Mathematics of Control, Signals, and Systems

Singh G., Singh S., Kumar K., Kaur A.

Over the years the review papers that explored the use of visualizations and interactivity in online algorithm education. In this paper sequence of algorithms are explained visually in an interactive manner that helps to realize the fundamental concept of sorting algorithm in a simple manner. It discusses the significance of interactive learning tools and their impact on student performance. Several innovative tools are introduced, including VSOSA for sorting algorithms and Algorithm Visualizer. The paper highlights the importance of adaptability in interactivity levels to cater to diverse learning styles and emphasizes the value of algorithm visualization in enhancing comprehension. Overall, it provides valuable insights into improving algorithm education and show that our approach provides better perceptive of sorting algorithms.

Jayanthi P., Bhama P.R., Abitha M., Shrijhaa R.

Sign language is a visual form of communication used by deaf and hearing-impaired people. It relies on hand gestures, body language, and facial emotions. Recognition of dynamic sign language is a more challenging problem compared to recognition of static sign language as it often involves sequence of multiple frames. This necessitates a significant memory resource to handle video dataset and high processing power for training the model. This paper presents a dynamic Indian Sign Language (ISL) recognizer system to recognize fifteen dynamic signs. It is built based on the federated learning approach, which utilizes three clients with identically distributed data and a single server system connected through a network for communication. The System takes hand gestures from video frames as input and process them using a custom trained 2-D Convolution Neural Network (2D CNN) model which predicts dynamic sign gestures and translates them into a respective natural language (audio). Experimental results demonstrate that the use of the Federated Learning approach reduces the memory and network requirements in distributed model training, resulting in an 87% accuracy.

Ponni S., Sabarivani A., Janney J.B.

Osteoporosis and Osteoarthritis in the knee are disorders that commonly affect the musculoskeletal system of an individual. It is majorly prevalent in elderly people and has an impact on the quality of their lives. The late diagnosis of these diseases leads to joint dysfunction, decreased mobility, and dependency on regular activities. Therefore, in the present article of review, we discussed works that applied deep learning methods, such as convolution neural networks (CNN), support vector machines and many more such algorithms for feature extraction and classification for early, precise, and rapid disease prediction. It involves the process of acquiring scan images, preprocessing and application of deep learning algorithms to identify the presence or absence of the diseases. Finally, in order to determine the efficiencies, the model's performance is compared and the best model is recommended for real time applications.

Takawale A.J., Paithane A.N.

Analysis of the human body has never diminished and explorations on it have never come to an end. An analysis of EEG for the study of the cognitive process for bio-medical appliances is a current research area. The EEG aids to understand the brain and its activities by offering thorough information on the present status of the brain. This is necessary data since it could aid in identifying or preventing varied abnormalities or diseases or damages in the brain. For that, the precise study of EEG signals is important. A review of 25 works on brain activity recognition (BAR) provides the basis for this survey. The diverse employed schemes in reviewed articles are evaluated that include classification and optimization techniques like Support Vector Machine (SVM) and so on. Moreover, the performance of the reviewed articles is examined and the better performances are analyzed. At last, a chronological review is done and the challenges of BAR are described.

Etuttu M.F., Bhatti G.K.

The potential correlation between renal function test indicators and lipid profiles in the population affected by cardiovascular disease is thoroughly investigated in this dissertation. The study aims to explain about analytical processing correlating renal function test markers and lipid profile in cardiovascular disease. Methodology: Analyzed renal function test and lipid profile in individuals with cardiovascular disease. Results: The patients’ ages ranged from 10 to 81 years. The majority of patients are in the 28–33 age range. The study was supported by 18% of the population. These individuals’ percentage readings for renal parameters were 47%, 51%, and 49%, respectively. A total cholesterol analysis was performed on 6% of the individuals in this investigation. The average number of patients overall was 2, with participants ranging from 0 to 4. Six percent of the recruited patients had triglycerides. The range was 6% to 0%, with 0% serving as the average. Six percent of newly enrolled individuals had high-density cholesterol. Zero was the lowest, zero was the usual, and six was the highest. Six percent of newly enrolled individuals had low-density cholesterol. With a low of o, a normal of 6, and a high of 0. Six percent of newly enrolled patients had very low-density lipoprotein. Normal was zero, low was zero, and high was six. Conclusion: Understanding the relationship between lipid profiles and renal function, demonstrating the high amount of creatinine and the average % of all parameters in the lipid profile.

