arXiv subjects

Raja Jurdak

Publications and source records attributed to Raja Jurdak.

At least 73 records · Page 4Linked to original sources

arXiv2020

Direct Modelling of Speech Emotion from Raw Speech

Speech emotion recognition is a challenging task and heavily depends on hand-engineered acoustic features, which are typically crafted to echo human perception of speech signals. However, a filter bank that is designed from perceptual evidence is not always guaranteed to be the best in a statistical modelling framework where the end goal is for example emotion classification. This has fuelled the emerging trend of learning representations from raw speech especially using deep learning neural networks. In particular, a combination of Convolution Neural Networks (CNNs) and Long Short Term Memory (LSTM) have gained great traction for the intrinsic property of LSTM in learning contextual information crucial for emotion recognition; and CNNs been used for its ability to overcome the scalability problem of regular neural networks. In this paper, we show that there are still opportunities to improve the performance of emotion recognition from the raw speech by exploiting the properties of CNN in modelling contextual information. We propose the use of parallel convolutional layers to harness multiple temporal resolutions in the feature extraction block that is jointly trained with the LSTM based classification network for the emotion recognition task. Our results suggest that the proposed model can reach the performance of CNN trained with hand-engineered features from both IEMOCAP and MSP-IMPROV datasets.

cs.SD↗

arXiv2020

Augmenting Generative Adversarial Networks for Speech Emotion Recognition

Generative adversarial networks (GANs) have shown potential in learning emotional attributes and generating new data samples. However, their performance is usually hindered by the unavailability of larger speech emotion recognition (SER) data. In this work, we propose a framework that utilises the mixup data augmentation scheme to augment the GAN in feature learning and generation. To show the effectiveness of the proposed framework, we present results for SER on (i) synthetic feature vectors, (ii) augmentation of the training data with synthetic features, (iii) encoded features in compressed representation. Our results show that the proposed framework can effectively learn compressed emotional representations as well as it can generate synthetic samples that help improve performance in within-corpus and cross-corpus evaluation.

cs.SD↗

arXiv2020

Deep Architecture Enhancing Robustness to Noise, Adversarial Attacks, and Cross-corpus Setting for Speech Emotion Recognition

Speech emotion recognition systems (SER) can achieve high accuracy when the training and test data are identically distributed, but this assumption is frequently violated in practice and the performance of SER systems plummet against unforeseen data shifts. The design of robust models for accurate SER is challenging, which limits its use in practical applications. In this paper we propose a deeper neural network architecture wherein we fuse DenseNet, LSTM and Highway Network to learn powerful discriminative features which are robust to noise. We also propose data augmentation with our network architecture to further improve the robustness. We comprehensively evaluate the architecture coupled with data augmentation against (1) noise, (2) adversarial attacks and (3) cross-corpus settings. Our evaluations on the widely used IEMOCAP and MSP-IMPROV datasets show promising results when compared with existing studies and state-of-the-art models.

cs.SD↗

arXiv2020

B-FERL: Blockchain based Framework for Securing Smart Vehicles

The ubiquity of connecting technologies in smart vehicles and the incremental automation of its functionalities promise significant benefits, including a significant decline in congestion and road fatalities. However, increasing automation and connectedness broadens the attack surface and heightens the likelihood of a malicious entity successfully executing an attack. In this paper, we propose a Blockchain based Framework for sEcuring smaRt vehicLes (B-FERL). B-FERL uses permissioned blockchain technology to tailor information access to restricted entities in the connected vehicle ecosystem. It also uses a challenge-response data exchange between the vehicles and roadside units to monitor the internal state of the vehicle to identify cases of in-vehicle network compromise. In order to enable authentic and valid communication in the vehicular network, only vehicles with a verifiable record in the blockchain can exchange messages. Through qualitative arguments, we show that B-FERL is resilient to identified attacks. Also, quantitative evaluations in an emulated scenario show that B-FERL ensures a suitable response time and required storage size compatible with realistic scenarios. Finally, we demonstrate how B-FERL achieves various important functions relevant to the automotive ecosystem such as trust management, vehicular forensics and secure vehicular networks.

