arXiv subjects

Maanak Gupta

Publications and source records attributed to Maanak Gupta.

51 records · Page 3Linked to original sources

arXiv2021

Autoencoder-based Anomaly Detection in Smart Farming Ecosystem

The inclusion of Internet of Things (IoT) devices is growing rapidly in all application domains. Smart Farming supports devices connected, and with the support of Internet, cloud or edge computing infrastructure provide remote control of watering and fertilization, real time monitoring of farm conditions, and provide solutions to more sustainable practices. This could involve using irrigation systems only when the detected soil moisture level is low or stop when the plant reaches a sufficient level of soil moisture content. These improvements to efficiency and ease of use come with added risks to security and privacy. Cyber attacks in large coordinated manner can disrupt economy of agriculture-dependent nations. To the sensors in the system, an attack may appear as anomalous behaviour. In this context, there are possibilities of anomalies generated due to faulty hardware, issues in network connectivity (if present), or simply abrupt changes to the environment due to weather, human accident, or other unforeseen circumstances. To make such systems more secure, it is imperative to detect such data discrepancies, and trigger appropriate mitigation mechanisms. In this paper, we propose an anomaly detection model for Smart Farming using an unsupervised Autoencoder machine learning model. We chose to use an Autoencoder because it encodes and decodes data and attempts to ignore outliers. When it encounters anomalous data the result will be a high reconstruction loss value, signaling that this data was not like the rest. Our model was trained and tested on data collected from our designed greenhouse test-bed. Proposed Autoencoder model based anomaly detection achieved 98.98% and took 262 seconds to train and has a detection time of .0585 seconds.

cs.CR↗

arXiv2021

Edge Centric Secure Data Sharing with Digital Twins in Smart Ecosystems

Internet of Things (IoT) is a rapidly growing industry currently being integrated into both consumer and industrial environments on a wide scale. While the technology is available and deployment has a low barrier of entry in future applications, proper security frameworks are still at infancy stage and are being developed to fit varied implementations and device architectures. Further, the need for edge centric mechanisms are critical to offer security in real time smart connected applications with minimal or negligible overhead. In this paper, we propose a novel approach of data security by using multiple device shadows (aka digital twins) for a single physical object. These twins are paramount to separate data among different virtual objects based on tags assigned on-the-fly, and are used to limit access to different data points by authorized users/applications only. The novelty of the proposed architecture resides in the attachment of dynamic tags to key-value pairs reported by physical devices in the system. We further examine the advantages of tagging data in a digital twin system, and the performance impacts of the proposed data separation scheme. The proposed solution is deployed at the edge, supporting low latency and real time security mechanisms with minimal overhead, and is light-weight as reflected by captured performance metrics.

cs.CR↗

arXiv2021

Detecting Anomalous User Behavior in Remote Patient Monitoring

The growth in Remote Patient Monitoring (RPM) services using wearable and non-wearable Internet of Medical Things (IoMT) promises to improve the quality of diagnosis and facilitate timely treatment for a gamut of medical conditions. At the same time, the proliferation of IoMT devices increases the potential for malicious activities that can lead to catastrophic results including theft of personal information, data breach, and compromised medical devices, putting human lives at risk. IoMT devices generate tremendous amount of data that reflect user behavior patterns including both personal and day-to-day social activities along with daily routine health monitoring. In this context, there are possibilities of anomalies generated due to various reasons including unexpected user behavior, faulty sensor, or abnormal values from malicious/compromised devices. To address this problem, there is an imminent need to develop a framework for securing the smart health care infrastructure to identify and mitigate anomalies. In this paper, we present an anomaly detection model for RPM utilizing IoMT and smart home devices. We propose Hidden Markov Model (HMM) based anomaly detection that analyzes normal user behavior in the context of RPM comprising both smart home and smart health devices, and identifies anomalous user behavior. We design a testbed with multiple IoMT devices and home sensors to collect data and use the HMM model to train using network and user behavioral data. Proposed HMM based anomaly detection model achieved over 98% accuracy in identifying the anomalies in the context of RPM.

