uses face authentication as a case study to propose a system that gives users access to their devices without requiring them to manually interact with the lock screen. MultiLock allows a user to categorize applications into various security bins based on their sensitivity. By doing so MultiLock can grant users access to different sensitivity applications, based on varying degrees of sureness that the device is being used by its rightful owner. Thus, allowing the device to be used even in adverse lighting conditions without hampering user experience. In our tests, MultiLock was able to grant access to users for 88% of the interactions on average, while passively running in the background. While we use face authentication as an example to demonstrate and propose MultiLock, our system can be used with any confidence based biometric system.
Authors: Shravan Aras (University of Arizona Tucson, Arizona), Chris Gniady (University of Arizona Tucson, Arizona), Hari Venugopalan (University of California, Davis Davis, California),
Hide Authors & Abstract

Show Authors & Abstract
16:45 - 17:00
PPCA: Privacy-Preserving Conditional Actions for IoT Environments Using Smart Contracts

Automated tasks play an important role in both consumer and industrial IoT environments. In many scenarios, the IoT tasks are performed given certain conditions. To facilitate the tasks, it is necessary to delegate a third party to listen to events that trigger the conditions and issue commands to the IoT resources accordingly. However, without restriction, the third party may be over-privileged and able to control the resources unconditionally. We define the third party's permission to act under some conditions as a conditional action. We propose PPCA, a privacy-preserving service that allows users to create conditional actions in a decentralized platform using smart contracts. PPCA can guarantee strict privilege at the third party that holds conditional actions. By generalizing a variety of conditions into simple forms of conditional logic, the conditions can be verified in a privacy-preserving manner. We build a prototype of PPCA on Ethereum. The performance shows the feasibility of PPCA in IoT environments.
Authors: Tam Le, Matt Mutka (Michigan State University),
Hide Authors & Abstract

Show Authors & Abstract
17:00 - 17:15
Quantifying Location Privacy in Permissioned Blockchain-Based Internet of Things (IoT)

Recently, blockchain has received much attention from the mobility-centric Internet of Things (IoT). It is deemed the key to ensuring the built-in integrity of information and security of immutability by design in the peer-to-peer network (P2P) of mobile devices. In a permissioned blockchain, the authority of the system has control over the identities of its users. Such information can allow an ill-intentioned authority to map identities with their spatiotemporal data, which undermines the location privacy of a mobile user. In this paper, we study the location privacy preservation problem in the context of permissioned blockchain-based IoT systems under three conditions. First, the authority of the blockchain holds the public and private key distribution task in the system. Second, there exists a spatiotemporal correlation between consecutive location-based transactions. Third, users communicate with each other through short-range communication technologies such that it constitutes a proof of location (PoL) on their actual locations. We show that, in a permissioned blockchain with an authority and a presence of a PoL, existing approaches cannot be applied using a plug-and-play approach to protect location privacy. In this context, we propose BlockPriv, an obfuscation technique that quantifies, both theoretically and experimentally, the relationship between privacy and utility in order to dynamically protect the privacy of sensitive locations in the permissioned blockchain.
Authors: Abdur Rahman Bin Shahid (Florida International University), Niki Pissinou (Florida International University), Laurent Njilla (US. Air Force Research Laboratory), Sheila Alemany (Florida International University), Ahmed Imteaj (Florida International University), Kia Makki (Technological University Of America), Edwin Aguilar (Florida International University),
Hide Authors & Abstract

Show Authors & Abstract
17:15 - 17:30
Saving Private Addresses: An Analysis of Privacy Issues in the Bluetooth-Low-Energy Advertising Mechanism

The Bluetooth Low Energy (BLE) protocol is being included in a growing number of connected objects such as fitness trackers and headphones. As part of the service discovery mechanism of BLE, devices announce themselves by broadcasting radio signals called advertisement packets that can be collected with off-the-shelf hardware and software. To avoid the risk of tracking based on those messages, BLE features an address randomization mechanism that substitutes the device address with random temporary pseudonyms, called Private addresses. In this paper, we analyze the privacy issues associated with the advertising mechanism of BLE, leveraging a large dataset of advertisement packets collected in the wild. First, we identified that some implementations fail at following the BLE specifications on the maximum lifetime and the uniform distribution of random identifiers. Furthermore, we found that the payload of the advertisement packet can hamper the randomization mechanism by exposing counters and static identifiers. In particular, we discovered that advertising data of Apple and Microsoft proximity protocols can be used to defeat the address randomization scheme. Finally, we discuss how some elements of advertising data can be leveraged to identify the type of device, exposing the owner to inventory attacks.
Authors: Guillaume Celosia (Univ Lyon, INSA Lyon, Inria, CITI), Mathieu Cunche (Univ Lyon, INSA Lyon, Inria, CITI),
Hide Authors & Abstract

Show Authors & Abstract

Dinner and Best Paper Award 18:00 - 21:00

Room #2

EFIOT Workshop 11:00 - 12:40

Edge, Fog, and Cloud Computing for the Internet of Things
11:00 - 11:25
Encryption Timings and Energy Measurements on Constrained Devices

In this paper we present our analysis of the processing power and energy consumption of various symmetric and asymmetric encryption algorithms on a constrained device, specifically the Texas Instruments cc2650 running Contiki 3. The algorithms of AES and ECC were tested, with the following implementations tinyAES, bcon’s AES, and Contiki AES.
Authors: Brandon Tsao, Yuhong Li (Santa Clara University), Behnam Dezfouli (Santa Clara University, USA),
Hide Authors & Abstract

Show Authors & Abstract
11:25 - 11:50
Lighter U-Net for Segmenting White Matter Hyperintensities in MR Images

White matter hyperintensities (WMH) is one of main consequences of small vessel diseases. Automated WMH segmentation techniques play an important role in clinical research and practice. U-Net has been demonstrated to yield the best precise segmentation results so far. However, sometimes it losses more detailed information as network goes deeper. In addition, it usually depends on data augmentation or a large number of filters. Large filters increase the complexity of model, which may be an obstacle for real-time segmentation on cloud computing. To solve these two issues, a new architecture, Lighter U-Net is proposed to reinforce feature use, to reduce the number of parameters as well as to retain sufficient receptive fields without losing resolution. The extensive experiments suggest that the proposed network achieves comparable performance as the state-of-the-art methods by only using 17% parameters of standard U-Net.
Authors: Jun Zhuang (Indiana University-Purdue University Indianapolis), Mingchen Gao (University at Buffalo), Mohammad Hasan (Indiana University-Purdue University Indianapolis),
Hide Authors & Abstract

Show Authors & Abstract
11:50 - 12:15
A game-theoretic framework for dynamic cyber deception in Internet of Battlefield Things

Cyber deception techniques are crucial to protect networks in battlefield settings and combat malicious cyber attacks. Cyber deception can effectively disrupt the surveillance process outcome of an adversary. In this paper, we propose a novel approach for cyber deception to protect important nodes and trap the adversary. We present a sequential approach of honeypot placement to defend and protect the network vital nodes. We formulate a stochastic game to study the dynamic interactions between the network administrator and the attacker. The defender makes strategic decisions about where to place honeypots to introduce new vulnerabilities to the network. The attacker's goal is to develop an attack strategy to compromise the nodes of the network by exploiting a set of known vulnerabilities. To consider a practical threat model, we assume that the attacker can only observe a noisy version of the network state. To this end, both players solve a partially observable stochastic game (POSG). Finally, we present a discussion on existing techniques to solve the formulated game and possible approaches to reduce the game complexity as part of our ongoing and future research.
Authors: Ahmed HEMIDA (Army Research Lab), Charles Kamhoua (Army Research Lab), Nandi Leslie (Army Research Lab),
Hide Authors & Abstract