cs.CR↗

arXiv2020

Energy-aware Demand Selection and Allocation for Real-time IoT Data Trading

Personal IoT data is a new economic asset that individuals can trade to generate revenue on the emerging data marketplaces. Typically, marketplaces are centralized systems that raise concerns of privacy, single point of failure, little transparency and involve trusted intermediaries to be fair. Furthermore, the battery-operated IoT devices limit the amount of IoT data to be traded in real-time that affects buyer/seller satisfaction and hence, impacting the sustainability and usability of such a marketplace. This work proposes to utilize blockchain technology to realize a trusted and transparent decentralized marketplace for contract compliance for trading IoT data streams generated by battery-operated IoT devices in real-time. The contribution of this paper is two-fold: (1) we propose an autonomous blockchain-based marketplace equipped with essential functionalities such as agreement framework, pricing model and rating mechanism to create an effective marketplace framework without involving a mediator, (2) we propose a mechanism for selection and allocation of buyers' demands on seller's devices under quality and battery constraints. We present a proof-of-concept implementation in Ethereum to demonstrate the feasibility of the framework. We investigated the impact of buyer's demand on the battery drainage of the IoT devices under different scenarios through extensive simulations. Our results show that this approach is viable and benefits the seller and buyer for creating a sustainable marketplace model for trading IoT data in real-time from battery-powered IoT devices.

cs.CR↗

arXiv2020

Lightweight Blockchain Framework for Location-aware Peer-to-Peer Energy Trading

Peer-to-Peer (P2P) energy trading can facilitate integration of a large number of small-scale producers and consumers into energy markets. Decentralized management of these new market participants is challenging in terms of market settlement, participant reputation and consideration of grid constraints. This paper proposes a blockchain-enabled framework for P2P energy trading among producer and consumer agents in a smart grid. A fully decentralized market settlement mechanism is designed, which does not rely on a centralized entity to settle the market and encourages producers and consumers to negotiate on energy trading with their nearby agents truthfully. To this end, the electrical distance of agents is considered in the pricing mechanism to encourage agents to trade with their neighboring agents. In addition, a reputation factor is considered for each agent, reflecting its past performance in delivering the committed energy. Before starting the negotiation, agents select their trading partners based on their preferences over the reputation and proximity of the trading partners. An Anonymous Proof of Location (A-PoL) algorithm is proposed that allows agents to prove their location without revealing their real identity. The practicality of the proposed framework is illustrated through several case studies, and its security and privacy are analyzed in detail.

cs.CR↗

arXiv2020

Tree-Chain: A Fast Lightweight Consensus Algorithm for IoT Applications

Blockchain has received tremendous attention in non-monetary applications including the Internet of Things (IoT) due to its salient features including decentralization, security, auditability, and anonymity. Most conventional blockchains rely on computationally expensive consensus algorithms, have limited throughput, and high transaction delays. In this paper, we propose tree-chain a scalable fast blockchain instantiation that introduces two levels of randomization among the validators: i) transaction level where the validator of each transaction is selected randomly based on the most significant characters of the hash function output (known as consensus code), and ii) blockchain level where validator is randomly allocated to a particular consensus code based on the hash of their public key. Tree-chain introduces parallel chain branches where each validator commits the corresponding transactions in a unique ledger. Implementation results show that tree-chain is runnable on low resource devices and incurs low processing overhead, achieving near real-time transaction settlement.

cs.CR↗

arXiv2020

Task Scheduling for Simultaneous IoT Sensing and Energy Harvesting: A Survey and Critical Analysis

