IoTBDS 2021 Abstracts

Area 1 - Big Data Research

Full Papers

Paper Nr:	24
Title:	Research on Optimization of 4G-LTE Wireless Network Cells Anomaly Diagnosis Algorithm based on Multidimensional Time Series Data
Authors:	Bing Qian, Chong Ma and Tong Zhang
Abstract:	With the continuous increase of network terminal equipment, the operation scenarios of 4G-LTE wireless networks are becoming more and more complex. The traditional manual method of analysis and screening of network cell equipment can no longer meet the needs of production. Therefore, an efficient wireless network cell abnormality diagnosis algorithm is needed to screen abnormalities of equipment to improve operation and maintenance efficiency. In view of the fact that the existing single-dimensional anomaly diagnosis algorithm cannot achieve fully automated detection and the existing multidimensional anomaly diagnosis algorithm has low detection efficiency on multidimensional time series data, there are a large number of errors and omissions. This paper proposes a multidimensional time series data based on 4G-LTE wireless network cell anomaly diagnosis optimization algorithm uses small-sample supervised algorithms to assist the training of massive-sample unsupervised algorithms, thereby improving the detection performance of unsupervised learning algorithms. This paper verifies the effectiveness of the optimization algorithm through experiments, and has a great improvement in the four commonly used unsupervised algorithms, which can well improve the anomaly detection capabilities of the existing algorithms.
Download

Paper Nr:	51
Title:	C*DynaConf: An Apache Cassandra Auto-tuning Tool for Internet of Things Data
Authors:	Lucas B. Dias, Dennis S. Silva, Rafael D. Sousa Junior and Maristela Holanda
Abstract:	Internet of Things environments may generate massive volumes of time series data, with specific characteristics that must be considered to facilitate its storage. The Apache Cassandra NoSQL database provides compaction strategies that improve data pages’ organization, benefiting the storage and query performance for time series data. This study exploits the temporal characteristics of IoT data, and proposes an engine called C*DynaConf based on the TWCS (Time Window Compaction Strategy), which dynamically changes its compaction parameters according to configurations previously defined as optimal, considering current metadata and metrics from the database. The results show that the engine’s use brought a 4.52% average gain in operations performed compared to a test case with optimal initial configuration that changes the scenario’s characteristics change over time.
Download

Short Papers

Paper Nr:	23
Title:	Challenges in Data Acquisition and Management in Big Data Environments
Authors:	Daniel Staegemann, Matthias Volk, Akanksha Saxena, Matthias Pohl, Abdulrahman Nahhas, Robert Häusler, Mohammad Abdallah, Sascha Bosse, Naoum Jamous and Klaus Turowski
Abstract:	In the recent years, the term big data has attracted a lot of attention. It refers to the processing of data that is characterized mainly by 4Vs, namely volume, velocity, variety and veracity. The need for collecting and analysing big data has increased manifolds these days as organizations want to derive meaningful information out of any data that is available and create value for the business. A challenge that comes with big data is inferior data quality due to which a lot of time is spent on data cleaning. One prerequisite for solving data quality issues is to understand the reasons for their occurrence. In this paper, we discuss various issues that cause reduced quality of the data during the acquisition and management. Furthermore, we extend the research to categorize the quality of data with respect to the identified issues.
Download

Paper Nr:	38
Title:	Approximate Query Processing for Lambda Architecture
Authors:	Aleksey Burdakov, Uriy Grigorev, Andrey Ploutenko and Oleg Ermakov
Abstract:	The lambda architecture is widely used to implement streaming data processing systems. These systems create batch views (subsets of data) at the Serving Layer to speed up queries. This operation takes significant time. The article proposes a novel approach to lambda architecture implementation. A new method for Approximate Query Processing in a system with Lambda Architecture (LA-AQP) significantly reduces aggregate (sum, count, avg) calculation error. This is achieved by using a new way of calculating the reading segments probabilities. The developed method is compared with the modern Sapprox method for processing large distributed data. Experiments demonstrate that LA-AQP almost equals Sapprox in terms of volume and time characteristics. The introduced accuracy measures (δ-accuracy and ε-accuracy) are up to two times better than Sapprox for total aggregate calculation. Aggregate values can vary greatly from segment to segment. It is shown that in this case the LA-AQP method gives a small error in the total aggregate calculation in contrast to Sapprox.
Download

Paper Nr:	59
Title:	Homicide Network Detection based on Social Network Analysis
Authors:	Victor Chang, Yeqing Mou, Qianwen A. Xu, Harleen Kaur and Ben S. Liu
Abstract:	This paper aims to explore the use of social network analysis in identifying the most active suspects and possible crime gangs in the network. The homicide dataset provided by White & Rosenfeld is employed and both the victim network and suspects network are structured by the use of Rstudio. This paper finds that the criminal gang and group of victims in homicide cases could be investigated by conducting centrality analysis and detecting cliques in these two one-mode networks. Moreover, the same features of victims or suspects are significant indicators for distinguishing and discovering victim groups or criminal gangs. As suspects or victims with the same features will be gathered into the same community in the community analysis of SNA, it is more effective to identify victim groups or criminal gangs by analyzing their characteristics, so that crimes can be resolved more efficiently or even prevented.
Download