Show Authors & Abstract
12:15 - 12:40
An Analysis of Fog Computing Data Placement Algorithms

This work evaluates three different data placement algorithms in Fog Computing, using the iFogSim simulator. The paper describes the three algorithms (OnlyCloud, Mapping, EdgeWard) in the con- text of an Internet of Things scenario, which has been based on an e-health system with variations in applications and network topology. Results achieved show that edge placement strategies are beneficial to assist cloud computing in lowering latency and Cloud energy expenditure.
Authors: Daniel Silva (Universidade Lusofóna de Humanidades e Tecnologias, Portugal; COPELABS - Cognitive and People-Centric Computing, Portugal), Godwin Asaamoning (Universidade Lusófona de Humanidades e Tecnologias), Hector Orrillo (Universidade Lusófona de Humanidades e Tecnologias), Rute Sofia (fortiss, Germany;Universidade Lusofóna de Humanidades e Tecnologias, Portugal; ISTAR-IUL, Instituto Universitário de Lisboa (ISCTE-IUL), Portugal), Paulo Mendes (Airbus Central Research and Technology),
Hide Authors & Abstract

Show Authors & Abstract
Day 2 13/11/2019
Room #1

Registration 08:30 - 09:00

Keynote Presentation by Prof. Gil Zussman 09:00 - 10:00

COSMOS Wireless Testbed and Experimentation with Compact Full-Duplex Wireless

Break 10:00 - 10:30

Session 4 10:30 - 12:00

Human-Centered Computing
10:30 - 10:45
AugToAct: Scaling Complex Human Activity Recognition with Few Labels

Human activity recognition (HAR) from wearable sensor data has recently gained widespread adoption in a number of fields. However, recognizing complex human activities, postural and rhythmic body movements (e.g. dance, sports) is challenging due to the lack of domain-specific labeling information, the perpetual variability in human movement kinematics profiles due to age, sex, dexterity and the level of professional training. In this paper, we propose a deep activity recognition model to work with limited labeled data,both for simple and complex human activities. To mitigate the intra and inter-user spatio-temporal variability of movements, we posit novel data augmentation and domain normalization techniques. We depict a semi-supervised technique that learns noise and transformation invariant feature representation from sparsely labeled data to accommodate intra-personal and inter-user variations of human movement kinematics. We also postulate a transfer learning approach to learn domain invariant feature representations by minimizing the feature distribution distance between the source and target domains.We showcase the improved performance of our proposed framework, AugToAct, using a public HAR dataset. We also design our own data collection, annotation and experimental setup on complex dance activity recognition steps and kinematics movements where we achieved higher performance metrics with limited label data compared to simple activity recognition tasks.
Authors: Abu Zaher Md Faridee (UMBC), Md Abdullah Al Hafiz Khan (Philiphs Research North America), Nilavra Pathak (Expedia Group), Nirmalya Roy (UMBC),
Hide Authors & Abstract

Show Authors & Abstract
10:45 - 11:00
MFE-HAR: Multiscale Feature Engineering for Human Activity Recognition Using Wearable Sensors

Human activity recognition plays a key role in the application areas such as fitness tracking, healthcare and aged care support. However, inaccurate recognition results may cause an adverse effect on users or even an unpredictable accident. In order to improve the accuracy of human activity recognition, multi-device and deep learning based approaches have been proposed. However, they are not practical on a daily basis due to the limitations that devices are difficult to wear, and deep learning requires large training dataset and incurs expensive computational costs. To address this problem, we propose a novel approach, multiscale feature engineering for human activity recognition (MFE-HAR), which exploits the properties of arm movement from global and local scales using the accelerometer and gyroscope sensors on a single wearable device. Our method takes advantage of having important features at multiple scales over previous single-scale methods. We evaluated the performance of the proposed method on two public datasets and achieved the mean classification accuracy of 93\% and 98\% respectively. Our proposed system performs better than the state of the art multi-device based approaches, and is more practical for real-world applications.
Authors: jianchao lu (Macquarie University), Jiaxing Wang (Zhejiang University of Technology), xi zheng (Macquarie University), Quan Z. Sheng (Macquarie University), Wanlei Zhou (University of Technology Sydney), Zawar Hussain (Macquarie University),
Hide Authors & Abstract

Show Authors & Abstract
11:00 - 11:15
MicPrint: Acoustic Sensor Fingerprinting for Spoof-Resistant Mobile Device Authentication

Smartphones are the most commonly used computing platform for accessing sensitive and important information placed on the Internet. Authenticating the smartphone’s identity in addition to the user’s identity is a widely adopted security augmentation method since conventional user authentication methods, such as password entry, often fail to provide strong protection by itself. In this paper, we propose a sensor-based device fingerprinting technique for identifying and authenticating individual mobile devices. Our technique, called MicPrint, exploits the unique characteristics of embedded microphones in mobile devices due to manufacturing variations in order to uniquely identify each device. Unlike conventional sensor-based device fingerprinting that are prone to spoofing attack via malware, MicPrint is fundamentally spoof-resistant since it uses acoustic features that are prominent only when the user blocks the microphone hole. This simple user intervention acts as implicit permission to fingerprint the sensor and can effectively prevent unauthorized fingerprinting using malware. We implement MicPrint on Google Pixel 1 and Samsung Nexus to evaluate the accuracy of device identification. We also evaluate its security against simple raw data attacks and sophisticated impersonation attacks. The results show that after several incremental training cycles under various environmental noises, MicPrint can achieve high accuracy and reliability for both smartphone models.
Authors: Yongwoo Lee (University of Wisconsin–Madison), Jingjie Li (University of Wisconsin–Madison), Younghyun Kim (University of Wisconsin–Madison),
Hide Authors & Abstract

Show Authors & Abstract
11:15 - 11:30
NextAct: A Hybrid Approach for High-resolution Human Activity Predictions

Emerging anticipatory applications that intervene or proactively support users in daily life require an in-depth understanding of prospective human behavior. However, accurate and fine-grained prediction of human behavior is complex and conventional machine learning approaches show limitations. In this paper, we present a novel approach, namely NextAct, to predict human activities by utilizing temporal and spatial features predicted from calendar events, user routines and sequences of next place predictions for remaining uncertain time slots in between. NextAct hereby makes spatial features in addition to the obvious temporal features available for the human activity predictor. We evaluate our hybrid approach on a four-week user-annotated dataset from 30 participants; we compare our forecasting results against prediction techniques proposed in the literature. The results show an F1-score up to 82.6% and a performance gain up to 16.6% compared to state of the art approaches. We also discuss the cold-start problem of individual models and show that we achieve adequate results after 4 days. NextAct opens a novel way for human activity prediction to proactively support a user and optimize his next activity or day.
Authors: Christian Meurisch (TU Darmstadt), Artur Gogel (TU Darmstadt), Bekir Bayrak (TU Darmstadt), Max Mühlhäuser (TU Darmstadt),
Hide Authors & Abstract