The Internet of Things (IoT) has important applications in our daily lives including health and fitness tracking, environmental monitoring and transportation. However, sensor nodes in IoT suffer from the limited lifetime of batteries resulting from their finite energy availability. A promising solution is to harvest energy from environmental sources, such as solar, kinetic, thermal and radio frequency, for perpetual and continuous operation of IoT sensor nodes. In addition to energy generation, recently energy harvesters have been used for context detection, eliminating the need for additional activity sensors (e.g. accelerometers), saving space, cost, and energy consumption. Using energy harvesters for simultaneous sensing and energy harvesting enables energy positive sensing -- an important and emerging class of sensors, which harvest higher energy than required for signal acquisition and the additional energy can be used to power other components of the system. Although simultaneous sensing and energy harvesting is an important step forward towards autonomous self-powered sensor nodes, the energy and information availability can be still intermittent, unpredictable and temporally misaligned with various computational tasks on the sensor node. This paper provides a comprehensive survey on task scheduling algorithms for the emerging class of energy harvesting-based sensors (i.e., energy positive sensors) to achieve the sustainable operation of IoT. We discuss inherent differences between conventional sensing and energy positive sensing and provide an extensive critical analysis for devising new task scheduling algorithms incorporating this new class of sensors. Finally, we outline future research directions towards the implementation of autonomous and self-powered IoT.

eess.SP↗

arXiv2020

Aggregating Buildings as Virtual Power Plant: Architecture, Implementation Technologies, and Case Studies

The prevalence of distributed generation drives the emergence of the Virtual Power Plant (VPP) paradigm. A VPP aggregates distributed energy resources while operating as a conventional power plant. In recent years, several renewable energy sources have been installed in the commercial/industrial/residential building sector and they lend themselves for possible aggregation within the VPP operations. At present, building-side energy resources are usually managed by building energy management systems (BEMSs) with autonomous objectives and policies, and they are usually not setup to be directly controlled for the implementation of a VPP energy management system. This article presents a Building VPP (BVPP) architecture that aggregates building-side energy resources through the communication between the VPP energy management system and autonomous BEMSs. The implementation technologies, key components, and operation modes of BVPPs are discussed. Case studies are reported to demonstrate the effectiveness of BVPP in different operational conditions.

eess.SP↗

arXiv2020

Vaccination strategies on dynamic networks with indirect transmission links and limited contact information

Infectious diseases are still a major global burden for modern society causing 13 million deaths annually. One way to reduce the morbidity and mortality rates from infectious diseases is through preventative or targeted vaccinations. Current vaccination strategies, however, rely on the highly specific individual contact information that is difficult and costly to obtain, in order to identify influential spreading individuals. Current approaches also focus only on direct contacts between individuals for spreading, and disregard indirect transmission where a pathogen can spread between one infected individual and one susceptible individual that visit the same location within a short time-frame without meeting. This paper presents a novel vaccination strategy that relies on coarse-grained contact information, both direct and indirect, that can be easily and efficiently collected. Rather than tracking exact contact degrees of individuals, our strategy uses the types of places people visit to estimate a range of contact degrees for individuals, considering both direct and indirect contacts. We conduct extensive simulations to evaluate the performance of our strategy in comparison to the state of the art's vaccination strategies. Results show that our strategy achieves comparable performance to the oracle approach and outperforms all existing strategies when considering indirect links.

cs.SI↗

arXiv2020

Identifying highly influential travellers for spreading disease on a public transport system

The recent outbreak of a novel coronavirus and its rapid spread underlines the importance of understanding human mobility. Enclosed spaces, such as public transport vehicles (e.g. buses and trains), offer a suitable environment for infections to spread widely and quickly. Investigating the movement patterns and the physical encounters of individuals on public transit systems is thus critical to understand the drivers of infectious disease outbreaks. For instance previous work has explored the impact of recurring patterns inherent in human mobility on disease spread, but has not considered other dimensions such as the distance travelled or the number of encounters. Here, we consider multiple mobility dimensions simultaneously to uncover critical information for the design of effective intervention strategies. We use one month of citywide smart card travel data collected in Sydney, Australia to classify bus passengers along three dimensions, namely the degree of exploration, the distance travelled and the number of encounters. Additionally, we simulate disease spread on the transport network and trace the infection paths. We investigate in detail the transmissions between the classified groups while varying the infection probability and the suspension time of pathogens. Our results show that characterizing individuals along multiple dimensions simultaneously uncovers a complex infection interplay between the different groups of passengers, that would remain hidden when considering only a single dimension. We also identify groups that are more influential than others given specific disease characteristics, which can guide containment and vaccination efforts.

