WELCOME TO BITS RMIT PHD APPLICATION PORTAL

Project Description

Title: Integration of microfluidic solutions to enhance AIE-carbon dots fluorescent biosensing platforms for the detection of important biomarkers The utilization of fluorescence technology has become increasingly popular due to its exceptional sensitivity and selectivity, ease of handling, and various other unique advantages. Several efficient organic fluorescent probes have been designed for the determination of biomarkers. Compared with traditional organic probes, fluorescent nanoparticles such as carbon dots (CDs) have garnered significant research interest due to their water solubility, biocompatibility, low toxicity, etc. The blue to green CDs are more common and red-emitting carbon dots (RCDs) are scarcely reported. Notably, aggregation-induced emission active CDs have emerged very recently and red-emissive AIE-CDs are rarely reported but presumed to be more efficient fluorescent nanomaterials for biomedical applications. In recent years, cobalt oxyhydroxide (CoOOH) nanoflakes have engrossed significant research interests due to their excellent water dispersion and simple and mild preparation procedure. With its oxidase-like activity, CoOOH nanoflakes in combination with AIE-CDs can be used as nanoenzymes for sensing applications. Microfluidic analytical devices have become a powerful tool for applications in different fields including clinical diagnostics. With the success of solution-based detection of biomarkers, microfluidic device platforms will be developed and tested for the detection and quantitation of biomarkers for early-phase detection of diseases. In this proposed project, we aimed to develop new fluorescence-based microfluidic sensing platforms for the selective detection of important enzyme biomarkers such as nitroreductase (biomarkers for hypoxia), ?-galactosidase (urinary biomarkers for prediction of diabetic kidney disease); esterase (carboxylesterase, biomarker candidate for hepatocellular carcinoma) utilizing a combination of red-emitting carbon dots (CDs) and cobalt oxyhydroxide (CoOOH) as the energy quencher. Objective: • Design and development of red emissive carbon dots with aggregation-induced emission properties and the use of a suitable greener method(s) for their synthesis. • Development of red carbon dots-CoOOH nanoflake based versatile sensory platform for three different enzyme biomarkers nitroreductase, ?-galactosidase, and esterase in aqueous phase. • Design and validation of microfluidic solutions for preci

BITS Supervisor

AMRITA CHATTERJEE

RMIT Supervisor

Francisco Tovar Lopez

Other Supervisor BITS

Other Supervisor RMIT

Cesar Sanchez Huertas

Required discipline background of candidate

Project Description

The research project proposes a secure and privacy-preserving solution for processing and analysing medical datasets. Healthcare information pertaining to patients’ data and transactions needs integrity, auditability, authenticity, scalability, and security features. The existing hospital information systems predominantly rely on cloud infrastructure managed by a single data provider, which exhibits multiple drawbacks, notably insufficient security protocols. Consequently, these weaknesses have resulted in numerous data breaches. The concentration of data in a centralized manner becomes a target for cyberattacks, which, in turn, gives rise to complications in maintaining a consistent patient data perspective across a network. Moreover, certain transactions demand a heightened emphasis on transparent and immutable record-keeping like procurement and shipping transactions within the supply chains of medical equipment and pharmaceuticals. There is a need to develop solutions to ensure data security, privacy and temporal resistance. The medical dataset is stored on a cloud platform, which makes secure data transmission and access control critical. Given the sensitive nature of medical data, it is essential to ensure that it is protected against unauthorized access and breaches during storage and transmission. This is particularly important considering that many machine learning algorithms are being developed to assist healthcare professionals by analyzing this data. The integrity, confidentiality, and availability of the dataset must be preserved to ensure accurate and reliable machine learning outcomes. The project aims to ensure the consistency, integrity, and authenticity of transactions within the context of Electronic Medical Record (EMR) by utilizing secure and privacy-preserving algorithms. Additionally, it extends these principles to the application of machine learning on medical datasets, ensuring that data security and privacy are maintained throughout the process.

BITS Supervisor

Prof. Subhrakanta Panda

RMIT Supervisor

Dr. Hai Dong

Other Supervisor BITS

Other Supervisor RMIT

Tabinda Sarwar

Required discipline background of candidate

Project Description

Traditional digital identity verification systems often face a difficult balancing act. Centralized systems, while efficient, store vast amounts of personal data, making them prime targets for hackers and raising concerns about data misuse. On the other hand, decentralized solutions can struggle to scale effectively for widespread use while still ensuring robust identity verification without compromising user privacy. Current methods often force users to reveal more information than necessary, creating unnecessary vulnerabilities and diminishing trust in digital systems. This project proposes to overcome these challenges by developing a zero-knowledge proof (ZKP) framework specifically for digital identity verification. The goal is to enable users to prove their identity or specific attributes without disclosing any unnecessary personal information. To achieve these ambitious goals, the project will follow a rigorous methodology. This includes a thorough review of existing research on ZKPs and identity verification to identify best practices and gaps. Efficient ZKP protocols specifically tailored for identity verification will be designed and implemented. These protocols will be integrated into a modular and extensible framework that can work seamlessly with current identity management systems and blockchain networks. The framework's performance, scalability, and security will be extensively evaluated.

BITS Supervisor

Amit Dua

RMIT Supervisor

Abebe Diro

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Recommendation systems identify a fraction of items from a very large inventory of items and recommend them to a user. They are trained with machine learning algorithms to maximize a user-centered utility or a business process-oriented one. Automated next-best action recommendation for each user in a sequential, dynamic and interactive context has been widely needed in natural, social and business decision-making. Personalized next-best action recommendation must involve multi-sequence interactions between states, behaviors and actions, and their reactions to their counterpart's actions. No existing modeling theories and tools, including Markovian decision processes, user and behavior modeling, deep sequential modeling, and personalized sequential recommendation, can quantify such complex decision-making on a personal level. In this project, we will take a data-driven approach to learn the next-best actions for personalized decision-making with deep learning. We will develop a recommendation system for marketing strategy predictions. We will focus on fully data-driven learning methods for generating recommendation data such as Collaborative Filtering (CF) and their state-of-the-art variants in deep learning such as Factorization machines (FMs) and Graph Neural Networks (GNNs). The data representation from raw data will be done in the extract-transform-load (ETL) process for data integration, cleaning, transformation, visualization, smoothing, and reduction. It will be extended with steps for data governance on the datasets procured from real-world application. Then the plan is to build more complex models having improvements like link prediction and graph pattern mining to CF algorithms. We will then collect user feedback and incorporate into the data analytics project lifecycles. CF is a popular technique for estimating preference of end users by discovering the implicit correlation between the revealed responses of users. Back-end data architectures will be proposed to create efficient data pipelines. Our research efforts will be directed towards improving the interpretability of the deep learning to give personalized recommendations to individual users. Observations that contradict existing classifications may point to emerging topics. Making the wrong choices can be expensive. So we want to build deep learning models to predict the best offerings to each user, but with the least cost.

BITS Supervisor

Dr. Aneesh Chivukula

RMIT Supervisor

Dr. Ashish Kumar

Other Supervisor BITS

Dr. Manish Kumar

Other Supervisor RMIT

Required discipline background of candidate

Project Description

In response to the growing imperative for sustainability, supply chains are under increasing pressure to shift from linear to circular models. Circular supply chains, rooted in the principles of narrowing, slowing, and closing, prioritise minimising raw material use, extending product durability, and maximising recycling and reuse at the end of a product's life. This approach replaces the conventional ‘end-of-life’ business model with a more sustainable one, emphasising reusing, recycling, reducing, and recovering resources. This research project aims to design a circular supply chain model for manufacturing firms, with a focus on those critical for trade between Australia and India. The study will investigate how these firms, spanning various industries, can effectively implement circularity in their supply chains. The research will delve into the intra and inter-organisational capabilities and institutional factors that interact to achieve a circular supply chain. The study will unfold in three phases. The first phase will encompass a comprehensive literature review to gain insights into the capabilities and factors that facilitate circular supply chains. The second phase will involve gathering qualitative data from manufacturing firms to understand how these identified factors operate in real-world scenarios. Drawing on the insights from Phases 1 and 2, the research will develop a model with hypotheses to illustrate the mechanisms involved in achieving circular supply chains. Finally, a large-scale survey will be conducted to test and validate the model, ensuring its applicability across diverse manufacturing sectors. This multi-phase, multi-methodology study aligns with the United Nations' Sustainable Development Goals, in particular Goal 12 (Responsible Consumption and Production) and Goal 9 (Industry, Innovation and Infrastructure). By incorporating the principles of narrowing, slowing, and closing in sourcing, manufacturing, and recycling, the research aims to guide manufacturing firms in designing circular supply chains to foster a regenerative approach to sustainability.

BITS Supervisor

RAJESH MATAI

RMIT Supervisor

Priyabrata Chowdhury

Other Supervisor BITS

Other Supervisor RMIT

Kwok Hung Lau

Required discipline background of candidate

Project Description

This study delves into the complex interrelationships among Corporate Social Responsibility (CSR) practices, climate risk, green innovation, corporate governance, and sustainable firm performance within the Indian business landscape. By analyzing financial reports spanning from 2010 to 2023, the research aims to uncover the implications, benefits, and challenges associated with integrating CSR considerations into business operations using robust panel data econometric models to control for industry and time effects as well as the potential endogeneity issues. This research will employ quantitative methods, including GLS regression, propensity score matching estimates, instrumental variable analysis, differences-in-difference, and system GMM, to estimate and analyze data extracted and manually collected from secondary sources. Moreover, it emphasizes the necessity of a comprehensive approach to corporate sustainability through climate risk measures. As businesses increasingly recognize the importance of sustainable practices, this study provides valuable insights for policymakers, corporate leaders, and scholars. Furthermore, the research examines the influence of green innovation and corporate governance on the relationship between CSR practices and firm performance. It investigates how the green initiatives of the firm and ownership attributes shape overall firm performance dynamics. Specifically, the study explores the impacts of regulatory measures such as the Companies Act (2013) on board and ownership structure and their subsequent effects on firm performance. Additionally, it assesses whether these measures contribute to sustainable growth and improved social practices, considering the different perspectives on CSR and public relations associated with diversity on boards and green initiatives.