Paper Nr:	5
Title:	Customer Relationship Management Improvement using IoT Data
Authors:	Christian Ploder, Reinhard Bernsteiner, Thomas Dilger and Sarah Huber
Abstract:	The Internet of Things (IoT) increasingly gains importance and costumers ale willing to pay for. Studies show that by 2020, more than 30 billion devices will be connected and the IoT platform market will grow to $ 7.6 billion in 2024. The purpose of this paper is to determine how IoT data could have a positive impact on customer relationship management (CRM). An empirical study has been conducted based on qualitative research methods with twelve experts in 2020 specialized in innovation marketing or CRM who have already participated in IoT projects in the retail industry. The results demonstrate that companies will be able to satisfy the customer’s needs in a more precise way and that it is possible to predict the customer’s behavior by analyzing generated data. Furthermore for most companies it is sufficient to implement a standardized CRM system because of their lack of knowledge in software development and interfacing opportunities. In this way, collected IoT data of the individual can be aggregated with already generated data from all other channels. Through this alignment, a holistic customer understanding about the purchased products, services, and wishes will be acquired and marketing activities can be targeted accordingly.
Download

Paper Nr:	13
Title:	A Comparison among Wi-Fi Direct, Classic Bluetooth, and Bluetooth Low Energy Discovery Procedures for Enabling Massive Machine Type Communications
Authors:	Abel R. Medel and Jose C. Brito
Abstract:	The exponential growth of the Internet of Things (IoT) devices bring the necessity to support massive Machine Type Communications (mMTC) for the Next-Generation networks. One of the enablers for mMTC is the Device-to-Device (D2D) communications. Since it is not possible to reduce the number of devices in massive communications, all the effort for collision and power consumption reduction should be applied to the discovery algorithm of the D2D technologies. The principal D2D technologies are Wi-Fi Direct, Classic Bluetooth, and Bluetooth Low Energy (BLE). However, most of these technologies were not originally designed to support massive communications. The main goal of this work is to assess Wi-Fi Direct, Classic Bluetooth, and BLE performances in terms of number of collisions, energy consumption, and discovery latency, in order to check out the more suitable technology for mMTC scenarios. The results show that Classic Bluetooth is faster than Wi-Fi Direct during the devices’ discovery, accelerating network access for the devices in massive communications. Besides, BLE incurs fewer collisions, less energy consumption, and less time for devices’ discovery than Classic Bluetooth. Thus, BLE is the more suitable D2D technology—out of the three analyzed in this work—to enable mMTC.
Download

Paper Nr:	18
Title:	An Ontological Approach to the Formation of an Excursion Route by Heritage Objects in GIS
Authors:	Andrii Honchar, Maryna Popova and Rina Novogrudska
Abstract:	The paper presents an ontological approach to the consolidation of heritage objects 3D models and geoinformation systems for the virtual excursion routes formation. Given research purpose is to provide the opportunity for free mass access to the digitized heritage of the world civilization using augmented and virtual reality in three-dimensional space. Modern approaches and the most common software for 3D modelling of heritage objects are analysed. Software solutions for data 3D representation and analysis in GIS are listed. It is shown that the most suitable for practical implementation is GIS with integrated 3D panoramas of heritage objects. An ontological approach to the consolidation of multi-format data is described, it provides a solution to the problems of heterogeneity and interoperability of transdisciplinary distributed information resources that describe heritage objects. An ontological model of an excursion route is presented; its taxonomy is implemented in the form of a graph.
Download

Paper Nr:	31
Title:	Analysis of the Message Queueing Telemetry Transport Protocol for Data Labelling: An Orthopedic Manufacturing Process Case Study
Authors:	Mangolika Bhattacharya, Reenu Mohandas, Mihai Penica, Mark Southern, Karl Vancamp and Martin J. Hayes
Abstract:	The recent paradigm shift in the industrial production systems, known as Industry 4.0, changes the work culture in terms of human machine interaction. Human labours are assisted by smart devices and machines as in human-machine cooperation and human-machine collaboration. For enhancing this process, data processing and analyses are needed. Therefore, data collection has become one of the most essential functions of large organizations. In this work, a data engineering experiment for a grinding process within a commercial orthotics manufacturing company is presented. The data collection and labelling is assessed for time stamp latency using the Message Queuing Telemetry Transport (MQTT) protocol. This step is necessary to determine if alarm prediction or ‘front running’ is feasible. The paper analyses the procured dataset and discusses its merits as an alarm predictor, using sparsity indicators and concludes that a new investment in sensor infrastructure is necessary. This work highlights some of the limits of performance that exist for the use of MQTT with existing sensor infrastructure when retrofitting machine learning based alarm prediction in an industrial use case setting. A road-map for potential solution to this problem is provided which needs to be assessed by the company management before further progress can be made.
Download