Show Authors & Abstract
11:30 - 11:45
Understanding the Impact of Number of CPU Cores on User Satisfaction in Smartphones

Understanding user experience/satisfaction with mobile systems in order to manage computational resources has become a popular approach in recent years. One of the key issues in this area is to gauge user satisfaction. In this paper, we study CPU configurations’ impact on user satisfaction and power consumption with real users. Specifically, we propose a system to save energy by altering CPU core count and frequency while keeping users satisfied. The system utilizes user-facing metrics such as frame rate and input lag to predict user satisfaction and sets CPU core and frequency to maximize satisfaction while minimizing power consumption in real-time. We first study a set of applications in-the-lab and show that we can accurately model satisfaction with the collected user-facing metrics. We then go intothe- wild in order to evaluate the proposed system in real environments. In the wild, we build a user-independent (useroblivious) and user-dependent (personal) model. Users test the two models and the default scheme for one-week duration, which composes 140 days of worth of data. When compared to default scheme, our results show that, without impacting satisfaction, userindependent and user-dependent models save 12.3% and 11.8% of total system energy on average, respectively.
Authors: Emirhan Poyraz, Prethvi Kashinkunti (Nvidia), Matt Schuchhardt (4C), Michael Kishinevsky (Intel), Niranjan Soundararajan (Intel), Gökhan Memik (Northwestern University),
Hide Authors & Abstract

Show Authors & Abstract
11:45 - 12:00
WiLay: Building Wi-Fi-Based Human Activity Recognition System through Activity Hierarchical Relationship

Recently, Wi-Fi-based human activity recognition technique has attracted attentions extensively. Due to its ease of access and low cost, Wi-Fi-based technique achieves great potential on building human activity recognition systems. However, this technique is limited because the Wi-Fi signal is less-informative and susceptible to environmental changes. To build a practical Wi-Fi-base human activity recognition system, in this paper, WiLay, a layer-structured human activity recognition system is proposed. To recognize 7 different activities, WiLay uses an activity-oriented process to select and extract features according to the hierarchical relationship between different activities, and trains multiple classifiers to build its layer-structured recognition system. We collected data on several different environments and tested our system. The experimental results with 95.4% accuracy and 89.1% recall rate indicate that our system has very well performance on recognition human activities and is robust to environmental changes.
Authors: Yongqiang Jiang (Beijing Key Laboratory of Digital Media, School of Computer Science and Engineering, Beihang University, Beijing 100191, China), Hai-Miao Hu (Beijing Key Laboratory of Digital Media, School of Computer Science and Engineering, Beihang University, Beijing 100191, China; State Key Laboratory of Virtual Reality Technology and Systems, Beihang University, Beijing 100191, China), Yanglin Pu (Beijing Key Laboratory of Digital Media, School of Computer Science and Engineering, Beihang University, Beijing 100191, China), Haoran Jiang (Information Technology Bureau of Shandong Province, China Post Group, Jinan 250000, China),
Hide Authors & Abstract

Show Authors & Abstract

Lunch 12:00 - 13:00

Session 5 13:00 - 14:30

Machine Learning for M-health
13:00 - 13:15
A cost-effective and non-invasive system for sleep and vital signs monitoring using passive RFID tags

Inadequate sleep has very bad impacts on human daily life activities. Sleep disorders are one of the main reasons for the inadequate sleep problem. Diagnosis of these disorders is a very challenging task because they occur at night when the patients are asleep and requires a full night monitoring to diagnose. Conventional sleep monitoring techniques are not feasible due to various reasons such as they require the person to wear multiple invasive sensors, the cost is high and also need special environment for monitoring. In this work, we propose a non-invasive and cost-effective solution based on passive RFID technology for sleep monitoring. We attach passive RFID tags to the shirt of the person and collect low-level data which can capture the information about the vital signs of the person. The proposed solution can detect on-bed movements which provide useful information about the sleep of the person. We propose an adaptive threshold-based technique which can effectively identify the stable breathing and apnea regions from the collected breathing signal. The stable region is used for estimating the breathing rate. Our solution can monitor vital signs such as respiration rate during the overnight sleep and can detect sleep apnea. We implement the proposed solution using COTS RFID tags in the real world scenarios and the results show that our system achieves 100\% accuracy for apnea detection and above 95\% for breath rate estimation.
Authors: Zawar Hussain (Macquarie University), Subhash Sagar (Macquarie University), Wei Zhang (The University of Adelaide), Quan Sheng (Macquarie University),
Hide Authors & Abstract

Show Authors & Abstract
13:15 - 13:30
Adaptive Robotic Rehabilitation using Muscle Fatigue as a Trigger

Fatigue is a pervasive symptom following brain injury or disease. It has been known to impact recovery and hence is an important factor in rehabilitation. Robotic rehabilitation may be one way to reduce fatigue because of the robot’s capability to adapt to user’s performance. This paper explores an adaptive rehabilitation system to provide personalized upper limb rehabilitation. The system collects EMG data from the major muscles responsible for movement and adapts the forces used for rehabilitation (assistive and resistive) in real time based on muscle fatigue. Experimental results and the user survey outcomes show that the system was able to detect the onset of fatigue within ± 10 seconds error margin. Overall, it was found that the subjects experienced lower fatigue and had a higher probability of compliance and engagement with the proposed robotic rehabilitation system.
Authors: Varun Kanal (University of Texas at Arlington), Maher Abujelala (University of Texas at Arlington), James Brady (University of Texas at Arlington), Glenn Wylie (Rocco Ortenzio Neuroimaging Center Kessler Foundation), Fillia Makedon (University of Texas at Arlington),
Hide Authors & Abstract

Show Authors & Abstract
13:30 - 13:45
DeepFit: Deep Learning based Fitness Center Equipment Use Modeling and Prediction

In today’s busy modern life, modeling and accurately predicting fitness center equipment usage and availability is essential for improving human fitness and well-being. Therefore, in this paper, we develop DeepFit, a deep learning based system that predicts future fitness center equipment usage based on historical data. To this end, we design a Long Short Term Memory (LSTM) based sequence-to-sequence model that captures the dependencies in the data. The sequence-to-sequence model comprises of an encoder and a decoder, each of which separately is a deep Recurrent Neural Network (RNN). We evaluate DeepFit on equipment usage data collected from a university campus fitness center over a period of 1.5 years and demonstrate that it is able to accurately predict future fitness center equipment usage. We show that DeepFit outperforms the linear regression and ARIMA baselines in terms of Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) providing 17% performance improvement on average. We also investigate the benefits of augmenting the deep learning model in DeepFit with features such as whether the school is in session and month of the year and observe that the enhanced DeepFit system obtains performance improvements of 35% and 32% over linear regression and ARIMA, respectively. Our experiments show that the trained DeepFit model requires limited computational resources at test time, thus making it an attractive system for practical deployment.
Authors: Adita Kulkarni (Binghamton University), Anand Seetharam (Binghamton University), Arti Ramesh (Binghamton University),
Hide Authors & Abstract