BITS Supervisor

Prof. R. L. MANOGNA

RMIT Supervisor

Prof. Daisy Chou

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Integrated sensing and communication (ISAC) is envisaged as a key technology for 6G communications finding applications in the optimisation of the radio resources. Integration of sensing and communication functionality would lead to resource optimization and also lead to increase in the gain. There are various challenges to the adoption of ISAC in the network which we aim to address through this proposal. This research focuses on ISAC to address spectrum congestion and meet the growing demands of 6G applications, such as smart cities, transportation, homes, and tactical networks. ISAC enhances spectral, energy, hardware, and economic efficiencies through resource and information sharing. Simultaneously, Reconfigurable Intelligent Surface (RIS) technology is considered promising for dynamically manipulating the wireless propagation environment efficiently. The outcome of this research will be a realization of high-accuracy, wide- coverage, and ultra-reliable sensing and communication functionalities. Further, the goal is to determine the superiority of existing solutions, examine performance trade-offs with and without RIS, and highlight the potential of RIS integration in ISAC systems

BITS Supervisor

Sandeep Joshi, Assistant Professor

RMIT Supervisor

Saman Atapattu, ARC Future Fellow (Senior Research Fellow)

Other Supervisor BITS

Other Supervisor RMIT

Kandeepan Sithamparanathan

Required discipline background of candidate

Project Description

The current landscape of cyber incident response is fraught with difficulties. Traditional systems, often reliant on manual intervention, struggle to keep up with the rapid pace of cyberattacks, leading to delays that can be exploited by malicious actors. Furthermore, the sheer volume and complexity of cyber threats pose a formidable challenge for human analysts, who are increasingly overwhelmed. While machine learning-based solutions offer a degree of automation, their lack of transparency and explainability hinders trust and effective human oversight. This disconnect between powerful but opaque AI models and the necessity for accountable decision-making in cybersecurity underscores the urgent need for innovative solutions. This project aims to revolutionize cyber incident response by developing an autonomous system powered by neurosymbolic AI. By moving beyond mere threat detection, the system will make robust, explainable decisions and execute swift, effective actions. This will be achieved through the development of a hybrid AI model that seamlessly integrates neural networks for pattern recognition with symbolic AI for reasoning and decision-making. The model will be trained on diverse cybersecurity datasets and rigorously validated against known and novel threats. Explainability techniques will be incorporated to ensure transparent and accountable decision-making. The system's ability to autonomously execute pre-defined response actions will be developed and tested in simulated and controlled environments. Continuous refinement through human feedback and oversight will further enhance the system's capabilities and ensure ethical and responsible AI behaviour.

BITS Supervisor

Dr. Ashutosh Bhatia

RMIT Supervisor

Abebe Diro

Other Supervisor BITS

Prof. Kamlesh Tiwari

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Sharing of health data for secondary uses such as research and public policy development has many benefits, but also risks if information about an individual's health record can be inferred. Ensuring healthcare data privacy preservation technology is designed to work as per requirements is essential. The requirements for correct functioning of these techniques includes training of both researchers and the individuals. Data is now gathered and possibly used to train AI models from various sources including clinical trials, participatory health-enabling technologies and other health/medical-related records. These are in addition linked, for research, population disease surveillance, risk prediction etc. The increasing use of participatory health enabling technologies and integration of person-generated data with formal healthcare provider data, adds another layer of complexity. Thus, the design of healthcare privacy preserving techniques needs to be considered from all perspectives. The aims of this project are: Design of a process for testing of techniques used for privacy preservation of Healthcare systems. Design process for verification of functionality with ML based systems. Design and implement customized ML scenarios which providing protection with bounds on information leakage. The project will explore various mechanisms to evaluate suitability of the existing techniques and their vulnerabilities given a set of constraints.

BITS Supervisor

Prof. Shubhangi Gawali

RMIT Supervisor

Prof. Asha Rao

Other Supervisor BITS

Neena Goveas

Other Supervisor RMIT

Amy Corman

Required discipline background of candidate

Project Description

Clinical notes obtained from various sources including electronic health records (EHRs), physician documentation, and medical literature contain valuable medical history and evidence associated with individuals. Leveraging this information can lead to more precise diagnoses, prognoses, and treatment plans. Explainable Clinical Natural Language Processing (ClinicalNLP) entails applying NLP techniques in healthcare with an emphasis on transparency and interpretability. NLP is utilized in healthcare settings to analyze and extract information from these clinical documents. However, employing neural network techniques to analyze such data may face challenges in terms of explainability. In ClinicalNLP, the concept of "explainable" emphasizes the significance of comprehending and interpreting the decisions generated by NLP models. This is especially critical in healthcare, where decisions directly influence patient care and outcomes. The goal of Explainable ClinicalNLP is to render the process and outcomes of NLP models transparent and comprehensible to healthcare professionals, patients, and other stakeholders. The objectives and proposed methodology of the project are as follows: (1) To develop a comprehensive quality control and pre-processing pipeline for clinical text: This involves creating a systematic and robust framework to ensure the accuracy, reliability, and usability of clinical text data. This pipeline will serve as a crucial step in the data preprocessing phase, where raw clinical text is transformed into a format suitable for further analysis and interpretation. (2) To Develop Contextualised, Interpretable Models: Using machine learning models that produce interpretable outputs. Further, analyzing the importance of different features or words in the NLP model's decision-making process,will provide insights into which factors influence the model's predictions. We will consider the context of clinical texts and consider domain-specific knowledge to improve the accuracy and relevance of NLP analyses. (3) To Develop Intuitive Visualization:This objective focuses on presenting the results of NLP analyses in a visually intuitive manner, to help understand the patterns and relationships in the data. Setting up interactive visualisations will reduce barriers for clinicians to directly interact with the data to make informed decisions. (4) Validation:We will closely work with healthcare professionals to set up validation pipelines.

BITS Supervisor

Satnik Mitra, Assistant Professor

RMIT Supervisor

Sonika Tyagi, Associate Professor

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

The project delves into developing protocols and algorithms tailored for adopting millimeter-wave (mmWave), free-space optical (FSO), and terahertz (THz) wireless technologies within integrated ground-air-space communication networks. The project is motivated by the complementary characteristics of THz and FSO for space communications while employing mmWave and RF technology in the terrestrial network. Addressing the challenges posed by dynamic scenarios within shared spectrum environments, the project prioritizes spectrum allocation, deep learning methodologies, interference mitigation, and radio channel adaptation. The project aims to autonomously optimize spectrum usage, mitigate interference, and adjust to fluctuating channel conditions in real-time by leveraging state-of-the-art techniques, notably machine learning algorithms. Objectives of the project are as follows: (a)Design and implement advanced protocols and algorithms optimized for the unique characteristics of mmwave, FSO, and THz wireless technologies, ensuring efficient integration and operation within integrated ground-air-space communication networks. (b) Develop solutions to address the dynamic challenges encountered in shared spectrum environments, focusing on spectrum allocation strategies, deep learning methodologies for interference prediction and mitigation, and adaptive radio channel adaptation techniques to maintain robust communication links. (c) Implement intelligent algorithms capable of autonomously optimizing spectrum utilization, dynamically adjusting transmission parameters, and mitigating interference sources in real-time, leveraging machine learning approaches to continuously adapt to changing environmental conditions and network dynamics. (d) Explore innovative methodologies and technologies to catalyze a paradigm shift in wireless networking, introducing novel approaches to spectrum management, interference mitigation, and dynamic resource allocation, with significant implications for critical applications, including disaster response, surveillance, and remote sensing. The project methodology involves rigorous simulations, mathematical modeling and analysis, and verification of developed algorithms through real-time empirical data. Through meticulous research, development, and deployment endeavors, the project endeavors to redefine the capabilities of integrated ground-air-space networks, paving the way for a more robust and efficient communication network.

BITS Supervisor

Syed Mohammad Zafaruddin, Associate Professor

RMIT Supervisor

Akram Hourani, Associate Professor

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Aim: To design a neural network system for an automatic diagnosis based on limited diagnostic data from different modality sensors. Methodology: We will design an algorithm that makes the medical diagnosis based on the combined information from multiple sources, such as different modality sensors or imaging methods. The data can have either single or multiple labels, such as, for example, the type of the disease and the stage of the disease. Current trends are to build very complex single-network multimodal and multi-label structures. The data, training time, and hardware requirements are incredibly high. We propose to replace it with a modular system of interconnected classifiers, each working on a different sub-task. The modules will have simple structures and relatively low data and training requirements. The final decision will be based on the type of connections and information flow between these modules. Each module can be easily re-trained; modules can be switched on or off depending on the application requirements.

BITS Supervisor

Abhijit Das, Dr

RMIT Supervisor

Margaret Lech, Professor

Other Supervisor BITS

Aritra Mukherjee

Other Supervisor RMIT

Richardt Wilkinson, Dr

Required discipline background of candidate

Project Description

The recent technological shifts have led to a tremendous rise in the miniaturized devices, components, features and functional surfaces which find their applications in space, optics, electronics, defense, medical and automotive industries. Various micro-manufacturing techniques generate micron-sized features on substrate materials or produce functional surfaces. Mechanical micromachining offers high precision, flexibility in the choice of work-piece materials, and high efficiency, which are attributable to low cycle time and ease of material removal. Several studies have been carried out in the mechanical micromachining of conventional engineering materials; however, the studies, including numerical and experimental investigations on additively manufactured difficult-to-cut materials, are limited. Additively manufactured difficult-to-cut alloys tend to cause challenges in machining due to continuous tool wear, irregular chip formation and deterioration in surface quality. Owing to continuous involvement between the cutting tool and the workpiece materials during the machining of microfeatures, the cutting edge is subjected to high stresses and consequent tool wear while machining high-strength materials. As the tool wear increases, the material removal mechanisms also tend to change, consequently affecting surface integrity. The surface may get affected in the form of defects such as micron-size burrs, chattering marks and tool marks. The proposed work suggests the modelling-based approach to predict tool wear mechanisms, material removal mechanisms and the effects of cutting forces and vibrations on the quality of the generated surface. Numerical modelling based on additively manufactured material parameters is proposed to simulate the micromachining process for cutting forces. The preliminary experiments will be carried out to validate the simulated results. The functional surface desired to perform a specific function (wettability or tribology-related) will be fabricated using mechanical micromachining. Post fabrication, surface quality will be thoroughly investigated regarding the machined features' size, surface finish, and subsurface damage.

BITS Supervisor

ANUJ SHARMA

RMIT Supervisor

Songlin Ding, Professor

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Natural weathering and anthropogenic activities resulted in percolation of various contaminants and pathogenic bacteria in the groundwater, thereby declining its quality. Arsenic and microbes are two such lethal contaminants which causes varieties of diseases in human body, ranging from neural apathy to cancer. Amongst different removal strategies used, (i.e., adsorption, precipitation and coagulation) membrane filtration is attractive due to its high retention and flux. But membrane fibers prepared by conventional wet spinning method has higher pore sizes and faces selectivity issues, after few cycles of operation. This method also requires expensive spinneret assembly that require space and maintenance. Rather, electrospinning produces nanofibers with high porosity, specific surface area and controllable membrane thickness to directly promote infiltration rate and contaminant rejection ratio. Additionally, a specific ion can be removed by impregnating functional materials, like, nanoparticles in the polymer matrix. These fibers are known as composite nanofibers and they have both filtration and adsorptive properties. Current proposal aims to develop these composite electrospun nanofibers, by impregnating iron nanoparticles (IP) in different polymer melts, such as, dimethyl formamide (DMF) dissolved polyacrylonitrile (PAN), cellulose acetate (CA) and polyethersulfone (PES) on a polywoven ester fabric substrate. Electrospinning conditions, such as, applied electric field, distance between the needle and collector, melt flow rate, and needle diameter will be optimized based on the performance of the fibers, mainly in terms of their permeability, porosity, cut off, hydrophilicity, arsenic and microbial uptake capacity. Apart from that, these fibers will also be characterized based on their morphology and structure and mineralogical parameters. Best fiber with a specific preparation condition and composition will be chosen following this assessment. This fiber will be used for continuous filtration against arsenic and water borne pathogens as a function of different operating conditions, i.e., pH, pressure, flow rate and coexisting ions in crossflow and batch mode. Equilibrium uptake capacity of the best nanofibers will also be performed to know about its maximum uptake capacity for arsenic and microbes. As membrane fouling will be an inherent limitation in such operations, regeneration and disposal strategy of finally exhausted membrane will be designed.