Paper Nr:	33
Title:	Role of Artificial Intelligence of Things (AIoT) in Covid-19 Pandemic: A Brief Survey
Authors:	Venkatesh K. Pappakrishnan, R. Mythili, V. Kavitha and N. Parthiban
Abstract:	Digital twins, Internet of Things (IoT) and Artificial Intelligence (AI), plays a proactive role in numerous ways during a pandemic such as COVID-19 by allowing us to make informed decisions using real-time data. According to World Health Organization (WHO), COVID-19 is an infectious disease caused by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) that predominantly spreads through body fluids, leading to a mild-to-severe respiratory illness. Considering the global health crisis due to COVID-19 and novelty of the SARS-CoV-2 virus due diligence is required in vaccine preparation and human trials. At the early stages of the pandemic, due to lack of complete knowledge on the virus, there are two main objectives: (1) treat patients as effectively as possible and (2) control the spread of the disease. IoT devices in healthcare empower the healthcare industry in identifying potential carriers of COVID-19 and quarantine. Even though IoT plays a major role in healthcare 4.0, decision making capabilities are limited due to the type of the algorithms and decision making paradigms used. Using AI, we will be able to identify critical medical conditions earlier and take necessary steps. Artificial Intelligence of Things (AIoT) implementation has the potential to greatly reduce the mortality rate allowing us in early identification of high-risk patients, monitoring the spread of the disease, methods to limit the spread, predict mortality risk by analyzing patient’s health history, remote or in-home treatments to reduce hospital occupancy, and other techniques to significantly control the spread and treat the patients effectively.
Download

Paper Nr:	39
Title:	Analysis Layer Implementation Method for a Streaming Data Processing System
Authors:	Aleksey Burdakov, Uriy Grigorev, Andrey Ploutenko and Oleg Ermakov
Abstract:	Analysis is an important part of the widely used streaming data processing. The frequency of flow element occurrence and their values sum are calculated during analysis. The algorithms like Count-Min Sketch and others give a big error in restoring the aggregate with a large number of elements. The article proposes application of a vector matrix. Each vector has a length of 'n'. If the number of different elements approaches 'n', then the window size is automatically reduced. This allows accurate storage of the aggregate without element loss. The SELECT operator for searching in a vector array is also proposed. It allows getting various slices of the aggregated data accumulated over the window. The comparison of the developed method with the Count-Min Sketch data processing method in the Analysis Layer was performed. The experiment showed that the method based on the vector matrix more than twice reduces memory consumption. It also ensures the exact SELECT statement execution. An introduction of a floating window allows maintaining the calculation accuracy and avoiding losing records from the stream. The same query sketch-based execution error reaches 200%.
Download

Paper Nr:	43
Title:	Smart Cities V2I Cloud based Infrastructure using Road Side Units
Authors:	Tamer Omar, Kevin Guerra, Christopher Mardoyan, Shannen Sharma and Xavier Rangel
Abstract:	Vehicles and city infrastructure can be interconnected through Road Side Units (RSUs) and On Board Units (OBUs) that utilize Radio-frequency identification (RFID) technology to send and receive information about various road conditions in real time. The objective of this work is to create and test a mesh network through Internet of Things (IoT) devices to emulate and test the RSUs capabilities. A vehicle-to-infrastructure (V2I) network is established through parent & child approach that rely on previously established infrastructure. The purpose for this design is to extend the reach of the system while limiting the amount of endpoint nodes needed. The mesh network is meant to target areas affected with road congestion. Using this information, the vehicles and users will be aware of traffic conditions on their routes in real time. Connected vehicles will be able to adjust their routes to experience more efficient commutes. The mesh network is capable of taking information from vehicles and transmitting it through the network until being uploaded to the cloud. In particular, the number of vehicles passing through an endpoint RSU within a certain time frame is collected and sent through the network, along with the location of the endpoint RSU. The parent node receives this information through a relay RSUs and uploads it to a cloud service where the data is collected and then analyzed through a data mining software. The software applies the k-means clustering algorithm to classify the traffic conditions of the road at a particular time. Results shows the capability of the algorithm to detect and classify the different traffic conditons.
Download

Paper Nr:	49
Title:	A Preliminary Overview of the Situation in Big Data Testing
Authors:	Daniel Staegemann, Matthias Volk, Matthias Pohl, Robert Häusler, Abdulrahman Nahhas, Mohammad Abdallah and Klaus Turowski
Abstract:	Due to the constantly increasing amount and variety of data produced, big data and the corresponding technologies have become an integral part of daily life, influencing numerous domains and organizations. However, because of its diversity and complexity, the necessary testing of the corresponding applications is a highly challenging task that lacks maturity and is still being explored. While there are numerous publications dealing with this topic, there is no sufficiently comprehensive overview to conflate those isolated pieces of information to a coherent knowledgebase. The publication at hand highlights this grievance by means of an unstructured literature review, proposes a starting point for a corresponding taxonomy to bridge this gap and highlights future avenues for research.
Download