Show Authors & Abstract
13:45 - 14:00
FitAssist: Virtual Fitness Assistant Based on WiFi

Regular exercise offers numerous health benefits and contributes to a healthy lifestyle. Doing exercise at home is an attractive choice for many people due to its convenience and low cost. Motivated by this, we propose FitAssist in this paper, a household virtual fitness assistant capable of performing fine-grained exercise recognition and exercise quality assessment based on commercial WiFi devices. Unlike wearable devices based systems, this system is more comfortable and device-free. In addition, compared to previous Wi-Fi based exercise monitoring systems, whose performance attenuates seriously when users stand out of the First Fresnel Zone (FFZ), FitAssist does not require users to stand on or near the line of sight (LoS) path. To achieve this, FitAssist extracts features from the fine-grained WiFi channel state information (CSI) and enables both exercise recognition and user identification via deep learning techniques. Moreover, FitAssist can provide personalized workout assessment to help users obtain effective workout and prevent injury. Extensive experimental results in real settings show that FitAssist achieves average accuracies of 97% and 98% for exercise recognition and user identification respectively, as well as giving accurate and useful feedback in various scenarios, which proves its effectiveness and robustness.
Authors: Yan Zhu (School of Software, Shanghai Jiao Tong University, Shanghai, China), Dong Wang (RFID and IOT lab, Shanghai Jiao Tong University, Shanghai, China), Run Zhao (Shanghai Jiao Tong University, Shanghai, China), Qian Zhang (RFID and IOT lab, Shanghai Jiao Tong University, Shanghai, China), Anna Huang (School of Software, Shanghai Jiao Tong University, Shanghai, China),
Hide Authors & Abstract

Show Authors & Abstract
14:00 - 14:15
RF-RVM: Continuous Respiratory Volume Monitoring With COTS RFID Tags

Continuous and accurate respiratory volume monitoring is quite helpful in many healthcare-related applications. Traditional respiratory volume monitoring approaches involve obtrusive devices which are uncomfortable for long-term monitoring, while unobtrusive approaches mainly focus on sensing the respiratory rate which is insufficient for many healthcare related applications. In this paper, we present RF-RVM, a wireless approach to sense the respiratory volume based on commercial off-the-shelf (COTS) RFID tags. In particular, RF-RVM continuously collects the temporal phase information from tags attached on the chest to extract the chest displacement caused by respiration. Then we assess the respiratory volume by training a BP neural network model between chest displacement and respiratory vol-ume. A reference tag under the user’s neck is used to eliminate the noise caused by slight movements of the upper body during respiration. We implement and evaluate RF-RVM using COTS RFID devices. Experimental results show that RF-RVM can continuously monitor user’s respiratory volume with a median accuracy from 88.4% to 92.1% for different chest sizes of different users.
Authors: Xiangmao Chang (Nanjing University of Aeronautics and Astronautics), XiaoXiang Xu (Nanjing Unversity of Aeronautics and Astronautics), Guoliang Xing (City University of Hong Kong), Linlin Tu (Michigan State University),
Hide Authors & Abstract

Show Authors & Abstract

Session 6 14:30 - 16:00

Mobile Networks
14:30 - 14:45
A Social-Relation-Based Game Model for Distributed Clustering in Cooperative Wireless Networks

In this paper, a novel framework for cluster detecting in cooperative wireless networks is proposed. This framework is modeled by a dynamic game with incomplete information. Each player in the game aspires to improve its position in the network by forming cooperative groups. Instead of static systems, the attention we paid in this paper is highly dynamic networks, in which the users' high mobility brings a huge challenge in clustering. In order to mitigate that impact, this paper proposes to analyze the mobile patterns of users for clustering from the perspective of social nature and designs measures to extract social nature of users. The introduction of social nature with generally long-term characteristics makes clustering framework more predictive and stable. Simulations on real-world networks show that the proposed approach can perform well in clustering in cooperative wireless networks.
Authors: Bowen Li (Institute of Information Engineering.CAS), Shunliang Zhang (Institute of Information Engineering.CAS), Xu Shan (Institute of Information Engineering.CAS), Yongming Wang (Institute of Information Engineering.CAS), Zhenxiang Gao (Department of Computer Science and Engineering, UConn),
Hide Authors & Abstract

Show Authors & Abstract
15:00 - 15:15
Efficient Content Caching for Named Data Network Nodes

Name Data Networking (NDN) is a promising Content-Centric Network (CCN) architecture that supports data distribution and data sharing by in-network ubiquitous content caching. In NDN, each router has content store to cache data packets passed by and, therefore, frequently requested content by consumers (e.g., end hosts) is cached at multiple routers in the network. Content caching at routers enables data delivery to consumers from a nearest location with minimal latency and thereby enhances overall network performance. Content store at nodes should have sufficient space to hold the current frame of locality of reference for attaining a good hit rate. The content store size requirement for each node is different due to their topological characteristics. Homogeneous caching mechanisms distribute the total cache budget equally among the nodes irrespective of their topological characteristics. In contrast, heterogeneous caching allocates cache to the nodes based on their topological importance. In this paper, a heterogeneous on-path cache budget distribution approach is proposed that distributes cache to the content stores based on reference locality of the nodes. The proposed cache distribution algorithm is evaluated for structured and unstructured network topology using the ndnSIM simulator. The results are compared with the homogeneous cache distribution mechanism and 14% improvement in cache hit rate is achieved.
Authors: Virendra Singh Shekhawat, Ankur Vineet (Birla Institute of Technology & Science, Pilani), Avinash Gautam (Birla Institute of Technology & Science, Pilani),
Hide Authors & Abstract

Show Authors & Abstract
15:15 - 15:30
Regression-based Network Monitoring in Swarm Robotic Systems

Mobile ad-hoc networks are becoming more common as robotic swarm technology becomes possible. One issue surrounding swarm technology is communication between robots. Communication costs time and energy, and can impact the performance of a swarm. In order to control the network, network state information must be acquired through network monitoring. We propose a novel REgression-based network Monitoring (REM) algorithm where robots in a swarm receive network state data only when necessary for the task at hand. This algorithm will save time and energy in communication by creating predictive models using regressions, minimizing network monitoring overhead.
Authors: Joshua Rands (Colorado School of Mines), Qi Han (Colorado School of Mines),
Hide Authors & Abstract

Show Authors & Abstract
15:30 - 15:45
Simulation of a big number of microservices in a highly distributed vast network.