BITS Supervisor

Dr. Somak Chatterjee

RMIT Supervisor

Prof. Namita Roy Chowdhury

Other Supervisor BITS

Prof. Banasri Roy

Other Supervisor RMIT

Seungju Kim, Dr

Required discipline background of candidate

Project Description

The electrocatalytic reduction of gaseous CO2 and NOx, powered by renewable electricity, is a promising strategy to not only mitigate the pollutants from the atmosphere, but also to valorize the pollutants to high energy-density fuels, like alcohol and ammonia. However, the state of the art is far from being optimal and the level of understanding of the mechanistic pathways is very poor at present. Also, there are still considerable breakthroughs to be made before it can be considered as a viable economical process. Moreover, the reactions suffer from low activity and poor product selectivity primarily due to the competitive hydrogen evolution reaction. Our proposal focuses on designing and development of high surface area, porous and highly conducting metal organic framework derived single atom catalysts (SACs) for the catalytic processes. The experienced collaboration will facilitate fine tuning of the appropriate SACs for efficient and selective electroreduction with high Faradaic Efficiency. In-situ spectroscopic studies along with DFT calculations will be made towards understanding the molecular mechanism with respect to the structural, morphological, and electronic properties of the synthesized SACs. The final aim will be to develop high-fidelity techno-economic-analysis and life cycle-analysis models to evaluate the economic and environmental benefits along with feasibility and scalability of the process.

BITS Supervisor

Prof. Sounak Roy

RMIT Supervisor

Prof. James Tardio

Other Supervisor BITS

Prof. B M Reddy

Other Supervisor RMIT

Prof. Suresh Bhargava

Required discipline background of candidate

Project Description

This project aims to develop novel materials with distinct electrochemical properties and Electrochemical energy storage (EES) systems are pivotal in transforming our lifestyle, as they are integrated into electronic components and electric vehicles (EVs). They also enhance the reliability of renewable energy production systems, such as fuel cells, solar cells, wind and tidal power, by providing a platform for large-scale energy storage. Among various EES systems, batteries and supercapacitors (SCs) are the primary systems capable of large-scale energy storage. However, they face challenges related to poor power and energy densities, respectively, which are primarily due to the limitations of the electrodes. Furthermore, issues with the long-term stability of electrode materials can lead to rapid degradation of storage cells, necessitating replacement after a limited number of cycles. The limited lattice space in bulk electrode materials restricts ion insertion, resulting in slow charge-discharge rates, poor power density, and electrode failure. While energy density can be increased by maximising ionic storage, bulk materials only offer a finite number of intercalation sites, and their surface is not fully available for charge storage. Additionally, the reversible intercalation of ions leads to the expansion and contraction of electrode materials, causing mechanical stresses that can result in electrode cracking or delamination from the current collectors. Some materials also undergo phase transformations that produce redox-inactive phases, reducing capacity. These mechanical stresses and phase changes significantly impact the efficiency and lifecycle of EES systems. Therefore, to enhance the stability and cycle life of electrode materials, their phase transformation reactions should be perfectly reversible, and there should be sufficient space to accommodate the resulting stress, which is only possible with atomic-level reactions on planar surfaces. Two-dimensional (2D) materials provide a promising platform for designing new electrode materials to overcome the limitations of various energy storage devices, particularly SCs and batteries. This project aims to develop heterostructures of these 2D materials with perfect face-to-face heterointerfaces at individual flakes. Both wet-chemical and physical methods will be employed to develop materials and explore their performance for different battery chemistries, such as sodium, potassium, zinc, and others.

BITS Supervisor

Sandip S. Deshmukh

RMIT Supervisor

Prof. Nasir Mahmood

Other Supervisor BITS

R. Parameshwaran

Other Supervisor RMIT

Muhammad Waqas Khan

Required discipline background of candidate

Project Description

The proposed project focusses on the unsaturated behaviour and reliability analysis of carbon sequestrated sustainable binders in soil stabilization. Addressing unsaturated soil behaviour is crucial as it significantly influences the performance of carbon-sequestrated sustainable binders in soft clay stabilization. The existing literature lacks sufficient insights into unsaturated conditions, highlighting a critical knowledge gap. Incorporating reliability analysis is essential to quantify uncertainties associated with binder performance, ensuring the project's findings are robust and applicable in real-world scenarios. By addressing these aspects, the research aims to provide comprehensive guidelines for sustainable soil stabilization practices. Specifically, the study seeks to 1) estimate the carbon sequestration capacities of the soft clay stabilized with sustainable binders (such as fly ash and steel slag), 2) investigate unsaturated soil properties, involving the development of the soil-water characteristic curve (SWCC) for carbonated sustainable materials, and 3) establish a reliability-based framework for the strength characteristics of soft clays stabilized with carbonated binders. This study is anticipated to address challenges related to soft clays, waste management, and carbon emissions. The outlined methodology includes a) collecting and conducting geotechnical characterization of soft clay and binders (fly ash and steel slag), b) evaluating the carbon sequestration potential of soft clay stabilized with binders, c) estimating the SWCCs of treated and untreated soft clay using the filter paper method, d) conducting unconfined compression strength tests on treated and untreated soft clays, and e) developing a reliability-based design approach by considering uncertainties associated with SWCC fitting parameters and the percentage of binder. Through this comprehensive approach, the project aims to provide a deeper understanding of the unsaturated behaviour of carbon sequestrated sustainable binders and establish a reliable framework for the geotechnical engineering community, thereby advancing sustainable practices in soft clay stabilization.

BITS Supervisor

Dr. Raghuram Ammavajjala

RMIT Supervisor

Prof. Annan Zhou

Other Supervisor BITS

Prof. Anasua GuhaRay

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Wear is an undesirable material deterioration phenomenon that affects awide range of technological systems, often leading to premature failures and causing health issues. In the context of modern automotive technology, the sintered pad and disc system are the main sources of non-exhaust pollution, releasing large amounts of wear debris or particulate matter during braking operations. Studies on these nano-sized airborne particles and their effects on human health have revealed that they are more hazardous to humans and the environment due to their increased surface area and higher reactivity. Therefore, the project aims to develop an advanced tribologically optimized novel friction material for the automotive brake pad-disc system. This material should support the wear reduction mechanism without compromising mechanical frictional performance, reducing brake dust load and extending service life. Typically, conventional brake pads consist of sintered friction material deposited on steel backing plates. In contrast, this project aims to develop a brake pad made of metal matrix composite (MMC) powders mixed with friction modifiers-transition metal dichalcogenides (TMDs) on a steel backing plate prepared using a novel 3-D fabrication process that eliminates the use of binders. The investigation will focus on how the prepared friction material with different alloy compounds can replace the thick and bulky conventional sintered brake pad-disc system. Along with the MMC fabricated brake pads, the tribological pair will include a high wear resistance and lightweight Al-SiC matrix rotor or disc that can withstand heavy axial and tangential load with excellent frictional properties. To achieve this, the principal investigator will concentrate on the tribological pair/system of 1. Fe-based MMC powder comprising Iron alloy, hard reinforcement phases, and TMDs as a friction modifier for the brake friction material fabricated using a novel 3-D fabrication process. 2. Al-SiC MMC-based counterpart disc with high wear resistance, strength, and lightweight properties. Developing and investigating practical and cost-effective advanced tribological systems for brake pad-disc systems to reduce brake dust load and extend service life is highly desirable. The tribologically optimized friction material is suitable not only for combustion engines but also for electric vehicles to meet low particulate matter emission requirements.

BITS Supervisor

Piyush Chandra Verma

RMIT Supervisor

Prof Raj Das

Other Supervisor BITS

Dr. Parikshit Sahatiya

Other Supervisor RMIT

Dr. Maciej Mazur Senior Lecturer

Required discipline background of candidate

Project Description

National and organisational cultural characteristics might influence requirements engineering (RE) and project management (PM) activities. This influence might have an impact on the choice of project management methodology, artefacts, and metrics. Where some artefacts and metrics are strongly associated with particular methodologies, the actual choice from the available range may vary. There have been several studies to identify the impact of cultural values on the choice of choice of project management methodology, e.g., [1,2]. However, there is a limited body of knowledge regarding potential correlations among cultural values and the choice of artefacts and metrics. The aim of this work is to analyse these correlations and their impact, with the focus on Indian culture. Methodology of the project will include the following core phases: 1. A systematic literature review will be conducted on the related works, 2. A survey and a set of interviews will be conducted with practitioners to identify correlations and their impact. On this basis, extension to the Alsanoosy’s framework [2] will be elaborated. 3. The extended framework will be implemented and evaluated both from correctness and usability perspective. [1] M. Spichkova et al., 2021. Impact of Organisational Culture on the Requirement Engineering Activities. RE, IEEE. [2] T. Alsanoosy et al., 2020. Does our culture influence requirements engineering activities? RE, IEEE.

BITS Supervisor

Dr Prajna Upadhyay, A/Prof

RMIT Supervisor

Dr Maria Spichkova, Senior Lecturer

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

The conventional solar energy storage systems are capable of storing only the thermal energy obtained from the solar radiation. However, they do suffer from huge thermal losses of the order of 15 °C to 20 °C with low energy density, self-discharge, requirement of huge volumes with associated high cost of construction and so on especially when the sensible heat storage materials are utilized. In this context, the well-organized group of investigators from the RMIT and BITS will thoroughly address the above-mentioned fundamental challenges with their complementary expertise gained over the years in this field of research. For the very first time, this research team aims at the development of a low cost and highly efficient next generation smart phase transition film (SPTFi) enriched with bio-polymers and nanomagnetic particles. This collaborative research is immensely significant, that adds value to the field of research in terms of achieving the following salient aspects (but not limited to): • Enhanced solar photo-to-thermal energy conversion efficiency (>85 %) using biopolymers. • Enhanced heat conduction and swift transfer of heat through the incorporation of nanomagnetic particles. • High thermal energy storage capability (>95 %) • Reliability up to 20000 thermal cycles. • Reduction in its overall cost by 30 % (when compared to the existing sensible heat storage materials). The research team will construct a small-scale workable model of the proposed smart phase transition film (SPTFi)-based energy storage system to demonstrate its real-time application for efficient conversion of solar photo-to-thermal energy without sacrificing the above-mentioned salient points.

BITS Supervisor

R. Parameshwaran

RMIT Supervisor

Dr Hamid Arandiyan

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Many energy sectors depend on fossil fuels, which release three-quarters of greenhouse gas emissions and lead to severe climate changes, bringing about catastrophic environmental changes and rising average atmospheric temperature. Hence, the research community is searching for energy transformation using renewable energy sources. Governments worldwide have set ambitious targets to achieve net-zero emissions by 2050, necessitating groundbreaking innovations and widespread deployment across various sectors. Key areas of focus include advancements in battery technology, the utilization of renewable hydrogen, and the development of CO2 capture and utilization technologies aimed at significantly curtailing CO2 emissions. Among these innovations, catalytic conversion processes for CO2 utilization have emerged as a promising avenue for reducing CO2 accumulation while yielding valuable chemicals. Notably, methanol synthesis through CO2 hydrogenation represents one such process garnering considerable attention for its potential to mitigate CO2 emissions and produce high-value products. The conversion of CO2 to methanol is still a great challenge because of CO2 low reactivity and thermodynamic stability. The methanol from CO2 is exothermic and thermodynamically favorable at low temperatures and high pressures. However, considering the reaction kinetics and the stable nature of CO2, a high reaction temperature is required. In addition, the reverse water gas shift (RWGS) reaction is endothermically favorable at high temperatures and low pressures that result in low methanol selectivity and low CO2 conversions. Moreover, the water from CO2 hydrogenation leads to catalyst deactivation and sintering. These challenges could be addressed by developing new catalysts for CO2 hydrogenation, leading to high methanol selectivity and high CO2 conversions. The proposed research aims to develop stable catalyst such as Cu with Zn, Ni, is supported on perovskite oxides and will be tested for CO2 hydrogenation. Bimetallic catalysts have shown a more synergetic effect than monometallic catalysts and can enhance physical and chemical properties like dispersion, active surface area, surface basicity, and CO2 adsorption results in high methanol selectivity with enhanced CO2 conversion.