cs.LG↗

arXiv2021

Analyzing Machine Learning Approaches for Online Malware Detection in Cloud

The variety of services and functionality offered by various cloud service providers (CSP) have exploded lately. Utilizing such services has created numerous opportunities for enterprises infrastructure to become cloud-based and, in turn, assisted the enterprises to easily and flexibly offer services to their customers. The practice of renting out access to servers to clients for computing and storage purposes is known as Infrastructure as a Service (IaaS). The popularity of IaaS has led to serious and critical concerns with respect to the cyber security and privacy. In particular, malware is often leveraged by malicious entities against cloud services to compromise sensitive data or to obstruct their functionality. In response to this growing menace, malware detection for cloud environments has become a widely researched topic with numerous methods being proposed and deployed. In this paper, we present online malware detection based on process level performance metrics, and analyze the effectiveness of different baseline machine learning models including, Support Vector Classifier (SVC), Random Forest Classifier (RFC), KNearest Neighbor (KNN), Gradient Boosted Classifier (GBC), Gaussian Naive Bayes (GNB) and Convolutional Neural Networks (CNN). Our analysis conclude that neural network models can most accurately detect the impact malware have on the process level features of virtual machines in the cloud, and therefore are best suited to detect them. Our models were trained, validated, and tested by using a dataset of 40,680 malicious and benign samples. The dataset was complied by running different families of malware (collected from VirusTotal) in a live cloud environment and collecting the process level features.

cs.CR↗

arXiv2021

Game Theory Based Privacy Preserving Approach for Collaborative Deep Learning in IoT

The exponential growth of Internet of Things (IoT) has become a transcending force in creating innovative smart devices and connected domains including smart homes, healthcare, transportation and manufacturing. With billions of IoT devices, there is a huge amount of data continuously being generated, transmitted, and stored at various points in the IoT architecture. Deep learning is widely being used in IoT applications to extract useful insights from IoT data. However, IoT users have security and privacy concerns and prefer not to share their personal data with third party applications or stakeholders. In order to address user privacy concerns, Collaborative Deep Learning (CDL) has been largely employed in data-driven applications which enables multiple IoT devices to train their models locally on edge gateways. In this chapter, we first discuss different types of deep learning approaches and how these approaches can be employed in the IoT domain. We present a privacy-preserving collaborative deep learning approach for IoT devices which can achieve benefits from other devices in the system. This learning approach is analyzed from the behavioral perspective of mobile edge devices using a game-theoretic model. We analyze the Nash Equilibrium in N-player static game model. We further present a novel fair collaboration strategy among edge IoT devices using cluster based approach to solve the CDL game, which enforces mobile edge devices for cooperation. We also present implementation details and evaluation analysis in a real-world smart home deployment.

cs.CR↗

arXiv2021

Towards Activity-Centric Access Control for Smart Collaborative Ecosystems

The ubiquitous presence of smart devices along with advancements in connectivity coupled with the elastic capabilities of cloud and edge systems have nurtured and revolutionized smart ecosystems. Intelligent, integrated cyber-physical systems offer increased productivity, safety, efficiency, speed and support for data driven applications beyond imagination just a decade ago. Since several connected devices work together as a coordinated unit to ensure efficiency and automation, the individual operations they perform are often reliant on each other. Therefore, it is important to control what functions or activities different devices can perform at a particular moment of time, and how they are related to each other. It is also important to consider additional factors such as conditions, obligation or mutability of activities, which are critical in deciding whether or not a device can perform a requested activity. In this paper, we take an initial step to propose and discuss the concept of Activity-Centric Access Control (ACAC) for smart and connected ecosystem. We discuss the notion of activity with respect to the collaborative and distributed yet integrated systems and identify the different entities involved along with the important factors to make an activity control decision. We outline a preliminary approach for defining activity control expressions which can be applied to different smart objects in the system. The main goal of this paper is to present the vision and need for the activity-centric approach for access control in connected smart systems, and foster discussion on the identified future research agenda.

cs.CR↗

arXiv2021

Reachability Analysis for Attributes in ABAC with Group Hierarchy

Attribute-based access control (ABAC) models are widely used to provide fine-grained and adaptable authorization based on the attributes of users, resources, and other relevant entities. Hierarchial group and attribute based access control (HGABAC) model was recently proposed which introduces the novel notion of attribute inheritance through group membership. GURAG was subsequently proposed to provide an administrative model for user attributes in HGABAC, building upon the ARBAC97 and GURA administrative models. The GURA model uses administrative roles to manage user attributes. The reachability problem for the GURA model is to determine what attributes a particular user can acquire, given a predefined set of administrative rules. This problem has been previously analyzed in the literature. In this paper, we study the user attribute reachability problem based on directly assigned attributes of the user and attributes inherited via group memberships. We first define a restricted form of GURAG, called rGURAG scheme, as a state transition system with multiple instances having different preconditions and provide reachability analysis for each of these schemes. In general, we show PSPACE-complete complexity for all rGURAG schemes. We further present polynomial time algorithms to solve special instances of rGURAG schemes under restricted conditions.