Rapidly increasing numbers of instances of services, especially microservices, and a growing complexity of systems create uncertainty if existing algorithms and solutions for problems like service discovery or load balancing are still applicable. Building huge infrastructure for verification is often too costly and impractical in production environment therefore it is necessary to find other ways to test architectural approaches and algorithms. One of them is using simulators. To be able to simulate very big systems like telecommunication radio access network the simulator must have certain features: capability of simulating huge numbers of services, possibility to model heterogenic and vast network, facility to add new type of services and logic. This paper presents the idea of such simulator of services.
Authors: Tomasz Bosak (AGH University of Science and Technology), Piotr Jantos (Ericsson sp. z o.o.), Krzysztof Boryczko (AGH University of Science and Technology),
Hide Authors & Abstract

Show Authors & Abstract
15:45 - 16:00
YanuX - Pervasive Distribution of the User Interface by Co-located Devices

We currently live surrounded by many different computing devices. Therefore, it is important to take better advantage of those devices by coming up with smart ways of integrating and combining them. We have been exploring the possibility of building applications that present user interfaces pervasively distributed across different co-located devices. We designed the YanuX framework, which generalizes and supports the development of this new type of applications. A key issue is the automatic distribution of user interface (UI) components among co-located devices.We created the tools set needed to describe the capabilities of each of the devices present in the environment and the requirements of each of the applications’ components as configured by the developers. Restrictions of the components should match the capabilities of the devices leading to a UI component distribution decision that should reflect developers’ intentions and expectations of the users. Besides detailing YanuX’s components, the paper also presents YouTube Viewer as a proof-of-concept application based on YanuX. The application was also used in a user study to evaluate the concept and the experience supported by the framework. The results presented here are positive and very promising.
Authors: Pedro Santos (NOVA LINCS, DI, FCT, Universidade NOVA de Lisboa), Rui Madeira (Sustain.RD, ESTSetúbal, Instituto Politécnico de Setúbal), Nuno Correia (NOVA LINCS, DI, FCT, Universidade NOVA de Lisboa),
Hide Authors & Abstract

Show Authors & Abstract

Break 16:00 - 16:30

Session 7 16:30 - 17:30

Data Mining and Data Management
16:30 - 16:30
An Intelligent Optimization Method for Information Recommendation

Accurate information recommendation to users can significantly improve the service efficiency of recommendation platforms, and improve users’ experience of the client. With the rapid development of the internet, information recommendation platforms such as e-commerce are facing problems of data sparsity or algorithms are difficult to deal with large-scale calculation, which cause the recommended information cannot meet the needs of users. In order to improve the accuracy of information recommendation, the need of users is regarded as the evaluation basis of information recommendation, and an intelligent optimization method for information recommendation is proposed in this paper. Firstly, Resource Evaluation, User Evaluation and Scenario Evaluation are taken as main factors of information recommendation, and each evaluation algorithm for every factor is designed. Then, the weight of each factor is optimally allocated by the adaptive genetic algorithm. Finally, the multiplication of the quantized value of each factor and its weight is accumulated to get the comprehensive evaluation result, which is compared with the threshold value to decide whether to recommend the information to the user. Taking the e-commerce platform as an example, simulation results show that the intelligent optimization method for information recommendation proposed in this paper is effective and can significantly improve both accuracy and precision of information recommendation.
Authors: Shunan Ma (Institute of Information Engineering, Chinese Academy of Science), Xunbo Shuai (Research Institute of Petroleum Exploration &Development, PetroChina), Yongcai Chai (Research Institute of Petroleum Exploration &Development-Northwest (NWGI),PetroChina),
Hide Authors & Abstract

Show Authors & Abstract
16:30 - 16:45
A Generalized Mechanism beyond NLP for Real-Time Detection of Cyber Abuse through Facial Expression Analytics

Abuse in cyber space is a problem requiring immediate attention. Unfortunately, despite advances in Natural Language Processing techniques, there are clear limitations in detecting instances of cyber abuse today. Challenges arising due to different languages that teens communicate today, and usage of codes along with code mixing and code switching make the design of a comprehensive approach very hard. Existing NLP based approaches for detecting cyber abuse thus suffer from a high degree of false negatives and positives. To this extent, in this paper, we investigate a new approach to detect instances of cyber abuse. Our approach is motivated by the premise that abusers tend to have unique facial expressions while engaging in an actual abuse episode, and if we are successful, such an approach will be language-agnostic. Here, using only four carefully identified facial features without any language processing, and realistic experiments with 15 users, our system proposed in this paper achieves 98\% accuracy for same-user evaluation and up to 74\% accuracy for cross-user evaluation in detecting instances of cyber abuse.
Authors: Atanu Shome (Bangladesh University of Engineering and Technology), Md. Mizanur Rahman (Bangladesh University of Engineering and Technology), Sriram Chellappan (University of South Florida), A. B. M. Alim Al Islam (Bangladesh University of Engineering and Technology),
Hide Authors & Abstract

Show Authors & Abstract
16:45 - 17:00
An Augmented Real-world Interactive Classroom for Developing Learning Numbers and Counting Skills for Preschool Children

High quality preschool can have large beneficial effects on educational outcomes, including social, physical, intellectual, cognitive, and emotional development. In this paper, we propose an augmented real-world interactive learning-environment for preschool students based on Quick Response (QR) codes, mobile technologies, and pedagogical concepts. The proposed application attempts to promote learners’ interactive play with their learning environment. We evaluate our proposed learning using the Android SDK, which allows for preschool children to navigate and count a set of QR tagged learning objects using a 1-to-1 counting correspondence and the cardinal counting principles. Preliminary results provide evidence for potential benefits of an interactive learning experience in an augmented real-world classroom environment for fostering children’s number-learning and counting skills at preschool.
Authors: Ismail Abumuhfouz (Western Kentucky University), Yaser Mowafi (Western Kentucky University),
Hide Authors & Abstract

Show Authors & Abstract
17:00 - 17:15
SCARF: A Scalable Data Management Framework for Context Aware Applications in Smart Environments

This paper proposes a context data management framework for efficient context collection and evaluation to support context-aware applications. The proposed framework named SCARF leverages two architectures: a server-based centralized architecture and an edge-based architecture to address the problem of scalability of context collection and evaluation. SCARF collects the context data efficiently from resource constrained mobile/edge and multiple in-situ sensors (not resource constrained) deployed in smart environments. In particular, SCARF takes a set of contextual queries from applications and optimizes the rate at which the context data needs to be collected and/or transmitted from the sensor to the server. Furthermore SCARF decides whether the code for evaluating context should be executed on the server or on the edge. Execution on the edge will result in reduced communication between the server and edge but can result in low performance due to resource constraints of the edge. On the other hand, evaluation of context on the server, will require SCARF to transmit sensor data from the sensor to the server. SCARF is an adaptive framework which generates an optimal context acquisition plan and a context evaluation plan in such a way that the overall latency is minimized.
Authors: Eun-Jeong Shin (University of California, Irvine), Dhrubajyoti Ghosh (University of California, Irvine), Sharad Mehrotra (University of California, Irvine), Nalini Venkatasubramanian (University of California, Irvine),
Hide Authors & Abstract

Show Authors & Abstract
17:15 - 17:30
User-Connection Behaviour Analysis in Service Management Using Bipartite Labelled Property Graph

Digital transformation is continuously disrupting business models and service delivery. This has resulted in a transition from physical contact in traditional service delivery to digitized user interactions on service platforms. Gleaning insight from patterns of user connections to these services is important for effective service management. We first construct a user-state bipartite la-belled property graph model for user-connection behaviour analyses. Unlike previous works, our model is neither restricted by the pre-partitioning of data nor the pre-aggregation of edge weights. Using this model, we performed flexible versions of static analysis conduct in earlier research works. These ad hoc graph traversal user-connection analyses typical of real world business scenarios reveal patterns for the range of user interest, user interest intensity and service utilization. We also illustrate with an example of how to perform service recommendation. In addition, we extend our model by enriching a one mode user projection of the bipartite network with connected features to predict user connection behaviour. Our approach is effective for the modelling of large user behavioural data sets. The method of analysis is suitable for flexible and expressive real-time analytics in service management in domains such data publishing, mobile service usage and telecommunications.
Authors: Michael Kpiebaareh (University of Electronic Science and Technology of China), Wei-Ping Wu (University of Electronic Science and Technology of China), Strato Bayitaa (University of Electronic Science and Technology of China), Charles Haruna (University of Electronic Science and Technology of China), Lawrence Tandoh (University of Electronic Science and Technology of China),
Hide Authors & Abstract