BITS Supervisor

Srinivas Appari, Associate Professor

RMIT Supervisor

Selvakannan Periasamy

Other Supervisor BITS

Srikanta Dinda, Professor

Other Supervisor RMIT

Jampaiah Deshetti

Required discipline background of candidate

Project Description

Automated processing of natural language requirements has been explored for over a decade [1]. These techniques cover different aspects of requirements engineering such as requirements analysis, elicitation, or quality assessment, including compliance checks. When the requirements come from different contexts (countries, organizations, or situations), it becomes challenging to perform a compliance check, and it is extremely crucial that we also refer to the regulations [2]. Regulatory text is hard to read and understand for laymen, and companies may have to pay fines if they cannot prove to ensure compliance with the existing regulations during auditing procedures. To address this critical need, we propose the development of a system that leverages the advanced text understanding capabilities of Large Language Models (LLMs), such as GPT, LLaMA, or Mistral, to automatically interpret and ensure compliance with regulatory requirements throughout the requirements engineering process. Specifically, our approach involves converting regulatory texts into Attempto Controlled English (ACE), a semi-formal representation that is both human-readable and suitable for automated reasoning. Our system will (i) translate diverse regulations into ACE, (ii) analyze the diversity of requirements, identifying ambiguities and contradictions, and (iii) enhance requirements analysis by enabling precise information retrieval and scenario-based querying. This will not only streamline compliance checks but also significantly reduce the risk of regulatory non-compliance, ultimately safeguarding organizations against costly penalties. 1. R. Sonbol et al. The Use of NLP-Based Text Representation Techniques to Support Requirement Engineering Tasks: A Systematic Mapping Review. Tech. Rep. 2021. 2. M. Spichkova et al. Structuring Diverse Regulatory Requirements for Global Product Development. ReLaw 2015. 3. S. Klaus et al. Summarizing Legal Regulatory Documents using Transformers. SIGIR 2022. 4. http://attempto.ifi.uzh.ch/site/ 5. O. Amaral et al. AI-Enabled Automation for Completeness Checking of Privacy Policies. IEEE Trans. on SE. Nov. 2022, pp. 4647-4674, vol. 48

BITS Supervisor

Dr Prajna Upadhyay, A/Prof

RMIT Supervisor

Dr Maria Spichkova, Senior Lecturer

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

The proposed PhD project focuses on sustainable water desalination by integrating a Membrane Distillation (MD) system for mineral recovery and achieving a Zero Liquid Discharge (ZLD) process. The main objectives of the study are to develop an innovative and environmentally friendly approach to desalination, one that not only produces fresh water but also recovers valuable minerals from the brine and eliminates liquid discharge, thus minimizing environmental impact. Aims: The primary aim of the research is to design and optimize a Membrane Distillation system for desalination, mineral recovery and zero liquid discharge. In conventional desalination processes, the concentrated brine, enriched with valuable minerals, is often discarded as waste, causing environmental concerns. The proposed project seeks to develop methods for the recovery of these minerals from the brine stream, turning a waste product into a valuable resource. Methodology: Literature Review: The initial phase of the research will involve an in-depth literature review to gain a comprehensive understanding of existing MD systems, mineral recovery techniques, and ZLD processes in desalination. MD System Design and Optimization: The researcher will work on designing and optimizing the MD system. This will include selecting appropriate membranes, conducting experiments to determine the best operating conditions, and evaluating the systems energy efficiency and performance. Mineral Recovery Techniques: Various mineral recovery techniques will be explored to extract valuable minerals from the brine. These techniques may include crystallization, precipitation, or other innovative approaches suitable for the specific minerals present. Zero Liquid Discharge Implementation: The researcher will focus on developing and implementing a comprehensive ZLD strategy, ensuring that no liquid discharge occurs throughout the entire desalination process.

BITS Supervisor

Manoj Kumar Soni, Professor

RMIT Supervisor

Abhijit Date, Associate Professor

Other Supervisor BITS

Other Supervisor RMIT

Kiao Inthavong

Required discipline background of candidate

Project Description

Stable surfactant foams are very effective in remediation of petroleum contaminated soils around the world but producing a highly stable surfactant foam remains the main barrier for remediation. In this project, stable surfactant foams will be generated with the aid of nanoparticles and will apply such foams in the remediation testing of petroleum contaminated desert soils in India and different soils in Australia. The soils involved in the work will be collected from field including actual petroleum industry contaminated sites and characterized in detail. Effect of shape, size and surface chemistry of the nanoparticles on the foam properties and the contaminated soil remediation will be studied. The work would include the use of nano-biochar, a promising material which, to date, has not been used in conjunction with surfactant foams. The stable foams produced will be also characterized with the help of Dynamic Foam Analyzer equipment available here in India. Also it is proposed to perform the foam simulations to develop better understanding on the contaminated soil remediation mechanism.

BITS Supervisor

Pradipta Chattopadhyay, Associate Professor

RMIT Supervisor

Dr. Jorge Paz-Ferreiro

Other Supervisor BITS

Prof. Banasri Roy

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Electric vehicles (EVs) have been considered an efficient solution to alleviate the intensifying energy and environmental crisis. Due to the high energy density, long cycling life, and high efficiency, lithium-ion batteries (LIBs) are selected as the primary power source for EVs. However, the performance of the LIBs is greatly affected by their operating temperature. During the charge and discharge processes, LIB's reasonable working temperature range is 25–40 ?C, and the temperature difference of the battery module should be within 5?C. The power capacity of the LIB decreases at low temperatures, while the risk of thermal runaway increases at high temperatures. Therefore, a battery thermal management system (BTMS) is significant in ensuring the thermal safety of the battery and enhancing the battery's electrical performance in the operating process. According to the different cooling mediums, BTMS can be classified into three methods: air cooling, liquid cooling, phase change material (PCM), and a combination of them. Although the PCM increases the installation space, weight, and cost of the BTMS, the higher thermal conductivity of PCM-based hybrid cooling excels in its efficacy over the air or liquid cooling approaches. As the PCM experiences continuous phase changes, an innovative cold plate design is essential to encapsulate the PCM and maximize the heat transfer. Therefore, it is decided to indigenously design a triply periodic minimal surface (TPMS) based cold plate to encapsulate the PCM. The following objectives are proposed to investigate the thermal management system of lithium-ion batteries integrated with TPMS structures. • Numerically explore the ideal cold plate design for battery thermal management using various sheet-based TPMS structures. • Development of TPMS structures-based cold plates for thermal management of Li-ion battery pack. • Explore the durability and thermal stress of the cold plate for volatile operating conditions. • Explore the performance enhancement during hybridization of the cold plate with phase change materials for thermal management of the battery pack.

BITS Supervisor

Santanu Prasad Datta

RMIT Supervisor

Dr. Maciej Mazur Senior Lecturer

Other Supervisor BITS

NITIN RAMESHRAO KOTKUNDE

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Due to efficiency and latency issues, Machine Learning (ML) services using resources at the Internet-of-Things network edge near the data sources has emerged as edge intelligence (EI). EI is expected to push Deep Learning (DL) computations from the cloud to the edge thus enabling various distributed, low-latency, high-throughput and reliable ML services. The research challenges facing Edge DL for ubiquitous intelligence are training, inference, optimizing the ML on Edge. Suitable network architecture, hardware and software support are required for the DL computations in ML services. In this project, we will apply Adversarial Machine Learning (AML) to data streams collected for EI. Deep transfer learning strategies will be researched to facilitate AML on such resource-constrained edge devices. Storing, reusing and transferring information and knowledge from previous datasets and tasks has the potential to improve sample efficiency in AML. Domain adaptation is a research area in transfer learning that can model distributional shifts in data available for validating the robustness of AML algorithms. It will result in AML paradigms such as incremental learning, utility learning, reinforcement learning and online learning with class and cost distribution information for the transferable feature representations in data streams. Tensor decomposition is useful to analyze the transferable feature representations in such AML formulated as a non-linear optimization problem. The resultant optimization algorithms can be sped up with randomized algorithms that select columns according to a biased probability distribution for tensor decompositions. They can be interpreted as generalizations of low-rank tensor approximation methods representing statistically significant portions of the training data obtained from real world processes. We shall work on transfer metric learning to analyze the low-rank approximations of tensors in EI data streams with optimization methods such as canonical correlation maximization. They will be integrated with existing efforts for running DL on the edge such as Model Input narrowing down the ML searching space, Model Structure of specialized models that waive the generality of standard ML models in exchange for faster inference, Model Selection in the edge by combining different compression techniques in DL models balancing between performance and resource constraints.

BITS Supervisor

Dr. Aneesh Chivukula

RMIT Supervisor

Professor Jenny Zhang

Other Supervisor BITS

Dr Nikumani Choudhury

Other Supervisor RMIT

Required discipline background of candidate

Project Description

This project seeks to develop a framework for enhancing the security of Smart City Internet of Things (IoT) applications by developing a real-time anomaly detection system utilizing federated learning and device-based multimodal anomaly scoring. With the rapid proliferation of IoT devices in urban environments, there is a critical need to ensure the security and integrity of these interconnected systems. The project aims to leverage existing smart city sensor networks (traffic cameras, environmental sensors, etc.) for data collection. The project aims to train a robust anomaly detection model collaboratively across edge devices without transferring raw sensor data. This protects privacy and reduces bandwidth consumption. The scope also includes proposing novel techniques for securing the model updatation and processing between IoT devices. Furthermore, the project will explore the development of predictive models that can anticipate security vulnerabilities and proactively recommend mitigation measures. By harnessing the predictive capabilities of machine learning techniques, Smart City administrators can preemptively address potential security risks, safeguarding critical infrastructure and ensuring the privacy and safety of citizens. Ultimately, this project seeks to establish a robust framework for securing Smart City IoT applications using machine learning techniques, such as Secure Aggregation for Federated Training (SAFT), paving the way for resilient and secure urban environments in the digital age. Through the integration of advanced technologies, Smart Cities can thrive as safe, efficient, and interconnected hubs of innovation and progress.