cs.CR↗

arXiv2020

Future Smart Connected Communities to Fight COVID-19 Outbreak

Internet of Things (IoT) has grown rapidly in the last decade and continue to develop in terms of dimension and complexity offering wide range of devices to support diverse set of applications. With ubiquitous Internet, connected sensors and actuators, networking and communication technology, and artificial intelligence (AI), smart cyber-physical systems (CPS) provide services rendering assistance to humans in their daily lives. However, the recent outbreak of COVID-19 (also known as coronavirus) pandemic has exposed and highlighted the limitations of current technological deployments to curtail this disease. IoT and smart connected technologies together with data-driven applications can play a crucial role not only in prevention, continuous monitoring, and mitigation of the disease, but also enable prompt enforcement of guidelines, rules and government orders to contain such future outbreaks. In this paper, we envision an IoT-enabled ecosystem for intelligent monitoring, pro-active prevention and control, and mitigation of COVID-19. We propose different architectures, applications and technology systems for various smart infrastructures including E-health, smart home, smart supply chain management, smart locality, and smart city, to develop future connected communities to manage and mitigate similar outbreaks. Furthermore, we present research challenges together with future directions to enable and develop these smart communities and infrastructures to fight and prepare against such outbreaks.

cs.CY↗

arXiv2020

AI assisted Malware Analysis: A Course for Next Generation Cybersecurity Workforce

The use of Artificial Intelligence (AI) and Machine Learning (ML) to solve cybersecurity problems has been gaining traction within industry and academia, in part as a response to widespread malware attacks on critical systems, such as cloud infrastructures, government offices or hospitals, and the vast amounts of data they generate. AI- and ML-assisted cybersecurity offers data-driven automation that could enable security systems to identify and respond to cyber threats in real time. However, there is currently a shortfall of professionals trained in AI and ML for cybersecurity. Here we address the shortfall by developing lab-intensive modules that enable undergraduate and graduate students to gain fundamental and advanced knowledge in applying AI and ML techniques to real-world datasets to learn about Cyber Threat Intelligence (CTI), malware analysis, and classification, among other important topics in cybersecurity. Here we describe six self-contained and adaptive modules in "AI-assisted Malware Analysis." Topics include: (1) CTI and malware attack stages, (2) malware knowledge representation and CTI sharing, (3) malware data collection and feature identification, (4) AI-assisted malware detection, (5) malware classification and attribution, and (6) advanced malware research topics and case studies such as adversarial learning and Advanced Persistent Threat (APT) detection.

cs.CR↗

arXiv2020

Learner's Dilemma: IoT Devices Training Strategies in Collaborative Deep Learning

With the growth of Internet of Things (IoT) and mo-bile edge computing, billions of smart devices are interconnected to develop applications used in various domains including smart homes, healthcare and smart manufacturing. Deep learning has been extensively utilized in various IoT applications which require huge amount of data for model training. Due to privacy requirements, smart IoT devices do not release data to a remote third party for their use. To overcome this problem, collaborative approach to deep learning, also known as Collaborative DeepLearning (CDL) has been largely employed in data-driven applications. This approach enables multiple edge IoT devices to train their models locally on mobile edge devices. In this paper,we address IoT device training problem in CDL by analyzing the behavior of mobile edge devices using a game-theoretic model,where each mobile edge device aims at maximizing the accuracy of its local model at the same time limiting the overhead of participating in CDL. We analyze the Nash Equilibrium in anN-player static game model. We further present a novel cluster-based fair strategy to approximately solve the CDL game to enforce mobile edge devices for cooperation. Our experimental results and evaluation analysis in a real-world smart home deployment show that 80% mobile edge devices are ready to cooperate in CDL, while 20% of them do not train their local models collaboratively.

cs.CR↗

arXiv2020

Enabling and Enforcing Social Distancing Measures using Smart City and ITS Infrastructures: A COVID-19 Use Case

Internet of Things is a revolutionary domain that has the caliber to impact our lives and bring significant changes to the world. Several IoT applications have been envisioned to facilitate data driven and smart application for the user. Smart City and Intelligent Transportation System (ITS) offer a futuristic vision of smart, secure and safe experience to the end user, and at the same time efficiently manage the sparse resources and optimize the efficiency of city operations. However, outbreaks and pandemics like COVID-19 have revealed limitations of the existing deployments, therefore, architecture, applications and technology systems need to be developed for swift and timely enforcement of guidelines, rules and government orders to contain such future outbreaks. This work outlines novel architecture, potential use-cases and some future directions in developing such applications using Smart City and ITS.