Area 2 - Emerging Services and Analytics

Full Papers

Paper Nr:	56
Title:	Deep Learning for COVID-19 Prediction based on Blood Test
Authors:	Ziyue Yu, Lihua He, Wuman Luo, Rita Tse and Giovanni Pau
Abstract:	The COVID-19 pandemic is highly infectious and has caused many deaths. The COVID-19 infection diagnosis based on blood test is facing the problems of long waiting time for results and shortage of medical staff. Although several machine learning methods have been proposed to address this issue, the research of COVID-19 prediction based on deep learning is still in its preliminary stage. In this paper, we propose four hybrid deep learning models, namely CNN+GRU, CNN+Bi-RNN, CNN+Bi-LSTM and CNN+Bi-GRU, and apply them to the blood test data from Israelta Albert Einstein Hospital. We implement the four proposed models as well as other existing models CNN, CNN+LSTM, and compare them in terms of accuracy, precision, recall, F1-score and AUC. The experiment results show that CNN+Bi-GRU achieves the best performance in terms of all the five metrics (accuracy of 0.9415, F1-score of 0.9417, precision of 0.9417, recall of 0.9417, and AUC of 0.91).
Download

Short Papers

Paper Nr:	10
Title:	Comparison between Filtered Canny Edge Detector and Convolutional Neural Network for Real Time Lane Detection in a Unity 3D Simulator
Authors:	Jurij Kuzmic and Günter Rudolph
Abstract:	This paper presents two methods for lane detection in a 2D image. Additionally, we implemented filtered Canny edge detection and convolutional neural network (ConvNet) to compare these for lane detection in a Unity 3D simulator. In the beginning, related work of this paper is discussed. Furthermore, we extended the Canny edge detection algorithm with a filter especially designed for lane detection. Additionally, an optimal configuration of the parameters for the convolutional neural network is found. The network structure of the ConvNet is also shown and explained layer by layer. As well known, a lot of annotated training data for supervised learning of ConvNet is necessary. These annotated training data are generated with the Unity 3D environment. The procedure for generation of annotated training data is also presented in this paper. Additionally, these two developed systems are compared to find a better and faster system for lane detection in a simulator. Through the experiments described in this paper the comparison of the run time of the algorithms and the run time depending on the image size is presented. Finally, further research and work in this area are discussed.
Download

Paper Nr:	45
Title:	A Deployable Data as a Service Architecture for Enterprises
Authors:	Adrián Tóth and Mouzhi Ge
Abstract:	Nowadays, data have been considered as one of the valuable assets in enterprises. Although the cloud computing and service-oriented architecture are capable of accommodating the data asset, they are more focused on software or platforms rather than the data per se. Thus, data management in cloud computing is usually not prioritized and not well organized. In recent years, Data as a Service (DaaS) has been emerged as a critical concept for enterprises. It benefits from a variety of aspects such as data agility and data quality management. However, it is still unknown for enterprises why and how to develop and deploy a DaaS architecture. This paper is therefore to design a deployable DaaS architecture that is based on the as-a-Service principles and especially tackles data management as a service. To validate the architecture, we have implemented the proposed DaaS with a real-world deployment.
Download

Area 3 - Internet of Things (IoT) Applications

Short Papers

Paper Nr:	25
Title:	A Low-Complexity Algorithm for NB-IoT Networks
Authors:	Salem Alemaishat
Abstract:	The NB-IoT is a brand-new narrowband IoT technology based on cellular networks. It is an international standard defined by the 3GPP organization. It can be widely deployed worldwide. It focuses on low-power wide area networks and operates based on licensed spectrum. It can be directly deployed in The LTE network has low deployment costs and smooth upgrade capabilities. One of the most influencing factor in NB-IoT networks is time delay, which affects the system performance. Therefore, this paper proposes an efficient algorithm to estimate such factor based on the idea of ICI cancellation method to gradually mitigate the interference between signals in each cell. The proposed algorithm deploys time-frequency cross-correlation overlapping in each iteration based on the conventional correlation method to further enhance the time delay estimation accuracy. Furthermore, based on the noise threshold, a first arrival path algorithm is proposed to eradicate the multipath fading. Simulation results show that the proposed algorithm can effectively improve the time delay as compared with existing algorithms.
Download

Paper Nr:	55
Title:	Development of Low-cost IoT System for Monitoring and Enhancing Renewable Energy Feed-in Tariff at Household in Hong Kong
Authors:	C. C. Lee and George Chan
Abstract:	Feed-in tariff schemes become more popular in many countries nowadays. In October 2018, the Hong Kong Government launched a new policy scheme which was comprised of feed-in tariff scheme and renewable energy certificates. The work mentioned in this paper aims at improving the efficiency of implementing the feed-in tariff scheme at household in Hong Kong by using an automatic intelligent and relatively low-cost IoT system. Besides improving the energy efficiency, the proposed system also includes a website and an App for monitoring the system performance. The experimental results showed that the energy generated by using the proposed system is three times more than the system without smart features. The payback period can be greatly reduced from 11 to 6 years by using the proposed system. It encourages the development of renewable energy sources in Hong Kong or other similar developed cities.
Download