BITS Supervisor

Vinay Chamola

RMIT Supervisor

Kandeepan Sithamparanathan

Other Supervisor BITS

Other Supervisor RMIT

Dr Jing Fu

Required discipline background of candidate

Project Description

Over the years, lattice structures have been studied extensively due to their exceptional properties, lightweight design, and energy absorption capacity. Recently, a novel class of lattice structures, the Triply Periodic Minimum Surface (TPMS) lattices, have evoked great interest due to their incredibly high strength and energy absorption capability. The main characteristic of TPMS structures is zero curvature, which allows for more uniform stress distribution that can carry more loads and produce smoother crushing. However, it is challenging to produce TPMS lattice structures using traditional manufacturing techniques due to their complex geometries. Additive manufacturing (AM) makes it possible to produce these complex TPMS structures. Specifically, Selective Laser Melting (SLM) has successfully processed TPMS structures using various materials. ASS 316L is a promising structural material used in various sectors such as automotive, aerospace and nuclear applications due to excellent mechanical and corrosion properties at high-temperature conditions. The implementation of these TPMS structures using ASS 316L provides a lot of opportunities in various structural applications with lightweight and better performance outcomes. Thus, it is necessary to explore the mechanical and corrosion performance of different ASS 316L TPMS structures. The knowledge built during this study will be helpful in exploring many possibilities to replace the conventional components used in the automotive, naval, and aerospace sectors. The aim of the study is to thoroughly understand the mechanical and corrosion behaviour of ASS 316L at a wide range of strain rates and temperature conditions. The following objectives are planned to achieve the proposed aim. • Understanding the deformation behaviour of various TPMS structures at quasi-statics strain rates (0.001 s-1) and different temperature conditions (Room temperature to 4000C.) • Analysis of mechanical response of TPMS structures under high strain rate loading conditions (100 s-1 to 3000 s-1). • Analysis of the corrosion behaviour of TPMS structures and its influence on the mechanical response of various TPMS structures. • Microstructural characterisation of various structures under various strain rates and temperature conditions. • Finite element analysis of the mechanical response of TPMS structures and validation with experimental findings. The following methodology is planned and discussed in five stages to achieve th

BITS Supervisor

NITIN RAMESHRAO KOTKUNDE

RMIT Supervisor

Dr. Maciej Mazur Senior Lecturer

Other Supervisor BITS

Pardha Saradhi Gurugubelli

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Summary : Over the past 50 years, the UAE has witnessed a significant transformation in its road network, providing citizens with well-paved highways and flyovers. However, escalating traffic demand, construction projects, and rising temperatures pose challenges, particularly road pavements. The expected temperature increase can impact pavement structural performance of asphalt pavements, which make up for more than 90% of the roads in the UAE. To address this, continuous monitoring is essential, moving from traditional periodic inspections to intelligent remote monitoring using IoT and embedded sensing technologies. A Digital Twin Model (DTM) is crucial, synchronizing real-time data from sensors to create a virtual representation of the pavement structure. This approach, gaining popularity across various infrastructure sectors, enables dynamic updates for timely maintenance decisions. A Digital Twin Model (DTM), as demonstrated for an asphalt pavement in Ireland, is essential, synchronizing real-time data from sensors to create a virtual representation of the pavement structure. The primary objective of this project is to develop a 3D digital twin model of asphalt pavement structure, to monitor the structural changes under real-time loading and environmental conditions. This can be achieved through the following research objectives. Objectives : 1) Develop a DTM using a 3D Finite Element Method (FEM) based model to monitor the critical pavement responses 2) Develop machine-learning techniques to predict the heat distribution in pavement layers from weather data and temperature probes 3) Embed temperature and strain sensors at various depths of pavement layers 4) Establish the physical-to-virtual connection to update the DTM model parameters 5) Update the structural reliability, and propose the optimal timings for pavement maintenance Methodology : The development of the DTM to monitor the structural responses of the pavement is carried out in two phases in this project. Phase 1: a thermal digital twin model is developed to predict the predict heat distribution in the pavement layers. Phase 2: the predicted temperature distribution will be used as input, along with the field strain measurements, to build the 3D FEM model. Steps : 1) Get telemetry data from the physical road structure 2) Detect and diagnose 3) Develop real-time FEM-based model 4) Model Calibration 5)Result Collection 6)Decision support based on AI algorithm

BITS Supervisor

Deepthi Mary Dilip

RMIT Supervisor

Mojtaba Mahmoodian Dr

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

To meet the ever-growing demand for electrical energy, there is a need to harvest energy from sustainable sources, such as solar, thermal and wind. For these energy harvesting methods to succeed, materials with suitable properties and energy efficiencies are required. This project aims to investigate new materials for their potential energy applications. We will focus on 2D van der Waals materials, which can be used in low dimensional devices due to their size, flexibility, mechanical stability and unique material properties owing to their dimensionality. Various new 2D materials can be designed by functionalization as well as forming van der Waals heterostructures from the classes of existing 2D materials such as transition metal dichalcogenides, phosphorene, MXene etc. The materials that we will design and study in this project would be semiconducting in nature having a band gap within a specific range and band alignment suitable for photovoltaic and photocatalytic applications. First principles-based density functional theory (DFT) calculations and methods beyond DFT will be adopted for calculating the structural, electronic, optical and vibrational properties of the materials. The materials will be assessed for their suitability in photovoltaic and photocatalytic hydrogen production applications via water splitting and other chemical energy conversion reactions.

BITS Supervisor

Prof. Swastibrata Bhattacharyya

RMIT Supervisor

Prof. Michelle Spencer

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Ransomware attack detection and mitigation using SmartNICs Summary Ransomware attacks have become increasingly sophisticated and prevalent, causing significant disruptions and financial losses to organisations worldwide. For example, in December 2022, AIIMS servers were infected with ransomware named Wammacry, Mimikatz and Trojan which encrypted 1.3 TB of data blocking e-hospital applications for several days. Sophisticated malware attacks take months to infiltrate the target networks. During this period, there is ample opportunity for the organisations to detect them at an early stage. Ransomware malware leverages already known legitimate tools such as PsEXec, nssm.exe, mimikatz, quarks pwdump etc. in the operating system which is known as Living Off the Land Binaries (LOLBins) to bypass the scrutiny of antivirus tools. After privilege escalation, malware turns off antivirus tools and downloads the encryption module to encrypt the data. Signature-based approaches which scan each packet or program are not able to detect such attacks. On the other hand, behaviour-based approaches were able to detect such attacks. Behaviour-based approaches monitor system calls, file changes, registry changes, or network activities over a period of time, extract features, and use ML-based methods to classify the behaviour as benign or malignant. Existing behaviour-based approaches are not scalable and are not able to detect attacks in real-time. Features are extracted from recorded traffic in the form of zeek logs or pcap files but not in real-time at line rate. Moreover in today’s world, due to widespread use of mobile internet, malicious traffic can come into the network bypassing the perimeter security such as firewalls and IDS. Collecting features from a few perimeter devices alone is not sufficient to capture the comprehensive behaviour of malware. Existing malware detection approaches implemented in the dataplane do per-packet or per-flow malware detection at the switch level, thus making them incapable of detecting sophisticated malware such as ransomware attacks. The project proposes to implement behaviour-based detection approaches in the dataplane combining the scale of programmable data plane and accuracy of behaviour-based approaches. Crucial aspect of behaviour-based detection is extracting and aggregating features over a period of months of time. Most of this work can be offloaded to dataplane. Each switch/SmartNIC extracts features on its own traffic

BITS Supervisor

Dr Hari Babu Kotakula

RMIT Supervisor

Prof. Iqbal Gondal, Associate Dean, Cloud, Systems & Security

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

The proposed project aims to develop an optimal solution towards sustainable electric vehicle (EV) charging infrastructure by introducing renewable energy sources (e.g; solar PV) integrated long-life battery storage (e.g; Vanadium Redox Flow Battery, VRFB), considering local power distribution grid scenarios of India and Australia. A low-cost smart energy management scheme will be developed to meet dynamic EV load demand. The impact of EV penetration and solar PV on distribution grid operation for both the Indian and Australian scenarios will also be investigated in detail. The proposed work will focus on the four major objectives; 1. Design and development of an efficient electrical interfacing system for long-life battery storage (e.g.; VRFB) with solar PV. 2. Development of an optimized battery management system (BMS) for 5kW 30kWh VRFB storage. 3. Development of IoT-based smart energy management scheme. 4. Investigations on the impact of dynamic EV penetration on local distribution grid parameters. The following available facilities will be used during the project; Indian supervisor side: 5kW 30kWh VRFB set-up, Power electronic test-bed, Programmable load bank for emulating EV penetration profiles, Raspberry-Pi communication kit. Australian Supervisor side: Solar and distribution grid simulator, EV lab facilities, DIgSILENT Power Factory software. The methodology is presented by following work packages; WP1: Integrated solar PV – VRFB – distribution grid design and simulation WP2: Efficient BMS design for VRFB storage WP3: IoT-based low-cost smart energy management scheme WP4: Impact of EV penetration on the local distribution grid WP5: Validation of the developed solution for both the Indian and Australian scenarios.

BITS Supervisor

Dr. Ankur Bhattacharjee

RMIT Supervisor

Dr. Kazi Hasan

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

The operational aspects of future power systems are expected to be influenced significantly by the increasing grid integration of electric vehicles (EVs). This PhD project will analyse the impact of EVs on future electricity grids and will investigate the enabling technologies to facilitate high EV integration into the electricity grids. This project aims to explore the following research contents: • Development of models for spatiotemporal EV charging patterns at the distribution substation based on data-driven approaches; • Articulation of an appropriate simulation model of a distribution network considering correlations among EV, solar PV, and system loads; • Detailed time-sequential simulation studies with collected historical data on the EV grid impact assessment; • Identification and validation of the mitigation techniques for EV grid integration problems; and • Provision of recommendations on the planning and operational strategies of the EV-rich distribution networks. To achieve the outcomes as per the abovementioned objectives, the following methodology will be followed: • Collecting historical EV charging data from a wide range of customers over the entire annual cycle • Developing spatiotemporal EV charging profiles in a representative distribution network • Performing power system simulation with the spatiotemporal EV data in a combined MATLAB-DIgSILENT power factory software package • Identifying distribution grid problems, including voltage violation, line overloading, and substation capacity limit violation • Proposing techno-economic enabling technologies for grid integration of electric vehicles • Validating the proposed solution approaches considering representative network operational scenarios

BITS Supervisor

Dr. Alivelu Manga Parimi and Professor

RMIT Supervisor

Dr. Kazi Hasan

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Recently, environmental pollution-induced respiratory and pulmonary ailments have emerged as a major health concern for the Indian and global medical community. Consequently, the non-invasive, low-cost, and reliable detection of different respiratory disease biomarkers in the human breath environment is considered to be of primary importance for disease detection, early diagnosis, and treatment monitoring. Compared to traditional diagnostic methods such as blood and urine tests, exhaled breath analysis offers advantages in terms of simple and real-time sample processing. The inorganic nanomaterials (both 1D and 2D) have emerged as an attractive candidate for electrochemical gas (CH4, NH3, and H2 etc) sensor design for breath analysis due to the high surface-to-volume ratio, a large number of surface reactive sites, diversity in their surface chemistry, and distinguishable electronic/transport properties that can be exploited in favor of high-performance gas sensor design. The sensitivity and selectivity of a specific gas sensing significantly depend on the structure, chemical compositions, and shape/size of such materials. However, inorganic nanomaterials mostly experience cross-sensitivity to various gases, making it difficult to discriminate between different analytes. The reproducibility of sensing performance among different batches of nanomaterials and their long-term stability under operational conditions are difficult to achieve. The sensitivity of inorganic nanomaterials is substantially less and often found to be affected by various environmental factors such as temperature, humidity, and presence. Thus, in this project, we will be using various kinds of inorganic-organic hybrid nanocomposite materials for gas sensing applications. As inorganic materials, we will mostly consider oxide materials such as ZnO, TiO2, and SnOx. On the other hand, as organic ligands, we will mostly focus on TCNQ fibrous growth. The composite materials will be thoroughly characterized by different techniques, such as FT-IR, XPS, PXRD, EPR, and DSC-TGA. From the response, we will estimate the sensitivity, and selectivity of the system, and evaluate the thermodynamic parameters of the interactions. This will provide insight on how we can improve the sensor performance in terms of sensitivity, selectivity, and response time while changing the composite morphology and content as well as operating conditions such as temperature.