cs.CY↗

arXiv2020

Analyzing CNN Based Behavioural Malware Detection Techniques on Cloud IaaS

Cloud Infrastructure as a Service (IaaS) is vulnerable to malware due to its exposure to external adversaries, making it a lucrative attack vector for malicious actors. A datacenter infected with malware can cause data loss and/or major disruptions to service for its users. This paper analyzes and compares various Convolutional Neural Networks (CNNs) for online detection of malware in cloud IaaS. The detection is performed based on behavioural data using process level performance metrics including cpu usage, memory usage, disk usage etc. We have used the state of the art DenseNets and ResNets in effectively detecting malware in online cloud system. CNN are designed to extract features from data gathered from a live malware running on a real cloud environment. Experiments are performed on OpenStack (a cloud IaaS software) testbed designed to replicate a typical 3-tier web architecture. Comparative analysis is performed for different metrics for different CNN models used in this research.

cs.CR↗

arXiv2020

Secure V2V and V2I Communication in Intelligent Transportation using Cloudlets

Intelligent Transportation System (ITS) is a vision which offers safe, secure and smart travel experience to drivers. This futuristic plan aims to enable vehicles, roadside transportation infrastructures, pedestrian smart-phones and other devices to communicate with one another to provide safety and convenience services. Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I) communication in ITS offers ability to exchange speed, heading angle, position and other environment related conditions amongst vehicles and with surrounding smart infrastructures. In this intelligent setup, vehicles and users communicate and exchange data with random untrusted entities (like vehicles, smart traffic lights or pedestrians) whom they don't know or have met before. The concerns of location privacy and secure communication further deter the adoption of this smarter and safe transportation. In this paper, we present a secure and trusted V2V and V2I communication approach using edge infrastructures where instead of direct peer to peer communication, we introduce trusted cloudlets to authorize, check and verify the authenticity, integrity and ensure anonymity of messages exchanged in the system. Moving vehicles or road side infrastructure are dynamically connected to nearby cloudlets, where security policies can be implemented to sanitize or stop fake messages and prevent rogue vehicles to exchange messages with other vehicles. We also present a formal attribute-based model for V2V and V2I communication, called AB-ITS, along with proof of concept implementation of the proposed solution in AWS IoT platform. This cloudlet supported architecture complements direct V2V or V2I communication, and serves important use cases such as accident or ice-threat warning and other safety applications. Performance metrics of our proposed architecture are also discussed and compared with existing ITS technologies.

cs.CR↗

arXiv2019

Next-Generation Big Data Federation Access Control: A Reference Model

This paper discusses one of the most significant challenges of next-generation big data (BD) federation platforms, namely, Hadoop access control. Privacy and security on a federation scale remain significant concerns among practitioners. Hadoop's current primitive access control presents security concerns and limitations, such as the complexity of deployment and the consumption of resources. However, this major concern has not been a subject of intensive study in the literature. This paper critically reviews and investigates these security limitations and provides a framework called BD federation access broker to address 8 main security limitations. This paper proposes the federated access control reference model (FACRM) to formalize the design of secure BD solutions within the Apache Hadoop stack. Furthermore, this paper discusses the implementation of the access broker and its usefulness for security breach detection and digital forensics investigations. The efficiency of the proposed access broker has not sustainably affected the performance overhead. The experimental results show only 1\% of each 100 MB read/write operation in a WebHDFS. Overall, the findings of the paper pave the way for a wide range of revolutionary and state-of-the-art enhancements and future trends within Hadoop stack security and privacy.

cs.CR↗

arXiv2019

Secure Cloud Assisted Smart Cars Using Dynamic Groups and Attribute Based Access Control

Future smart cities and intelligent world will have connected vehicles and smart cars as its indispensable and most essential components. The communication and interaction among such connected entities in this vehicular internet of things (IoT) domain, which also involves smart traffic infrastructure, road-side sensors, restaurant with beacons, autonomous emergency vehicles, etc., offer innumerable real-time user applications and provide safer and pleasant driving experience to consumers. Having more than 100 million lines of code and hundreds of sensors, these connected vehicles (CVs) expose a large attack surface, which can be remotely compromised and exploited by malicious attackers. Security and privacy are serious concerns that impede the adoption of smart connected cars, which if not properly addressed will have grave implications with risk to human life and limb. In this research, we present a formalized dynamic groups and attribute-based access control (ABAC) model (referred as \cvac) for smart cars ecosystem, where the proposed model not only considers system wide attributes-based security policies but also takes into account the individual user privacy preferences for allowing or denying service notifications, alerts and operations to on-board resources. Further, we introduce a novel notion of groups in vehicular IoT, which are dynamically assigned to moving entities like connected cars, based on their current GPS coordinates, speed or other attributes, to ensure relevance of location and time sensitive notification services to the consumers, to provide administrative benefits to manage large numbers of smart entities, and to enable attributes and alerts inheritance for fine-grained security authorization policies. We present proof of concept implementation of our model in AWS cloud platform demonstrating real-world uses cases along with performance metrics.

cs.CR↗