Paper Nr:	32
Title:	IoT based Circadian Rhythm Monitoring using Fuzzy Logic
Authors:	K. Sornalakshmi, Revathi Venkataraman, N. Parthiban and V. Kavitha
Abstract:	A healthy body and happy mind is essential to lead a successful life. For the betterment of health, it is important to develop positive health habits. The human body has a 24-hour internal clock called circadian rhythm that can be affected by our lifestyle. It is a natural intelligence of the human body to perform certain tasks including hormone secretion, memory functions, immune system functions, etc. during certain periods of the day. This rhythm is synchronized with the light and dark cycle of the environment. However, when there are variations in light, sleeping at unusual times, exposure to bright lights at night and traveling across time zones, certain functions of the body may get activated and deactivated at inappropriate times. With smart devices, the Internet of Things (IoT) has a great impact on our everyday lives. Healthcare IoT systems use the data provided by the IoT devices to make automated decisions or to provide recommendations to users. This work is concerned with a health IoT system consisting of different IoT devices used by users who want to know about their circadian rhythms. In addition to this, fuzzy logic has been used to evaluate the circadian rhythm based on the effects of the time at which an individual starts to sleep, wake up time and mobile light exposure time. It classifies the circadian rhythm as aligned, intermediate and disrupted.
Download

Paper Nr:	52
Title:	An Arduino-based Device to Detect Dangerous Audio Noises
Authors:	Lorenzo de Lauretis, Tiziano Lombardi, Stefania Costantini and Ludovica Clementini
Abstract:	The issue of noise pollution is becoming more and more relevant in our today’s way of life. Studies have shown that some noise waves are especially damaging, triggering continuous harm to the nervous scheme with the resulting failure of listening capacity in some instances. Thanks to the latest technological findings, noises can be sampled and analyzed even on very tiny appliances that can possibly be carried anywhere. By testing the noise via a condenser microphone and evaluating the outcome of applying the Fast Fourier Transform to the sampled samples, we can identify the existence of frequencies that are considered detrimental to the auditory system, warning a person in real-time about the prospective risk to which (s)he is facing. Moreover, using a buzzer to alert the user when malicious sounds are detected, our device is also suitable for blind people.
Download

Area 4 - Internet of Things (IoT) Fundamentals

Full Papers

Paper Nr:	3
Title:	A Real-time and Energy-aware Framework for Data Stream Processing in the Internet of Things
Authors:	Egberto A. R. de Oliveira, Flavia C. Delicato, Atslands R. da Rocha and Marta Mattoso
Abstract:	The Internet of things (IoT) has transformed the internet, enabling the communication between every kind of objects (things). The growing number of sensors and smart devices increased the possibilities of data generation and collection. This led to an explosion of data streams being produced which are challenging to be processed in real-time. Regarding the nature of the data, the huge volume, heterogeneity, continuity, disordering, noise and unpredictable rate are some challenging aspects to tackle. Regarding the data processing, the core activities from the data acquisition to the production of high-level knowledge also pose challenges related to limited computational and energy resources and high network latency. In this context, we propose a framework to support activities of a data stream processing workflow for IoT. It aims allowing real-time data processing with low power consumption. Edge computing is used to bring the data processing closer to the data sources and allow actions to be triggered quickly. An adaptive sampling strategy combined with a data prediction model are adopted to reduce the network traffic, thus decreasing the power consumption of the network devices. Experiments show that the proposed framework is able to achieve up to 60.58% average energy consumption savings to sensor nodes and still meet a strict execution time threshold of 1s without compromising the accuracy of the output data on different scales of input streams.
Download

Short Papers

Paper Nr:	36
Title:	Selforganisational High Efficient Stable Chaos Patterns
Authors:	Bernhard Heiden, Volodymyr Alieksieiev and Bianca Tonino-Heiden
Abstract:	The aim of this paper is to provide a new solution for the problem of a simple application of swarm robots, and here the model and its simulation, which shall be later implemented in these Internet of Things (IoT) devices. For this reason this paper describes how, swarm robots, robot-multirobots, a series of entangled robots or robot-os, form predictable selforganisational room-time patterns, as a function of a binary sensor and a binary actor signal interaction, in a triangular cellular automata fashion. The influence of the outer border compared to the inner border of robot-os is investigated, to answer the question, whether and how they can be distinguished. So this process can then be regarded as a different level border-order-entity or as a ’solidification process’ of the robot-o. By means of this, the robot-o is itself ’recognising’, as an extended self, that is identified by the robot-o as the environment. Border as direction change of signal, hence, can be regarded as a basic selforganisational driving force. Above described sensor actor processes can be regarded as bidirectional ordering process, according to orgiton theory, a further development of the theory of selforganisation. Based on the Shannon information entropy, measuring this is methodically demonstrated. Application programs and respective patterns are given in Mathcad and Witness simulations in detail. These prepare for IoT robot-os applications, for future research applications, especially in the open source robot-os of Elmenreich et al., that our work refers to and builds upon.
Download