BITS Supervisor

Nilanjan Dey

RMIT Supervisor

Dr Ylias Sabri

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Carbon nanodots (C-dots) are a new type of carbon-based ultra-small nanomaterials. They possess a variety of properties such as luminescence, anti-corrosiveness, sensing capability, high solubility, biocompatibility, non-toxicity, easy synthesis and cost-efficiency, and the capability of accepting and donating electrons. C-dots have shown advantages in many fields, including bioimaging, biosensing, and energy conversion. Recent research has indicated that C-dots can increase the strength of polymer networks. This project will investigate how to empower fiber-reinforced polymer composites (FRC) with C-dots. FRC possess excellent structural properties such as high aspect ratio, high flexibility, lower imperfections owing to their small size, high resistance to corrosion, and fatigue failure. Consequently, FRC have served and assisted in the growth of a wide variety of industrial sectors like aeronautical and aerospace, automotive, high-speed trains, marine, military, sports, wind energy, civil infrastructure, biomedical prosthetics, etc. The properties of FRC primarily depend on the properties of constituents and the extent of interfacial adhesion between the filler and the matrix. High surface area nanoparticles like carbon nanotubes and graphene have been extensively researched to modify the fiber/matrix interface. However, keeping in mind the multifunctional requirements of modern-day engineering structures, C-dots look fascinating. A detailed study is proposed here to incorporate C-dots as a secondary filler in FRC to prepare a hybrid composite and investigate the effects on the composite's mechanical, electrical and sensing capabilities, followed by characterisation including SEM, XRD, FT-IR, etc. This can be achieved by preparing a C-dots modified polymer sizing to be coated on the fiber surface, and/or direct dispersion of C-dots in the matrix. Wettability analysis carried out through drop-on-fiber test will tell the efficacy of polymer modification. Laminate composite structures often contact metallic parts, so the effect of anti-corrosive properties of the hybrid composite will also be assessed. Another interesting aspect to be explored is the damage-sensing capability of the hybrid composite during the progressive failure of laminates. To this end, the inherent electrical conductivity as well as the luminescence of carbon nanodot particles can be used for continuous structural health monitoring, crack detection, and aging resistance.

BITS Supervisor

Harpreet Singh Bedi

RMIT Supervisor

Dr Lei Bao

Other Supervisor BITS

Dr Ravindra G Bhardwaj

Other Supervisor RMIT

Required discipline background of candidate

Project Description

This project seeks to revolutionize patient data management in healthcare by harnessing the power of blockchain technology to overcome the critical limitations of current centralized systems. Motivated by the urgent need for robust data security, enhanced patient control, and improved interoperability, our initiative aims to create a decentralized platform that not only secures patient data against breaches but also empowers patients with unprecedented control over their personal information. The project is driven by the goal to address specific challenges including scalability—to efficiently handle the increasing volumes of healthcare data, privacy and security—to protect sensitive patient information while ensuring it remains accessible for authorized use, interoperability—to facilitate seamless data exchange across various healthcare systems, and data ownership and consent management—to enable patients to manage who has access to their data and for what purpose. Furthermore, it aims to lay the groundwork for the integration of medical AI with blockchain, opening new avenues for AI-based research that can leverage secure, anonymized patient data to advance healthcare outcomes. The project expects to develop novel blockchain scalability solutions tailored for medical data management, innovate privacy-preserving techniques based on zero-knowledge proofs, and create APIs for data interoperability along with smart contracts for data ownership and consent management. The importance of this research lies in its potential to provide a blockchain-powered healthcare data management system that not only meets the complex demands of modern healthcare systems but also promotes the ethical use of data in AI-based research, paving the way for innovations that could transform patient care and medical research methodologies.

BITS Supervisor

Dr. Ashutosh Bhatia

RMIT Supervisor

Abebe Diro

Other Supervisor BITS

Prof. Kamlesh Tiwari

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Summary: Moisture sensors have applications in various sectors including agriculture, automobiles, food processing industry, medical equipment, pharmaceutical and waste water treatment plants, to name but a few. Traditional moisture content estimation includes thermogravimetric analysis and multi-spectral remote sensing-based approaches relying on absorptive properties of water in near-infrared or shortwave-infrared wavelengths. However, the conventional moisture measurement methods involve cumbersome steps and lack sensitivity. Hence, there is an unmet need to come up with a simple and sensitive tool for real time quantification of moisture. Aims: 1. Investigation of suitable coatings on sensor surfaces for more moisture adsorption sites 2. Development of cost-effective and scalable sensor for rapid and real time monitoring of moisture 3. Comparison of the developed moisture sensor with the state-of-the-art techniques Methodology: Quartz crystal resonator (QCR) is entirely electronic and simple in configuration and has therefore gained wide attention as a rapid and on-line detection sensor in both liquid and gaseous mediums [1]. Adoption of fixed frequency drive (FFD) method in conjunction with microfluidics and a temperature controlled QCR will be an exemplary application for real-time monitoring of moisture. In FFD method, a QCR is driven continuously at a fixed frequency and amplitude and the response is analytically interpreted to obtain resonance frequency and dissipation shifts employing imaginary and real components of experimentally recorded electrical impedance respectively. FFD technique will be applied to detect moisture employing coated QCRs which are obtained using either drop casting, spin coating or spray coating techniques. Various coatings including graphene oxide and indium oxide quantum dots, to name but a few will be explored for the project. The interaction between coating modified QCR surface and moisture will be quantified in terms of QCR resonance frequency and dissipation (acoustic energy loss) shifts with respect to baseline signal devoid of moisture. Initial experiments will be carried out using 14.3 MHz QCRs. QCRs with higher fundamental frequencies ranging from 50 to 250 MHz will be explored for sensitive detection of moisture. Reference [1] A. Alassi, M. Benammar, and D. Brett, “Quartz crystal microbalance electronic interfacing systems: A review,” Sensors (Switzerland), vol. 17, no. 12, pp. 1–41, 2017, doi: 10.3390/s1712

BITS Supervisor

Dr Arnab Guha, Assistant Professor

RMIT Supervisor

Dr Henin Zhang, Senior Lecturer

Other Supervisor BITS

Satish Kumar Dubey

Other Supervisor RMIT

Dr Amirali Khodadadian Gostar, Senior Lecturer

Required discipline background of candidate

Project Description

Background – In the age of foundational models that are constructed based on deep learning architectures (e.g., transformer models), we now possess the capability to process vast datasets to learn comprehensive representations of data, forming generalizable knowledge across different applications. So far, most advancements in foundation models have focused on a single data modality. For example, large language models (LLMs), such as GPT-3, BERT, and RoBERTa, are purely text-based. They excel in tasks like text generation and encoding, but they lack a comprehensive understanding and processing of other data types. However, multimodal perception is a fundamental component for achieving general artificial intelligence, as it is crucial for knowledge acquisition and interaction with the real world. Furthermore, the application of multimodal inputs greatly expands the potential of language models in high-value domains, such as multimodal robotics, document intelligence, and robot technology. Hence, development of multimodal LLMs that seamlessly integrate multiple data types has attracted extensive attention recently. In this project, we seek to address the pressing need for robust multimodal language models (LLMs) tailored specifically for clinical and healthcare data analysis. While recent advancements, exemplified by GPT-4's ability to process both images and texts, mark significant progress in multimodal LLMs, there remain notable challenges within the domain of healthcare data integration. Our project aims to fill this critical gap by developing a foundational model capable of seamlessly integrating diverse data modalities prevalent in clinical and healthcare settings. By leveraging advanced deep learning architectures and large-scale datasets, we seek to enhance the performance and versatility of multimodal LLMs, thereby facilitating more comprehensive and accurate analyses of medical information.

BITS Supervisor

Professor Poonam Goyal

RMIT Supervisor

Dr Jiayuan He

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

Quality control measures for alcoholic beverages, especially wines, are of paramount importance to ensure that wines are safe for consumption. This includes monitoring for contaminants such as pesticides, heavy metals, and microbiological hazards, which could pose health risks to consumers if present above permissible levels. This also ensures that each batch of wine maintains consistency in flavour, aroma, and overall sensory characteristics. This is particularly important for brands that strive to offer a distinct taste profile, as any deviation from the expected flavour can disappoint consumers and harm brand loyalty. Furthermore, proper quality control measures help ensure that wines are aged and stored correctly, preserving their quality over time. Factors such as temperature, humidity, and exposure to light can affect the aging process and ultimately affect the taste and quality of wine. Quality control should extend beyond the winery to the entire supply chain, including grape growers, suppliers of equipment and packaging materials, and distribution channels. By implementing rigorous quality control measures at every stage, wineries can ensure that only the highest quality ingredients and materials are used in the production process. The adulteration of alcoholic beverages, including wine, can occur through various means. The practice of adding water or excess sugar without regard to quality can lead to imbalances in flavour and may violate regulatory standards. Sometimes, despite their hazardous impact on the body, artificial colourants or synthetic flavour enhancers are used to improve the taste of low-quality wines or to cover undesirable characteristics. Some producers add chemical preservatives, such as sulfites, sugar, acids, and glycerin, to prolong the shelf life of alcoholic beverages and prevent spoilage. This project aims to develop optical sensors for the detection and estimation of chemical adulterants and preservatives in commercially available alcoholic beverages such as wine. In addition to detection, we will attempt to remove artificial adulterants from alcoholic beverages without hampering their flavour, aroma, and taste. This low-cost sensor will enable the preservation of the taste and quality of beverages. Additionally, the sensors and technologies developed through this project will be benchmarked with industry standards in collaboration with industry partners from the RMIT supervisor.

BITS Supervisor

Nilanjan Dey

RMIT Supervisor

Rajesh Ramanathan, Professor

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

The digital age has brought unprecedented business opportunities but has also introduced new security challenges. This project aims to address the threats posed by the arrival of quantum computing to traditional encryption methods and explore the potential of blockchain technology to provide a secure and tamper-proof solution. The primary objectives of this project are twofold. First, it seeks to investigate the vulnerabilities of existing cryptographic systems to quantum computing attacks and evaluate the readiness of businesses to adopt post-quantum cryptography. Second, it aims to explore integrating quantum-resistant cryptographic methods, such as lattice-based and hash-based signatures, into blockchain platforms. The methodology will involve a comprehensive literature review to understand the state-of-the-art in quantum computing, blockchain technology, and post-quantum cryptography. Computational resources will be allocated to run simulations and benchmarking tests on various cryptographic algorithms and blockchain implementations. Additionally, test networks will be established to evaluate the performance and security of quantum-safe blockchain platforms. Through this project, we aim to raise awareness among businesses and organisations about the potential risks posed by quantum computing to their data security and the need to adopt quantum-resistant strategies. Furthermore, a framework will be developed to assess organisations' readiness for post-quantum cryptography and provide recommendations for implementing quantum-resistant blockchains. By redefining business security in the digital age, this project will contribute to developing robust and future-proof solutions, ensuring the protection of sensitive data, and promoting trust in digital transactions. The findings and recommendations will be disseminated through top-tier academic publications, industry workshops, and outreach activities, empowering businesses to navigate the challenges of the quantum era.