Show Authors & Abstract
Room #2

Workshop 10:30 - 12:10

UAV Swarm Networking for Mobile and Ubiquitous Systems
10:30 - 10:55
Performance Evaluation of Wi-Fi for Underground Robots

Underground environments present their unique challenges for wireless communication. This paper presents an empirical study of WiFi performance where aerial-ground robots are used to map, navigate, and search in an unknown underground environment. While wireless signal attenuates significantly around corners, WiFi's overall performance is encouraging.
Authors: Lixiao Zhu (Colorado School of Mines), Xuejin Wen (Colorado School of Mines), Qi Han (Colorado School of Mines), Alexander Fryer (Colorado School of Mines), Neil Suttora (Colorado School of Mines), Andrew Petruska (Colorado School of Mines),
Hide Authors & Abstract

Show Authors & Abstract
10:55 - 11:20
Channel Prediction Based on Adaptive Structure Extreme Learning Machine for UAV mmWave Communications

In unmanned aerial vehicle (UAV) millimeter wave (mmWave) communications, the inter-UAV wireless channel is fast varying because the high mobility of the UAV transmission platform. In such dynamic scenarios, it is very costly to obtain the inter-UAV channel state information (CSI) with the conventional pilot-aided channel estimation. Aiming to address this critical issue, in this paper, we propose a novel adaptive-structure extreme learning machine (ASELM) enabled fast channel predication to obtain the CSI in a proactive fashion, which can further support agile beam-based inter-UAV mmWave communication. In particular, ASELM copes with the channel variations by adaptively adjusting the number of neurons in the hidden-layer of ELM. Moreover, a sliding window prediction mechanism (SWPM) predicts subsequent-CSI by efficiently reuses the predicted concurrent-CSI to train the ASELM, which is able to save the pilot overheard for channel sampling (estimation) towards longer-range channel prediction and improve prediction accuracy at affordable costs. Simulation results show that the proposed ASELM enabled fast channel predication can achieve lower normalized mean square error than traditional prediction algorithm in the considered inter-UAV mmWave communication scenarios.
Authors: Hongxing Zhang (Beijing University of Posts and Telecommunications), Hui Gao (Beijing University of Posts and Telecommunications), Xin Su (Tsinghua University),
Hide Authors & Abstract

Show Authors & Abstract
11:20 - 11:45
Constructing 3D Maps for Dynamic Environments using Autonomous UAVs

This paper presents the implementation of an Unmanned Aerial Vehicle (UAV), which can navigate autonomously in dynamic environments. The goal of the project is to minimize the risks to workers' safety by deploying UAVs to inaccessible places which are frequently found in the Oil & Gas Industry such as confined pipelines. The autonomous UAV can fly through a series of pipes to generating a 3D map of the flight path. We used Light Detection and Ranging (LIDAR) technology to map the surrounding environment as the UAV flies through the environment. The feedback from the LIDAR sensors is used for real-time autonomous navigation and obstacle avoidance. The route is also logged for subsequent navigation. As a UAV navigates the environment, it records a video of all it sees which can then be watched by the maintenance engineers. Our approach involves running a simulation using the Robotics Operating System (ROS) to assert and fine-tune our navigation algorithms before applying them directly to the physical hardware. At this stage, we have successfully implemented the autonomous navigation using LIDAR scanners in the ROS simulation environment. We also implemented an algorithm to manage the battery life of the UAV through which it can use to return home when the battery level drops down to a certain percentage. We expect that this research will help autonomous UAVs to safely navigate new spaces by themselves in different domains such as in industrial maintenance and rescue operations.
Authors: Ahmed Abdelmoamen Ahmed (Prairie View A&M University), Abel Olumide (Prairie View A&M University), Adeoluwa Akinwa (Prairie View A&M University), Mohamed Chouikha (Prairie View A&M University),
Hide Authors & Abstract

Show Authors & Abstract
11:45 - 12:10
A Novel Routing Metric for IEEE 802.11s-based Swarm-of-Drones Applications

With the proliferation of drones in our daily lives, there is an increasing need for handling their numerous challenges. One of such challenge arises when a swarm-of-drones are deployed to accomplish a specific task which requires coordination and communication among the drones. While this swarm-of-drones is essentially a special form of mobile ad hoc networks (MANETs) which has been studied for many years, there are still some unique requirements of drone applications that necessitates revisiting MANET approaches. These challenges stem from 3-D environments the drones are deployed in, and their specific way of mobility which adds to the wireless link management challenges among the drones. In this paper, we consider an existing routing standard that is used to enable meshing capability among Wi-Fi enabled nodes, namely IEEE 802.11s and adopt its routing capabilities for swarm-of-drones. Specifically, we propose a link quality metric called SrFTime as an improvement to existing Airtime metric which is the 802.11s default routing metric to enable better network throughput for drone applications. This new metric is designed to fit the link characteristics of drones and enable more efficient routes from drones to their gateway. The evaluations in the actual 802.11s standard indicates that our proposed metric outperforms the existing one consistently under various conditions.
Authors: Oscar Bautista Chia (Florida International University), Nico Saputro (Parahyangan Catholic University), Kemal Akkaya (Florida International University), Selcuk Uluagac (Florida International University),
Hide Authors & Abstract

Show Authors & Abstract
Day 3 14/11/2019
Room #1

Coffee break ready 07:30 - 07:45

Bus pick-up at the hotel 07:45 - 08:00

Pick up at the hotel for NASA social program

Social Program (NASA) & Networking 09:00 - 13:00

Lunch 13:00 - 14:00

Session 8 14:00 - 15:30

Localization and Tracking
14:00 - 14:00
DeepLoc: Deep Neural Network-based Telco Localization

Recent years have witnessed unprecedented amounts of telecommunication (Telco) data generated by Telco radio and core equipment. For example, measurement records (MRs) are generated to report the connection states, e.g., received signal strength at the mobile device, when mobile devices give phone calls or access data services. Telco historical data (e.g., MRs) have been widely analyzed to understand human mobility and optimize the applications such as urban planning and traffic forecasting. The key of these applications is to precisely localize outdoor mobile devices from these historical MR data. Previous works calculate the location of a mobile device based on each single MR sample, ignoring the sequential and temporal locality hidden in the consecutive MR samples. To address the issue, we propose a deep neural network (DNN)-based localization framework namely DeepLoc to ensemble a recently popular sequence learning model LSTM and a CNN. Without skillful feature design and post-processing steps, DeepLoc can generate a smooth trajectory consisting of accurately predicted locations. Extensive evaluation on 6 datasets collected at three representative areas (core business, urban and suburban areas in Shanghai, China) indicates that DeepLoc greatly outperforms 10 counterparts.
Authors: Yige Zhang (School of Software Engineering, Tongji University), Yu Xiao (Department of Communications and Networking, Aalto University), Kai Zhao (Robinson College of Business, Georgia State University), Weixiong Rao (School of Software Engineering, Tongji University),
Hide Authors & Abstract