Area 5 - IoT Technologies

Short Papers

Paper Nr:	21
Title:	Internet of Things based Product-Service System in the Maritime Industrial Sector
Authors:	Islam Abusohyon and Flavio Tonelli
Abstract:	The continuous progress of technology affects all aspects of life and business, dynamically. In order to take the advantaging of technology development, business owners need to adapt themselves with the changes stemming from it. Digitalization of production and service processes is one of the directions that alignment with it will bring many privileges. Internet of Things, cyber-physical system, and artificial intelligence are the popular components of digitalization that constantly undergo evolution. Utilizing these advanced components enables business owners to transform the product-centric processes to smart control digital service-oriented ones. The main motivation of current research work is analysis a theoretical thematic of literature on IoT and CPS servitization topics to shed the light on the main areas that the researchers are focusing on since 2009 and bridge the gap that exists in the literature regarding the implementation of these technologies in the remote monitoring processes in the maritime sector. The result of the literature examination revealed five dominant areas. Through utilizing these disclosed areas, a ten-step approach block diagram for IoT-based ‘smart product servitization’ was designed. The proposed framework supports companies to take the first steps toward remote monitoring servitization through the implementation of IoT and CPS to produce a fully integrated smart monitor system to improve assets’ health and performance and reduce costs and waiting time. Moreover, a case study of a smart injector for marine engine is analysed to propose a working framework supporting the implementation of IoT and CPS to communicate the added-value data within the smart system built on five modules: process control module, process diagnosis module, healing module, storage module, and human interaction module.
Download

Area 6 - Security, Privacy and Trust

Full Papers

Paper Nr:	7
Title:	Detection of Malicious Binaries by Applying Machine Learning Models on Static and Dynamic Artefacts
Authors:	Anantha R. Chukka and V. S. Devi
Abstract:	In recent times malware attacks on government and private organizations are rising. These attacks are carried out to steal confidential information which leads to loss of privacy, intellectual property issues and loss of revenue. These attacks are sophisticated and described as Advanced Persistent Threats(APT). The payloads used in this type of attacks are polymorphic and metamorphic in nature and contains stealth and root-kit components. As a result the conventional defence mechanisms like rule-based and signature-based methods fail to detect these malware. So modern approaches rely on static and dynamic analysis to detect sophisticated malware. However this process generates huge log files. The domain expert needs to review these logs to classify whether the binary is malicious or benign which is tedious, time consuming and expensive. Our work uses machine learning models trained on the datasets, created using the analysis logs, to overcome these problems. In this paper a number of supervised machine learning models are presented to classify the binary as malicious or benign. In this work we have used automated malware analysis framework to collect run time behavioural artefacts. Static analysis mainly focuses on collecting binary meta information, import functions and opcode sequences. The dataset is created by collecting malware from online sources and benign files from windows operating system and third party software.
Download

Paper Nr:	11
Title:	Malware Detection for IoT Devices using Automatically Generated White List and Isolation Forest
Authors:	Masataka Nakahara, Norihiro Okui, Yasuaki Kobayashi and Yutaka Miyake
Abstract:	The number of cyber-attacks using IoT devices is increasing with the growth of IoT devices. Since the number of routes malware infection is increasing, it is necessary not only to prevent infection but also to take measures after infection. Therefore, high-performance detection techniques are required, but many existing technologies require large amounts of data and heavy processing. Then, there is a need for a system that can detect malware infection while reducing the processing load. Therefore, we have proposed an architecture for detecting malware traffic using flow data of packets instead of whole packet information. We performed the malware traffic detection on the proposed architecture by using machine learning algorithms focusing on the behavior of IoT devices, and could detect malware with some degree of accuracy. In this paper, in order to improve the accuracy, we propose a hybrid system using machine learning and the white list automatically generated using the rule of Manufacturer Usage Description (MUD). The white list eliminates benign packets from the target of malware traffic detection, and it can decrease the false positive rate. We evaluate the performance of proposed method and show the effectiveness.
Download

Paper Nr:	26
Title:	SIMBIoTA: Similarity-based Malware Detection on IoT Devices
Authors:	Csongor Tamás, Dorottya Papp and Levente Buttyán
Abstract:	Embedded devices connected to the Internet are threatened by malware, and currently, no antivirus product is available for them. We present SIMBIoTA, a new approach for detecting malware on such IoT devices. SIMBIoTA relies on similarity-based malware detection, and it has a number of notable advantages: moderate storage requirements on resource constrained IoT devices, a fast and lightweight malware detection process, and a surprisingly good detection performance, even for new, never-before-seen malware. These features make SIMBIoTA a viable antivirus solution for IoT devices, with competitive detection performance and limited resource requirements.
Download

Paper Nr:	27
Title:	The VoIP PBX Honeypot Advance Persistent Threat Analysis
Authors:	N. Mcinnes and G. Wills
Abstract:	PBX hacking is a multi-billion dollar per year criminal and terrorism funding source. This paper follows on from a previous 10-day Honeypot experiment, to run a VoIP PBX Honeypot for a longer period of 103-day to not only validate any similarities, but to also analyse non-VoIP methods hackers use in an attempt to gain access to a VoIP System. Over the 103-day data collection period, the Honeypot recorded over 100 million SIP messages. Different techniques were used (including SQL injections in Invites) and hackers of the same IP subnet also attempted using web vulnerabilities in different telephony phone systems to gain access. Of specific interest, over the Christmas period of 2018, attack intensity decreased significantly. To validate these findings, the Honeypot experiment was also conducted for a short period over the Christmas period of 2019 which found that unlike Christmas 2018, attacks increased. The sophistication, scale and complexity of the fraud would suggest an Advance Persistent Threat exists with an aim to infiltrate a VoIP system (including a PBX) to conduct Toll Fraud and where possible to also add that system to a botnet of infected voice systems.
Download