BITS Supervisor

Dr Nikumani Choudhury

RMIT Supervisor

Malka N. Halgamuge

Other Supervisor BITS

Prof. Barsha Mitra

Other Supervisor RMIT

Dr Akanksha Saini

Required discipline background of candidate

Project Description

Political scientists and researchers are concerned about the impact of fake news on democracy. On the other hand, fake news could also escalate to life-threatening problems, for example, UNESCO states that during COVID-19 pandemic fake news is putting lives at risk, like Coronavirus (COVID-19) pandemic. Fake news continues to spread, so does people's behaviour and emotions about the fake news via social media platforms. Fake news via social media platforms. This opens up the back door for cyber-criminals to entice people (i.e., taking advantage of victims' emotional and behavioural aspects) to click on links (e.g., phishing links) associated with fake news when reading. Therefore, in this project we investigate how people's emotional and behavioural features influence reading and diffusing fake news in social media and proposes a fake news detection model incorporating people's behavioural features and their emotions to better detect fake news in social media. Furthermore, we study how individuals are emotionally manipulated to perform various malicious behavioural activities such as clicking on a link associated with fake news. The project will also explore fake news diffusion patterns and attacker strategies in social media. The research findings can be used in a risk prediction model to predict the risk score of any news content being fake in social media. Eventually, a browser plugin can be implemented through persuasive design principles for social media platforms that alert the potential risks to users when dealing with (fake) news. Methodology: The proposed methodology consists of several steps. First, we extract textual sentiment, visual sentiment, behavioural and metadata features from fake and real news in a dataset. Then, we analyse the data and evaluate the correlations between these set of features, as well as compare features between real and fake content. Based on the analysis, we identify important and impactful features for fake news detection, i.e., to distinguish between real and fake news. Then, we build a classifier using machine learning techniques, which will try to classify a post as fake or not, using these features as training data. In the following subsections, we will describe our methodology in detail. Finally, a browser plugin is implemented through persuasive design principles (i.e., HCI methods) for social media platforms that alert the potential risks to users when dealing with (fake) news.

BITS Supervisor

Yashvardhan Sharma and Professor, Department of Computer Science & Information Systems & Faculty In-

RMIT Supervisor

Nalin Asanka Gamagedara Arachchilage

Other Supervisor BITS

Other Supervisor RMIT

Prof. Matthew Warren

Required discipline background of candidate

Project Description

With the rise in energy demand, exploring efficient and economical ways of sustainable energy production techniques to reduce our dependency on fossil fuels, the primary source of the global greenhouse effect and climate change is essential. The energy requirement perceived an unparalleled growth in renewable energy production, with wind energy as a leader among others. In 2019, 7.3% of U.S. energy requirements were met by wind energy, with that percentage forecasted to increase to 20% by 2030 and 35% by 2050. In the wind turbine system, the mechanism involves the rotation of the rotor, which is directly connected to a generator through different bearings and a series of gearboxes (which speed up the rotation). The transition of wind energy to the rotation of a generator is what produces electricity. The premature failures (i.e., surface-initiated fatigue, white etching cracks [WECs], scuffing/smearing, spalling, and dents/indentations) of bearing in wind turbines are the prime concerns and significant causes of downtime in energy production. For this reason, the project aims to develop a high entropy alloy (HEA) coating with constituent elements of equal molar ratio, which will be fabricated using the magnetron sputtering technique. The investigating hypothesis will be the effect of metal disulfide content on the microstructure, composition, phase constitution, and wear and corrosion phenomenon on the magnetron sputtering HEA coatings. A tribological system consists of two same/different materials rolling/sliding against each other in dry or lubricating conditions. The investigating hypothesis will be how HEA hard-coated and lubricated tribological pairs could improve the endurance limit of the wind turbine bearings. In this, the P.I. aims to develop a hard and lubricated coated tribological pair that can support heavy axial and radial turbine load with minimum friction. The principal investigator will focus on developing an advanced tribological system consisting of Titanium alloy-coated bearing rollers in linear contact with the metal disulfide-coated raceways in dry and lubricated rolling conditions. The lubricant incorporated with metal disulfide nanoparticles as a friction modifier will be further tested for its auxiliary lubricating behavior in lubricated conditions. Thus, developing and investigating effective and economically advanced tribological systems for wind energy turbines for controlling turbine bearing failures are highly desirable.

BITS Supervisor

Piyush Chandra Verma

RMIT Supervisor

Prof. Nasir Mahmood

Other Supervisor BITS

Himanshu Aggarwal

Other Supervisor RMIT

Prof Raj Das

Required discipline background of candidate

Project Description

Harmful Algal Blooms (HABs) are increasingly recognised as critical global environmental issues, with significant repercussions for ecosystems, economies, and public health. These blooms, often exacerbated by the influx of nutrients from agricultural runoff, sewage discharge, and the broader impacts of climate change, lead to the proliferation of toxic algae in water bodies. Such proliferation not only diminishes water quality but also threatens aquatic life through oxygen depletion and toxin release, posing serious risks to human health, affecting the livelihoods dependent on fisheries, and undermining recreational water-based activities. Traditional methods for monitoring HABs, which typically involve in-situ sampling followed by laboratory analyses, are not only resource-heavy but cannot provide a comprehensive picture of the blooms' spatiotemporal dynamics, making it challenging to respond effectively to their rapid development and spread. The emergence of remote sensing technologies marks a paradigm shift in detecting and monitoring HABs. Equipped with advanced high-resolution multispectral and hyperspectral sensors mounted on satellites and Unmanned Aerial Vehicles (UAVs), these technologies offer a synoptic, timely, and cost-efficient means to observe and analyse the extent and biomass of algal blooms across vast and often inaccessible aquatic environments. Such technological advancements enable the continuous surveillance of water bodies, offering a valuable tool for environmental monitoring and management practices aimed at mitigating the impacts of HABs. This research aims to refine remote sensing techniques for enhanced HAB monitoring by tackling challenges in detecting blooms in turbid and coastal waters and developing precise models for estimating HAB characteristics. It will improve atmospheric correction algorithms, create and test new analytical and machine learning models, and integrate multisource remote sensing data for detailed HAB analysis. Cloud computing will enable efficient processing of large datasets for timely global HAB surveillance. Additionally, standardising methodologies and reporting in HAB remote sensing studies will boost the reliability and comparability of findings.

BITS Supervisor

Anirban Roy, Associate Professor

RMIT Supervisor

Dr Amirali Khodadadian Gostar, Senior Lecturer

Other Supervisor BITS

Snehanshu Saha and Professor

Other Supervisor RMIT

Dr. Sara Vahaji

Required discipline background of candidate

Project Description

1. Graphene-based polymer nanocomposites can be fabricated using solvent processing, in-situ polymerization, and melt blending. Polymers as flexible TE materials provide low thermal conductivity and mechanical flexibility. But the low electrical conductivity of polymers can be fixed by combining graphene derivatives with conjugated polymer composites. Ultrasonic and molding methods are used to create PVA nanocomposites for TE films. 2. Characterization includes thickness and mass measurement, hardness testing, thermal stability, oxidative stability, lifetime of the product, moisture, and volatile components using the Thermogravimetric Analyzer (TGA). The sheet resistance of thin films is measured using a four-point probe system. The Seebeck coefficient can be obtained with a four-probe setup along with two T-type thermocouples, two copper wires, a Keithley 2000 multimeter, a temperature controller, and data acquisition software. The electronic carrier concentration and mobility can be found using the ASTM F76-08 method using a van der Pauw geometry setup. An optimization study will be carried out to find the optimum concentration of graphene nanofillers in the polymer. 3. The present proposal also incorporates a numerical simulation study using COMSOL. The thermoelectric behavior of the system will be modeled in the COMSOL software and analyzed by varying parameters such as temperature and bending stress to optimize the thermoelectric behavior and increase the TE efficiency. 4. The optimization of material and thermoelectric device characteristics will be done using machine learning techniques.

BITS Supervisor

Dr Ravindra G Bhardwaj

RMIT Supervisor

Prof. Xu Wang

Other Supervisor BITS

Harpreet Singh Bedi

Other Supervisor RMIT

Professor Sumeet Walia

Required discipline background of candidate

Project Description

1) Summary: The proposed project aims to utilize Artificial Intelligence (AI) enabled digital twin technology for monitoring the structural integrity of Reinforced Concrete (RC) girder bridges. These bridges, subject to repetitive loading and environmental stressors like seismic activities and wind loads, face significant risks of structural damage. To ensure their continued safety amidst gradual deterioration, systematic monitoring is essential, necessitating the use of advanced methods like structural health monitoring (SHM). The project involves selecting an RC bridge for monitoring, creating a validated digital model aligned with the physical structure, and generating diverse vibration response datasets. These datasets are crucial for training AI models using advanced algorithms to identify damage location and extent accurately. By integrating AI and digital twin technology, the project aims to enhance structural health assessment, ensuring robust vigilance and resilience in critical structures like RC bridges 2) Aim & Objectives: The aim is to develop an advanced SHM system utilizing digital twin modelling and AI to proactively detect and assess structural damages in RC girder bridges, ensuring safety, averting failures, and enhancing infrastructure efficiency. The objectives are: • To develop a digital model technology for field tested RC girder bridge with validations • To generate a vibration response data set of the chosen structure using digital model with possible structural damages at different locations and extent of damage. • To train the AI models to identify the location and amount of damage in the RC girder bridge using suitable machine learning algorithms. • To enable an automated structural health monitoring system to continuously track the condition of a RC girder bridge. 3) Methodology: 1. Developing a digital model of the RC bridge using digital twin technology. 2. Validating the model with real bridge vibrations using accelerometer sensors. 3. Introducing simulated damages to the model to create a comprehensive database. 4. Processing data for feature extraction like wavelet transform or spectrograms. 5. Training AI models to identify damage and implementing an automated monitoring system. 6. Validating the AI and digital twin technique's ability to identify real damage in tested RC bridges.

BITS Supervisor

Mohan S C Dr

RMIT Supervisor

Mojtaba Mahmoodian Dr

Other Supervisor BITS

Other Supervisor RMIT

Prof. Sujeeva Setunge

Required discipline background of candidate

Project Description

Environmental pollution is a global phenomenon. There are different types of pollutants, among which organic chemical contaminants (phenolic compounds, petroleum, pesticides, pharmaceuticals, etc.) may adversely affect the ecosystem and induce microbial resistance to antibiotics. Various remediation techniques have been applied for organic contaminants in water; however, adsorption with catalytic activity can be considered one of the most promising due to several environmental and techno-economic merits, and sometimes, such techniques may produce value-added chemicals from the pollutants. To identify the optimal material for organic chemical degrading might be challenging. Polymer nanocomposites (PNCs), combining the benefits of nanoparticles with polymers, may open up new possibilities for overcoming this difficulty [1]. References 1. https://doi.org/10.1016/j.scitotenv.2023.165772 2. https://doi.org/10.1016/j.jenvman.2022.116596 3. https://doi.org/10.1002/slct.201900470 Global surplus agro-residue (AR) generation rate is around 3300 MT, which is underutilized and could be a cause of environmental pollution for stubble burning. We aim to utilize this AR to synthesize high surface area and porosity carbonaceous materials (AC, GO) and then use those as a part of the nano composite with polymer to adsorb the pollutants. Nanoscale range zero valent transition metals or their bimetallic compositions are extremely efficient in eliminating different pollutants present in our external environment [2]. Those nanoparticles would be immobilized in some polymers to provide them to react with the targeted contaminants, while inhibiting their reactivity with the surroundings [3]. Soil and/or water remediation will be target of this work.