cs.SI↗

arXiv2020

How mobility patterns drive disease spread: A case study using public transit passenger card travel data

Outbreaks of infectious diseases present a global threat to human health and are considered a major health-care challenge. One major driver for the rapid spatial spread of diseases is human mobility. In particular, the travel patterns of individuals determine their spreading potential to a great extent. These travel behaviors can be captured and modelled using novel location-based data sources, e.g., smart travel cards, social media, etc. Previous studies have shown that individuals who cannot be characterized by their most frequently visited locations spread diseases farther and faster; however, these studies are based on GPS data and mobile call records which have position uncertainty and do not capture explicit contacts. It is unclear if the same conclusions hold for large scale real-world transport networks. In this paper, we investigate how mobility patterns impact disease spread in a large-scale public transit network of empirical data traces. In contrast to previous findings, our results reveal that individuals with mobility patterns characterized by their most frequently visited locations and who typically travel large distances pose the highest spreading risk.

cs.SI↗

arXiv2020

Multi-Task Semi-Supervised Adversarial Autoencoding for Speech Emotion Recognition

Inspite the emerging importance of Speech Emotion Recognition (SER), the state-of-the-art accuracy is quite low and needs improvement to make commercial applications of SER viable. A key underlying reason for the low accuracy is the scarcity of emotion datasets, which is a challenge for developing any robust machine learning model in general. In this paper, we propose a solution to this problem: a multi-task learning framework that uses auxiliary tasks for which data is abundantly available. We show that utilisation of this additional data can improve the primary task of SER for which only limited labelled data is available. In particular, we use gender identifications and speaker recognition as auxiliary tasks, which allow the use of very large datasets, e.g., speaker classification datasets. To maximise the benefit of multi-task learning, we further use an adversarial autoencoder (AAE) within our framework, which has a strong capability to learn powerful and discriminative features. Furthermore, the unsupervised AAE in combination with the supervised classification networks enables semi-supervised learning which incorporates a discriminative component in the AAE unsupervised training pipeline. This semi-supervised learning essentially helps to improve generalisation of our framework and thus leads to improvements in SER performance. The proposed model is rigorously evaluated for categorical and dimensional emotion, and cross-corpus scenarios. Experimental results demonstrate that the proposed model achieves state-of-the-art performance on two publicly available datasets.

cs.SD↗

arXiv2020

Trust Management in Decentralized IoT Access Control System

Heterogeneous and dynamic IoT environments require a lightweight, scalable, and trustworthy access control system for protection from unauthorized access and for automated detection of compromised nodes. Recent proposals in IoT access control systems have incorporated blockchain to overcome inherent issues in conventional access control schemes. However, the dynamic interaction of IoT networks remains uncaptured. Here, we develop a blockchain based Trust and Reputation System (TRS) for IoT access control, which progressively evaluates and calculates the trust and reputation score of each participating node to achieve a self-adaptive and trustworthy access control system. Trust and reputation are explicitly incorporated in the attribute-based access control policy, so that different nodes can be assigned to different access right levels, resulting in dynamic access control policies. We implement our proposed architecture in a private Ethereum blockchain comprised of a Docker container network. We benchmark our solution using various performance metrics to highlight its applicability for IoT contexts.