Paper Nr:	50
Title:	Secure Key Management in Embedded Systems: A First Proposal
Authors:	Gorazd Jank, Silvia Schmidt and Manuel Koschuch
Abstract:	The Internet-of-Things (IoT) domain is highly heterogeneous and comprises a multitude of different devices. Because of this variety, many projects require unique compositions of tools, systems, and use cases. In addition, embedded devices are highly optimized and due to that are subject to different constraints. The interconnection of such products for data analysis or cooperation simultaneously increases the attack surface, which leads to requiring efficient cryptographic methods for the protection of data and communication. To enable this, a secure key management approach is needed. In practice however, there are still difficulties regarding the implementation and associated decision making of said management. All the more so since a generic one-size-fits-all approach in such a complex heterogeneous environment as the IoT simply does not exist. This paper aims to provide initial guidelines to argue the choice of a secure key management approach. To do so the state-of-the-art is presented and benefits as well as limits are evaluated. After that a set of factors and a first taxonomy are presented, which influence the final key management solution.
Download

Short Papers

Paper Nr:	6
Title:	Cloud-based Private Querying of Databases by Means of Homomorphic Encryption
Authors:	Yassine Abbar, Pascal Aubry, Thierno Barry, Sergiu Carpov, Sayanta Mallick, Mariem Krichen, Damien Ligier, Sergey Shpak and Renaud Sirdey
Abstract:	This paper deals with several use-cases for privately querying corpora of documents in both settings where the corpus is public or private with respect to an honest-but-curious infrastructure executing the query. We address these scenarios using Fully Homomorphic Encryption (FHE) hybridized with other techniques such as Symmetric Searchable Encryption (SSE) and Private Information Retrieval (PIR) to achieve acceptable system level performances. The paper also presents the prototypes developed to validate the approach and reports on the performances obtained as well as their capacity to scale.
Download

Paper Nr:	8
Title:	Detection of Malicious Binaries by Deep Learning Methods
Authors:	Anantha R. Chukka and V. S. Devi
Abstract:	Modern day cyberattacks are complex in nature. These attacks have adverse effects like loss of privacy, intellectual property and revenue on the victim institutions. These attacks have sophisticated payloads like ransom-ware for money extortion, distributed denial of service(DDOS) malware for service disruptions and advanced persistent threat(APT) malware to posses complete control over the victims computing resources. These malware are metamorphic and polymorphic in nature and contains root-kit components to maintain stealth and hide their malicious activity. So conventional defence mechanisms like rule-based and signature based mechanisms fail to detect these malware. Modern approaches use behavioural analysis(static analysis, dynamic analysis) to identity this kind of malware. However behavioural analysis process is hindered by factors like execution environment detection, code obfuscation, anti virtualization, anti-debugging, analysis environment detection etc. Behavioural analysis also requires domain expert to review the large amount of logs produced by it to decide on the nature of the binary which is complex, time consuming and expensive. To deal with these problems we proposed deep learning methods, where convolutional neural network model is trained on the image representation of the binary to decide the binary nature as malicious or benign. In this work we have encoded the binaries into images in a unique way. Deep convolution neural network is trained on these images to learn the features to identify the binary as malicious or normal. The malware and benign samples for the dataset creation are collected from online sources and windows operating system along with compatible third party application software respectively.
Download

Paper Nr:	12
Title:	Use of Machine Learning for Expanding Realistic and Usable Routes for Data Analysis on Sustainable Mobility
Authors:	Fabian Schirmer, Andreas Freymann, Anamaria Cristescu and Niklas Geisinger
Abstract:	The current mobility or the transition to more sustainable alternatives are constantly changing. For Promoting a sustainable mobility and for investing in a proper infrastructure, we need accurate data regarding the mobility behavior. Gathering location information such as GPS can help to improve the charging infrastructure and the bicycle or pedestrian paths. This motivates the citizens to use sustainable means of transportation such as bicycles or electric cars. However, using personal information via GPS data can cause some challenges: preserving data privacy while keeping data quality to get useful analysis results. This paper presents an advanced approach of processing GPS data based on machine learning and spatial cloaking in contrast to current approaches focusing on common algorithms only. The evaluation has been conducted by generating simulated GPS trips. As a result, the presented approach provides an algorithm that prevents a complete loss of useful data while protecting the privacy of each user in cases where cloaking areas are close together.
Download

Paper Nr:	22
Title:	Identifying Food Fraud using Blockchain
Authors:	Hoi W. Leung, Adriane Chapman and Nawfal F. Fadhel
Abstract:	Cross-contamination, counterfeit ingredients, false packaging, and labelling are all issues that contribute to food fraud which is a major concern undermining the integrity of the food supply chain and consumers health. Therefore, there is a need for an on-demand traceable, transparent food supply chain. This is a universal problem and blockchain presents itself as a means to maintain traceable, transparent food supply. This paper presents an innovative consensus algorithm and simulates the usage of it to identify the precision and recall of fraudulent food detection. This protocol aims to solve the issue of malicious leader node selection in common voting-based consensus protocols while achieving efficiency. Thus, providing a single version of truth for foods in a long food supply chain, preventing information asymmetries.
Download