BITS Supervisor

Prof. Banasri Roy

RMIT Supervisor

Dr. Fugen Daver - Associate Professor

Other Supervisor BITS

Dr. Sarbani Ghosh - Assistant Professor

Other Supervisor RMIT

Prof. Namita Roy Chowdhury

Required discipline background of candidate

Project Description

The increasing complexity and distributed nature of AI model supply chains, often spanning multiple organizations and third-party providers, present significant security challenges. These challenges can lead to severe consequences, including data breaches that expose sensitive information, model tampering that undermines the integrity of AI systems, and unauthorized access that enables malicious exploitation. To proactively address these concerns and their potential impact, this project aims to enhance the security of AI model supply chains by developing a comprehensive and adaptable security framework. This framework will be grounded in a rigorous identification and analysis of vulnerabilities and threats across the entire AI model lifecycle, from data collection and model development to deployment and maintenance. Current security approaches often fall short in addressing the unique challenges posed by AI model supply chains, such as the dynamic nature of models, the reliance on third-party components, and the potential for adversarial attacks. By conducting a meticulous risk assessment and threat modeling, this project will identify and prioritize specific security risks, enabling the development of targeted mitigation strategies. The resulting framework will be structured and adaptable, providing clear guidelines and best practices for securing AI model supply chains across diverse organizations and use cases. The framework's effectiveness and practicality will be rigorously evaluated and validated through real-world scenarios and case studies, with a focus on quantifiable metrics such as reduction in vulnerabilities, improved detection rates for threats, and increased overall resilience of AI systems. The successful implementation of this framework is expected to foster greater trust and confidence in the deployment of AI systems, promoting their responsible and secure use across various domains.

BITS Supervisor

Dr. Ashutosh Bhatia

RMIT Supervisor

Abebe Diro

Other Supervisor BITS

Prof. Kamlesh Tiwari

Other Supervisor RMIT

Required discipline background of candidate

Project Description

This project aims to explore the integration and application of Generative Artificial Intelligence (AI) and Digital Twin technology within the healthcare sector, specifically focusing on their potential to revolutionize medical research, diagnosis, and treatment methodologies. Generative AI, a cutting-edge technology capable of producing images, text, and various media forms from human prompts, holds significant promise in healthcare innovation. It can contribute to personalized medicine, predictive diagnostics, and tailored treatment plans. Concurrently, Digital Twin technology, which involves creating virtual replicas of physical entities, presents an opportunity to advance patient care by developing individualized digital counterparts. These digital twins can simulate patients' health conditions in real time, offering unprecedented insights into disease progression, treatment outcomes, and personalized healthcare interventions. Incorporating these technologies into the Smart City paradigm, this project envisions developing a robust healthcare system that leverages the capabilities of Generative AI and Digital Twin technology. This system will enhance patient care, improve health outcomes, and optimize resource allocation in future smart cities. By harnessing the synergy between these advanced technologies, the project aims to establish a healthcare framework that is innovative and responsive to the dynamic needs of urban populations, setting a new standard for healthcare delivery in the digital age.

BITS Supervisor

Vinay Chamola

RMIT Supervisor

Ibrahim Khalil

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

The gold standard for measuring cardiovascular parameters is the electrocardiogram (ECG). The ECG measures small changes in electrical potential at the surface of the body caused by contraction of the heart. ECG can provide a rich picture of cardiac health and disease. However, measuring the ECG requires specialised skill and equipment, and is typically not compatible with medical imaging modalities such as magnetic resonance imaging (MRI). There is a need for reliable non-invasive techniques for physiological monitoring to aid faster and comfortable diagnosis of cardiovascular health. Remote photo-plethysmography (rPPG) is an alternative technique for estimating heart rate from facial videos by measuring changes in reflected light intensity caused by changes in blood flow in the skin. The accuracy of rPPG is typically limited by the confounding effects of skin colour and interference due to changes in illumination. A related technique, image ballistocardiography (iBCG), also uses facial video to estimate heart rate but uses micro-vibrations of facial features caused by propagating pressure waves in the blood vessels. This technique is more robust to skin pigmentation and changes in illumination, but is susceptible to noise due to the rigid and non-rigid motion of the subject. Combining rPPG and iBCG derived from facial video has the potential to produce robust physiological estimates with accuracy exceeding either technique alone. This project therefore aims to investigate estimation of heart rate and a rich set of physiological parameters from simultaneous rPPG and iBCG signals extracted from facial videos. This work has the potential to enable low-cost, non-contact, physiological monitoring of cardiovascular health in the home or clinics.

BITS Supervisor

Amalin Prince A

RMIT Supervisor

Shaun Cloherty

Other Supervisor BITS

Other Supervisor RMIT

Dr Priya Rani

Required discipline background of candidate

Project Description

Summery: In the current proposal, we are aiming to develop advanced and reliable gas/VOC sensors for breath marker detection. The advanced 2D materials are the broad alternative of the metal oxides in gas/VOC sensing applications. Their inherent advantages, such as single atom thickness which provides large effective surface area for molecular adsorption, low carrier concentration which can modulate conductivity easily due to adsorption of a few molecules of target analytes, high mobility which provides an exceptional fast carrier transport and finally the 2D materials having easy functionalization possibility. The gas sensing performance of 2D materials can be enhanced using dopants/ nanocomposite/ heterostructures by modifying the band structure, charge transfer mechanisms, physisorption, and surface reactivity, which enhances its capacity for gas adsorption. Aims: 1. Chemical vapor deposition (CVD) growth of few layered 2D transition metal chalcogenides (TMDCs) e.g. MoSe2, WSe2, MoTe2 etc. 2. Integration of various oxide/sulfide materials with few layer TMDCs (selenides and tellurides) to develop various 2D heterostructures. 3. Morphological, structural and chemical characterizations of the pure and heterostructured 2D chalcogenides. 4. Fabrication of sensors/sensor array by implementing interdigitized noble metal electrodes on various substrates like SiO2/Si, Al2O3 etc. 5. Testing of sensing capabilities of the sensors in the exposure of popular breath markers gases/VOCs like NH3, acetone, NO, ethanol, CO, H2S etc. Methodology: 2D chalcogenide synthesis will be performed with thermal and low pressure CVDs. A few layer/monolayer 2D selenides/tellurides will be oxidized under controlled oxygen ambient to introduced heterostructure like MoO3/MoSe2 with multilayer stack like structure. Controlled sulfurization of 2D selenides/tellurides will be implemented to develop MoS2/MoSe2 or WS2/WSe2 type heterstructures. The simultaneous use of two metal oxide sources or co-deposition of two different metals through DC sputtering and further sulfurization/ selenization through CVD route can provide different 2D heterostructures. As grown/synthesized 2D materials will widely be characterized using various in-house (RMIT/BITS) analytical techniques. To fulfill the requirement of sensor fabrication, both the institutes (BITS/RMIT) have state of the art fabrication facilities. The cutting edge gas sensing facilities are available with both the research groups.

BITS Supervisor

Prof. Arnab Hazra

RMIT Supervisor

Prof. Yongxiang Li

Other Supervisor BITS

Other Supervisor RMIT

Required discipline background of candidate

Project Description

An increased awareness of environmental, social, and economic impacts associated with business operations has driven supply chains towards sustainability. As supply chain links multiple functions, including logistics, production, procurement, and marketing and sales, an integrated end-to-end approach is essential. AIs capacity to analyze vast amounts of data and comprehend intricate relationships assists in developing sustainable supply chains by identifying opportunities to reduce waste, enhance transparency, and support ethical practices. Procurement function accounting for 50 to 80 percent of a company’s cost has potential to implement AI that drives sustainable supply chains (McKinsey, 2023). Through scraping websites for data on finances, sustainability scorecards, diversity scores, and customer ratings of suppliers, AI has the potential to scout for new suppliers who are sustainable. Meanwhile, onboarding new suppliers is expensive and time-consuming. So, AI can be used to assess whether existing suppliers can provide additional materials. AI technologies can monitor existing suppliers' ethical and social practices by analyzing data related to labor conditions, human rights, and environmental impact. These practices ensures that suppliers are in align with environmental and social responsibility standards of supply chain. Moreover, adoption of AI in procurement process can increase organisation’s ROI through reducing the cost associated with procurement activities (KPMG, 2023). Given the importance of AI in procurement, industry professionals are looking for avenues of implementation while research is limited (Allal-Chérif et al., 2020; Guida et al., 2023). Hence, the aim of the research is to use AI based techniques such as genetic algorithm in driving procurement process towards sustainable supply chain. In 2022-23, agriculture sector in India contributes 18.3% to the Gross Value Added(GVA) and has generated an export revenue of $50.2 billion (Economic Times, 2023). Despite the significance, the agricultural supply chain efficiency is impacted by the shortages of resources such as land, water, and soil health, along with exhausted work force. As a result, there is a growing pressure on Indian agriculture sector to move from resource-intensive methods to sustainable and environment-friendly farming practices.  sustainable agriculture sector in India has a potential to emerge as a significant supplier to international markets seeking sustainability.

BITS Supervisor

Dr. Aneesh Chivukula

RMIT Supervisor

Dr. Aswini Yadlapalli

Other Supervisor BITS

Professor Srikanta Routroy

Other Supervisor RMIT

Required discipline background of candidate

Project Description

The integration of advanced technology in healthcare has revolutionized patient outcomes in numerous ways. From the advent of electronic health records (EHRs) to the use of artificial intelligence (AI) in diagnostics and treatment planning, the healthcare sector has witnessed a transformation that has significantly improved patient care and outcomes. One of the most notable advancements is the implementation of EHRs, which has streamlined the management of patient information. EHRs ensure that a patient’s medical history is readily available to healthcare providers, leading to more accurate diagnoses and personalized treatment plans. This accessibility reduces the chances of medical errors, which can be life-threatening. Moreover, EHRs facilitate better coordination among healthcare providers, ensuring that patients receive comprehensive and continuous care, especially for chronic conditions. Telemedicine is another technological breakthrough that has reshaped patient care. By leveraging telecommunication technologies, patients can now access medical consultations and follow-up appointments from the comfort of their homes. This has been particularly beneficial for individuals living in remote areas with limited access to healthcare facilities. Telemedicine has also proved to be a critical tool during the COVID-19 pandemic, allowing healthcare systems to continue providing care while minimizing the risk of virus transmission. Studies have shown that telemedicine can be as effective as in-person visits for a variety of conditions, contributing to improved patient satisfaction and adherence to treatment plans. Artificial intelligence and machine learning are making significant strides in enhancing diagnostic accuracy and treatment efficacy. AI algorithms can analyze vast amounts of medical data, identifying patterns and correlations that might be missed by human clinicians. For instance, AI-powered diagnostic tools can detect anomalies in medical imaging, such as X-rays and MRIs, with remarkable precision. This early detection is crucial for conditions like cancer, where timely intervention can dramatically improve survival rates. Additionally, AI is being used to develop predictive models that can foresee disease outbreaks and patient deterioration, allowing for proactive and preventive measures. Robotic surgery is another area where technology is making a profound impact. Robotic-assisted surgical systems provide surgeons with enhanced precision, flexib

BITS Supervisor

siddhi sharma

RMIT Supervisor

bitsrmit.phd@rmit.edu.au

Other Supervisor BITS

Rupendra Singh

Other Supervisor RMIT

Nicole Woodham

Required discipline background of candidate