cs.CR↗

arXiv2020

Towards Optimal Kinetic Energy Harvesting for the Batteryless IoT

Traditional Internet of Things (IoT) sensors rely on batteries that need to be replaced or recharged frequently which impedes their pervasive deployment. A promising alternative is to employ energy harvesters that convert the environmental energy into electrical energy. Kinetic Energy Harvesting (KEH) converts the ambient motion/vibration energy into electrical energy to power the IoT sensor nodes. However, most previous works employ KEH without dynamically tracking the optimal operating point of the transducer for maximum power output. In this paper, we systematically analyse the relation between the operating point of the transducer and the corresponding energy yield. To this end, we explore the voltage-current characteristics of the KEH transducer to find its Maximum Power Point (MPP). We show how this operating point can be approximated in a practical energy harvesting circuit. We design two hardware circuit prototypes to evaluate the performance of the proposed mechanism and analyse the harvested energy using a precise load shaker under a wide set of controlled conditions typically found in human-centric applications. We analyse the dynamic current-voltage characteristics and specify the relation between the MPP sampling rate and harvesting efficiency which outlines the need for dynamic MPP tracking. The results show that the proposed energy harvesting mechanism outperforms the conventional method in terms of generated power and offers at least one order of magnitude higher power than the latter.

eess.SP↗

arXiv2020

Poster Abstract: Towards Scalable and Trustworthy Decentralized Collaborative Intrusion Detection System for IoT

An Intrusion Detection System (IDS) aims to alert users of incoming attacks by deploying a detector that monitors network traffic continuously. As an effort to increase detection capabilities, a set of independent IDS detectors typically work collaboratively to build intelligence of holistic network representation, which is referred to as Collaborative Intrusion Detection System (CIDS). However, developing an effective CIDS, particularly for the IoT ecosystem raises several challenges. Recent trends and advances in blockchain technology, which provides assurance in distributed trust and secure immutable storage, may contribute towards the design of effective CIDS. In this poster abstract, we present our ongoing work on a decentralized CIDS for IoT, which is based on blockchain technology. We propose an architecture that provides accountable trust establishment, which promotes incentives and penalties, and scalable intrusion information storage by exchanging bloom filters. We are currently implementing a proof-of-concept of our modular architecture in a local test-bed and evaluate its effectiveness in detecting common attacks in IoT networks and the associated overhead.

cs.CR↗

arXiv2020

Towards Energy Positive Sensing using Kinetic Energy Harvesters

Conventional systems for motion context detection rely on batteries to provide the energy required for sampling a motion sensor. Batteries, however, have limited capacity and, once depleted, have to be replaced or recharged. Kinetic Energy Harvesting (KEH) allows to convert ambient motion and vibration into usable electricity and can enable batteryless, maintenance free operation of motion sensors. The signal from a KEH transducer correlates with the underlying motion and may thus directly be used for context detection, saving space, cost and energy by omitting the accelerometer. Previous work uses the open circuit or the capacitor voltage for sensing without using the harvested energy to power a load. In this paper, we propose to use other sensing points in the KEH circuit that offer information rich sensing signals while the energy from the harvester is used to power a load. We systematically analyse multiple sensing signals available in different KEH architectures and compare their performance in a transport mode detection case study. To this end, we develop four hardware prototypes, conduct an extensive measurement campaign and use the data to train and evaluate different classifiers. We show that sensing the harvesting current signal from a transducer can be energy positive, delivering up to ten times as much power as it consumes for signal acquisition, while offering comparable detection accuracy to the accelerometer signal for most of the considered transport modes.

eess.SP↗

arXiv2019

A journey in applying blockchain for cyberphysical systems

Cyberphysical Systems (CPS) are transforming the way we interact with the physical world around us. However, centralised approaches for CPS systems are not capable of addressing the unique challenges of CPS due to the complexity, constraints, and dynamic nature of the interactions. To realize the true potential of CPS, a decentralized approach that takes into account these unique features is required. Recently, blockchain-based solutions have been proposed to address CPS challenges.Yet, applying blockchain for diverse CPS domains is not straight-forward and has its own challenges. In this paper, we share our experiences in applying blockchain technology for CPS to provide insights and highlight the challenges and future opportunities.

cs.CR↗