Show Authors & Abstract
14:00 - 14:15
A scheme for Anomalous RFID Trajectory Detection based on Improved Clustering Algorithm under Digital-Twin-Driven

Anomaly analysis of trajectories is one of the means to maintain indoor safety. Effective track anomaly detection should be based on the current road network structure. However, the indoor environment usually contains many obstacles flexible to move. The changes in the positions of obstacles frequently cause the changes of the road network structure. When the road network changes, it takes time and effort to manually redraw the road network and simultaneously losses the guarantee of real-time performance. At the same time, manual drawing is forbidden in some confidential places. In our paper, the scheme of RFID track anomaly detection combined with digital-twin technology is proposed to provide a real-time actual road network for anomaly analysis. The accurate mapping and virtual simulation functions of digital-twin technology are used to achieve the dynamic real-time rendering and the maintenance of the indoor road network structure. We also improve a clustering algorithm to make it suitable for indoor RFID track clustering. In our paper, deviation thresholds and digital-twin model are used to detect anomalies. The trajectory is judged to be abnormal when it appears in the restricted areas or exceeds the deviation thresholds in any terms of position, velocity, or direction. We use an improved scan-line algorithm and F1-scores to determine the threshold values. Our proposed anomaly detection method provides more effective and intuitive results for researchers to analyze.
Authors: Mengnan Cai (Institute of Information Engineering, Chinese Academy of Sciences), Siye Wang (Institute of Information Engineering, Chinese Academy of Sciences), Xinling Shen (Institute of Information Engineering, Chinese Academy of Sciences), Yijia Jin (The Boeing Company),
Hide Authors & Abstract

Show Authors & Abstract
14:15 - 14:30
Indoor localization based on subcarrier parameter estimation of LoS with Wi-Fi

With the wide application of MIMO-OFDM technology, Channel State Information (CSI) as a fine-grained feature can be extracted from PHY layer with Wi-Fi. Although CSI has a better performance on expressing the spatial and temporal features of wireless signal, it is more sensitive to the multipath reflection. As a result, Line-of-Sight (LoS) identification and corresponding subcarrier parameter estimation play an important role in improving positioning accuracy. In this paper, we propose a complete parameter processing framework, which involves phase calibration, phase ambiguity elimination, subcarrier parameter (amplitude and phase) estimation of LoS, fingerprint feature extraction and relationship mapping from fingerprint feature to position estimate. The experimental results show that, compared with existing algorithm, our proposed algorithm improves the positioning accuracy by 2.3% in LoS and 10.7% in NLoS cases.
Authors: Bobai Zhao (Institute of Information Engineering, Chinese Academy of Sciences), Dali Zhu (Institute of Information Engineering, Chinese Academy of Sciences), Tong Xi (Institute of Information Engineering, Chinese Academy of Sciences), Siye Wang (Institute of Information Engineering, Chinese Academy of Sciences), Di Wu (Institute of Information Engineering, Chinese Academy of Sciences),
Hide Authors & Abstract

Show Authors & Abstract
14:30 - 14:45
MMLOC: Multi-Mode Indoor Localization System Based on Smart Access Points

Indoor localization based on Wi-Fi fingerprints has been an active research topic for years. However, existing approaches do not consider the instability of access points (APs) which may be unreliable in practice, particularly the ones deployed by individual users. This instability impacts the localization accuracy severely, due to the unreliable or even wrong Wi-Fi fingerprints. Ideally, the localization should be done using only the well-deployed APs (e.g., deployed by facility teams). However, in many places the number of these APs is too few to achieve a good localization accuracy. To solve this problem, we leverage emerging smart APs equipped with multi-mode antennas, and build a new indoor localization system called MMLOC to reduce the number of necessary APs. The key idea is controlling the modes of AP antennas to generate more fingerprints with fewer APs. A clustering based localization strategy is designed to enable a mobile terminal to figure out the RSSI (Received Signal Strength Indicator) for different antenna modes without requiring any synchronization. We have implemented a prototype system using smart APs and commercial smartphones. Experimental results demonstrate that MMLOC can reduce the number of necessary APs by 50%, and achieve the same or even better localization accuracy.
Authors: Tuo Yu (University of Illinois at Urbana-Champaign), Wenyu Ren (University of Illinois at Urbana-Champaign), Klara Nahrstedt (University of Illinois at Urbana-Champaign),
Hide Authors & Abstract

Show Authors & Abstract
14:45 - 15:00
Multi-Destination Vehicular Route Planning with Parking and Traffic Constraints

This paper aims to provide an efficient solution for people in a city who drive their cars to visit several destinations, where they need to park for a while, but do not care about the visiting order. This instance of the multi-destination route planning problem is novel in terms of its constraints: the real-time traffic conditions and the real-time free parking conditions in the city. The paper proposes a novel Multi-Destination Vehicle Route Planning (MDVRP) system to optimize the travel time for all drivers. MDVRP's design has two components: a mobile app running on the drivers' smart phones that submits real-time route requests and guides the drivers toward destinations, and a server in the cloud that optimizes the routes by finding the most efficient order to visit the destinations. MDVRP uses TDTSP-FPA, an algorithm that finds the fastest route to the next destination and also assigns free curbside parking spaces that minimize the total travel time for drivers. We evaluate MDVRP using a driver trip dataset that contains real vehicular mobility traces of over two million drivers from the city of Cologne, Germany. By learning the spatio-temporal distribution of real driver destinations from this dataset, we build a novel experimental platform that simulates real, multi-destination driver trips. Extensive simulations executed over this platform demonstrate that TDTSP-FPA delivers the best performance when compared to three baseline algorithms.
Authors: Abeer Hakeem (Department of Computer Science, New Jersey Institute of Technology), Narain Gehani (Department of Computer Science, New Jersey Institute of Technology), Xiaoning Ding (Department of Computer Science, New Jersey Institute of Technology), Reza Curtmola (Department of Computer Science, New Jersey Institute of Technology), Cristian Borcea (Department of Computer Science, New Jersey Institute of Technology),
Hide Authors & Abstract

Show Authors & Abstract
15:00 - 15:15
PerFlow: Configuring the Information Flow in a Pervasive Middleware via Visual Scripting

The plethora of ubiquitous information devices create smart environments with new ways to interact. Environments such as pervasive classrooms or smart offices provide multiple services and connect a variety of heterogeneous and mostly mobile devices. Besides all the advantages of this seamless communication, it is a complex task to control the information flow within these environments. We propose PerFlow, a middleware that allows to configure the information flow within a pervasive environment at runtime. The middleware is designed to enable technically unskilled personnel, such as a lecturer in a classroom might be, to define rules about who is able to send what to whom. Therefore, we have implemented a visual scripting tool into PerFlow. Our evaluation shows that PerFlow enforces the information flow within a pervasive environment with no noticeable overhead. Further, we demonstrate the usability of the visual scripting tool in a two-stage user study.
Authors: Jens Naber (University of Mannheim), Martin Pfannemüller (University of Mannheim), Janick Edinger (University of Mannheim), Christian Becker (University of Mannheim),
Hide Authors & Abstract