Paper Nr:	34
Title:	Information Flow Secure CAmkES
Authors:	Amit Goyal, Akshat Garg, Digvijaysingh Gour, R. K. Shyamasundar and G. Sivakumar
Abstract:	Component Architecture for microkernel-based Embedded Systems (CAmkES) is a framework used to build embedded systems software on the top of seL4. seL4, a general purpose microkernel, uses the underlying Discretionary Access Control (DAC) capability model to ensure confidentiality and integrity of the systems built on it. These systems are not information flow secure as DAC model only considers direct read/write accesses and does not consider the indirect accesses. In indirect access, an unauthorized subject can get access to an object through another subject which has the direct access to that object. In this paper, we model and implement information flow secure CAmkES (IFS-CAmkES) which ensures complete mediation by RWFM monitor which is based upon Readers Writers Flow Model (RWFM), a Mandatory Access Control (MAC) model. IFS-CAmkES can be considered as CAmkES enriched with MAC based security. Prototypes of some real life examples have been implemented on IFS-CAmkES. We also compare the performance of CAmkES and IFS-CAmkES based systems.
Download

Paper Nr:	48
Title:	Malicious Activity Detection using Smart Contracts in IoT
Authors:	Mwangi Eric, Hany F. Atlam and Nawfal Fadhel
Abstract:	Internet of Things (IoT) is a unique element in the realm of Cybersecurity. It constitutes countless applications, including defense, health, agriculture, finance, amongst other industries. The majority of existing studies focus on various developments of IoT products and services essential to our day-to-day activities, with little emphasis on the security of developed systems. This has led to the proliferation of IoT solutions acquired through rapid development and overlooking the need for a structured security framework during the systems’ development stages. IoT security capability can be improved by using complementary technologies. This paper explores applying Risk-Based Access Control Model using Blockchain to control access to IoT devices. Although current access control models provide efficient security measures to control who can access the system resources, there is no way to detect and prevent malicious attacks after granting access. The proposed solution utilizes smart contracts under the Hyperledger Fabric (HLF) Blockchain Framework to create access permissions and measure the security risks associated with any event in the IoT system and create access permissions to determine what processes may be performed. This will allow the detection of any malicious activity at the early stages of the attack and grant or deny access based on the risk associated with any activity.
Download

Paper Nr:	53
Title:	Detection of Anomalous User Activity for Home IoT Devices
Authors:	Vishwajeet Bhosale, Lorenzo de Carli and Indrakshi Ray
Abstract:	Home IoT devices suffer from poor security, and are easy to commandeer for unskilled attackers. Since most IoTs cannot run host-based detection, detecting compromise via analysis of network traffic is in many cases the only viable option. Unfortunately, traditional Deep Packet Inspection techniques are not applicable: many IoT devices encrypt their traffic and common attacks (e.g., credential stuffing) cannot be described via signatures. Anomaly detection on traffic features, while effective to identify egregious misbehavior (e.g., a DDoS) cannot identify privacy violations, where an attacker triggers legitimate functions (e.g., streaming video, unlocking a door), but without consent of the user. In this paper, we propose a novel anomaly detection technique based on the analysis of user activities. Our approach builds a model to identify user-performed activities on the device from packet sequences, and uses unsupervised learning to identify deviations from normal user behavior in activity sequences. Thus, it can flag situations where an attacker misuses an IoT device, even when such attacks do not involve protocol-level exploits and do not result in significant anomalies in traffic-level features. Preliminary results show that our approach can effectively map device traffic to activities, and suggest that such activities can be used to distinguish malicious and benign users.
Download

Paper Nr:	9
Title:	A Principled Approach to Enriching Security-related Data for Running Processes through Statistics and Natural Language Processing
Authors:	Tiberiu Boros, Andrei Cotaie, Kumar Vikramjeet, Vivek Malik, Lauren Park and Nick Pachis
Abstract:	We propose a principled method of enriching security related information for running processes. Our methodology applies to large organizational infrastructures, where information is properly collected and stored. The data we use is based on the Hubble Stack (an open-source project), but any alternative solution that provides the same type of information will suffice. Using statistical and natural language processing (NLP) methods we enrich our data with tags and we provide an analysis on how these tags can be used in Machine Learning approaches for anomaly detection.
Download

Paper Nr:	58
Title:	P2S2O: Pseudonymous Polling System for Small Organizations
Authors:	Liuyang Dong, Jiliang Li, Yanxing Li, Jing Ren, Shizhong Xu, Gang Sun and Victor Chang
Abstract:	Voting or polling is often used for decision-making in small organizations, but it is very difficult to achieve a complex balance between transparency and privacy. The requirements of an open-audit voting or polling system for small organizations are discussed. To satisfy the requirements, Pseudonymous Polling System for Small Organizations (P2S2O) is designed. To avoid anyone identifying the voter based on the ballot cast and the MAC address of the corresponding voting device cast the ballot, randomly generated private MAC addresses are used. And to take advantage of the fact that all voters are present at the meeting place, two phases of confirmation are adopted to allow participants to verify their own ballots.