Show Authors & Abstract

Break 15:30 - 16:00

Session 9 16:00 - 17:30

Sensing and Energy Optimization
16:00 - 16:15
PatientSense: Patient Discrimination from In-Bottle Sensors Data

Accurately accounting for medication use is important for the efficacy and safety of patients and family members. Monitoring is also important for medication adherence. This work investigates identification of persons taking medication using a sensor-equipped pill bottle. The bottle is equipped with inertial and switch sensors in both the cap and body, making the added hardware unobtrusive, low-cost, and wireless. Our system uses inertial data to build a patient discrimination model using classification techniques. We evaluated the system using 16 subjects. Our results show that using binary Support Vector Machine (SVM), the system can discriminate one patient among 16 subjects with 94% accuracy, and has a 93% using a single sensor. Identifying the exact person in a set of 3 subjects has an accuracy higher that 91%.
Authors: MURTADHA ALDEER (Rutgers The State University of New Jersey), Jorge Ortiz (Rutgers the state university of new jersey), Richard Howard (Rutgers the state university of new jersey), Richard Martin (Rutgers the state university of new jersey),
Hide Authors & Abstract

Show Authors & Abstract
16:15 - 16:30
Distances for WiFi Based Topological Indoor Mapping

For localization and mapping of indoor environments through WiFi signals, locations are often represented as likelihoods of the received signal strength indicator. Different measures are proposed in the literature for determining the similarity of these likelihoods. They are usually evaluated in studies with specific settings. In this work we compare, in a daily-life setting, various measures of distance between such likelihoods in combination with different methods for estimation and representation. In particular, we show that among the considered distance measures the Earth Mover’s Distance is the most beneficial for the localization task.
Authors: Bastian Schäfermeier (L3S Hannover), Tom Hanika (Humboldt-Universität zu Berlin), Gerd Stumme (Universität Kassel),
Hide Authors & Abstract

Show Authors & Abstract
16:30 - 16:45
Examining the Energy Impact of Sorting Algorithms on Android: An Empirical Study

With the advent of mobile application development a new software quality concern - energy consumption - was introduced. For mobile software developers knowledge about software and algorithm design choices and their implications on energy consumption are crucial. However, software developers either lack this knowledge or tools to support them in estimating the energy consumption of their applications and therefore are unable to reflect on their design choices. In this empirical study we examine the energy consumption of 12 sorting algorithms and the resulting energy impact when used with different data-types. We propose a methodology to obtain energy readings and relate them to application execution traces. Our results show that the choice of data-type together with algorithm design can have significant impact on the energy profile of an application.
Authors: Andreas Schuler (Advanced Information Systems and Technology (AIST), University of Applied Sciences Upper Austria), Gabriele Anderst-Kotsis (Department of Telecooperation, Johannes Kepler University Linz),
Hide Authors & Abstract

Show Authors & Abstract
16:45 - 17:00
On Optimization of Ad-blocking Lists for Mobile Devices

Online advertisements and third-party web tracking has gained much attention in recent years. Advertisers gather as much data and information about the users to provide targeted advertisement. Though this leads to better user experience, it comes at the cost of privacy intrusive tracking. To this end, ad-blocking lists (or filter-lists, blacklists) have been introduced which prevent third-party tracking. Ad-blocking lists operate in a crowd-sourced manner, where new tracking domains (or rules) are continuously added by privacy activists and the redundant domains are discarded from the filter-list. Over time, the number of rules added out grow the number of rules omitted, becoming hard to manage the filter-lists. We observe that the filter-lists mostly detect different ad and tracking domains, and the number of rules that are used by filter-lists on Alexa top 5,000 websites are less than 1%. This suggests the need to curate optimized filter-lists that provide high coverage and require less time to scan for a given domain. We develop an aggregated and filtered blacklist that is more than 15,000% less bulky, and provides same coverage as the union of the blacklists on Alexa top 5,000 websites. We also develop an update mechanism to incorporate new ad and tracking domains in the blacklist in a resource efficient manner.
Authors: Saad Hashmi (Macquarie University), Muhammad Ikram (Macquarie University), Stephen Smith (Macquarie University),
Hide Authors & Abstract

Show Authors & Abstract
17:00 - 17:15
Participatory Location Fingerprinting through Stationary Crowd in a Public or Commercial Indoor Environment

The training phase of indoor location fingerprinting has been traditionally performed by dedicated surveyors in a manner that is time and labour intensive. Crowdsourcing process is more efficient, but is impractical in public or commercial buildings because it requires occasional location fix provided explicitly by the participant, the availability of an indoor map for correlating the traces, and the existence of landmarks throughout the area. Here, we address these issues for the first time in this context by leveraging the existence of stationary crowd that have timetabled roles, such as desk-bound employees, lecturers and students. We propose a scalable and effortless positioning system in the context of a public/commercial building by using Wi-Fi sensor readings from its stationary occupants' smartphones combined with their timetabling information. Most significantly, the entropy concept of information theory is utilised to differentiate between good and spurious measurements in a manner that does not rely on the existence of known trusted users. Our analysis and experimental results show that, regardless of such participants' unpredictable behaviour, including not following their timetabling information, hiding their location or purposefully generating wrong data, our entropy-based filtering approach ensures the creation of a radio-map incrementally from their measurements. Its effectiveness is validated experimentally with two well-known machine learning algorithms.
Authors: A. K. M. Mahtab Hossain (University of Greenwich), George Loukas (University of Greenwich),
Hide Authors & Abstract

Show Authors & Abstract
17:15 - 17:30
Time-Aware Reactive Storage in Wireless Edge Environments

Nowadays, smart mobile devices generate huge amounts of data in all sorts of gatherings. Much of that data has localized and ephemeral interest, but can be of great use if shared among co-located devices. However, in those scenarios, mobile devices often experience poor connectivity, leading to availability issues if applications' storage and logic are fully delegated to a remote cloud infrastructure. The edge computing paradigm pushes computations and storage beyond the data center, closer to the end-user devices where data is generated and consumed. Thus, enabling the execution of certain components of edge-enabled systems directly and cooperatively on edge devices. In this paper, we address the challenge of supporting reliable and efficient data storage and dissemination among co-located wireless mobile devices without resorting to centralized services or network infrastructures. We propose THYME, a novel time-aware reactive data storage system for wireless edge networks, that exploits synergies between the storage substrate and the publish/subscribe paradigm. We present the design of THYME and evaluate it through simulation, characterizing the scenarios best suited for its use. The evaluation shows that THYME allows for reliable notification and retrieval of relevant data with low overhead and latency.
Authors: João Silva (NOVA LINCS - Universidade NOVA de Lisboa), Hervé Paulino (NOVA LINCS - Universidade NOVA de Lisboa), João Lourenço (NOVA LINCS - Universidade NOVA de Lisboa), João Leitão (NOVA LINCS - Universidade NOVA de Lisboa), Nuno Preguiça (NOVA LINCS - Universidade NOVA de Lisboa),
Hide Authors & Abstract

Show Authors & Abstract

Concluding Remarks 17:30 - 18:00