Yang Y., Tomar A.

The rapid advancement of large language models, such as the Generative Pre-trained Transformer (GPT) series, has had significant implications across various disciplines. In this study, we investigate the potential of the state-of-the-art large language model (GPT-4) for planning tasks. We explore its effectiveness in multiple planning subfields, highlighting both its strengths and limitations. Through a comprehensive examination, we identify areas where large language models excel in solving planning problems and reveal the constraints that limit their applicability. Our empirical analysis focuses on GPT-4’s performance in planning domain extraction, graph search path planning, and adversarial planning. We then propose a way of fine-tuning a domain-specific large language model to improve its Chain of Thought (CoT) capabilities for the above-mentioned tasks. The results provide valuable insights into the potential applications of large language models in the planning domain and pave the way for future research to overcome their limitations and expand their capabilities.

Adithya N.N., Sah P.V.

The contemporary landscape of technology is abuzz with the terms “Artificial Intelligence” and “Machine Learning.“ However, for many, the process of developing and deploying end-to-end machine learning applications can be quite daunting. In this paper, our objective is to elucidate the step-by-step procedure for creating a comprehensive machine learning workflow. To achieve this, we have chosen to work with the chronic kidney disease dataset, a publicly available open-source dataset. Our intention is to provide a clear framework that can serve as a valuable resource for students, academics, and practitioners, allowing them to grasp the essence of a machine learning project's workflow. Our approach encompasses a meticulous and detailed procedure, aiming to demystify the complex steps involved in building machine learning applications. Within the realm of machine learning, there is an abundance of algorithms, each designed for specific applications. Selecting the most suitable algorithm for your specific purpose is a challenge. In our paper, we apply widely used algorithms that can be adapted to a broad range of use cases. This approach is designed to provide insight into the potential enhancements achievable through parameter tuning and fine-tuning, tailored to meet specific needs.

Saxena S., Darak S., Patel J., Joshi D.

Sign language recognition and detection systems play an increasingly major role in bridging the communication gap between the deaf community and the hearing world. This paper aims to provide an overview of research work on existing sign language recognition and detection systems by examining the different methodologies, techniques, and tools used by researchers to develop accurate and efficient solutions. Through this review, various insights will be gained that shed light on the strengths, limitations, and potential avenues for advancement. By deriving a variety of perspectives from researchers’ work, this paper further contributes to the discussion surrounding inclusive technologies that focus on the empowerment of individuals with hearing impairments while enabling seamless end-to-end communication.

Joshi V., Biradar M., Mathapathi B., Veershetty S.

The other sensor nodes in the cluster are sending their detected data to the CH, which is aggregating and sending it to the BS. A cluster head is considered to have greater capability and energy than other sensor nodes. The cluster head also reduces energy usage and provides scalability for large node counts. Yet, selecting the optimum CH is difficult since it has the greatest influence on the network's energy usage. In order to maximize secure and energy-aware routing in wireless sensor networks, this research presents a novel cluster-based routing model with optimal CHS. The best CH will be selected in the present investigation depending on the requirements as follows: distance, security (risk level evaluation), distance, delay, energy, and trust evaluation (direct and indirect trust). The routing model also considers how the route quality (reliability) of a cluster is calculated. This work proposes a new self-adaptive Beluga whale optimization method (SABWO) that uses it as the optimization issue. Finally, the findings of simulations confirm the effectiveness of the proposed strategy with respect to delay, throughput, residual energy, etc.

Aadhitya S., Premkumar J., Janney J.B.

Human eye is one of the compound sensory organs which is responsible for providing color vision. Any disturbance in eyes, or aging might lead to serious problems such as Glaucoma, Cataract, Hypertensive Retinopathy, macular degeneration, Pathological Myopia etc.… in the long run. Early and accurate diagnosis of these diseases is essential in effective treatment and prevention of blindness. Machine Learning (ML) has evolved as promising tools for automating the diagnosis of multiple eye diseases. This review work emphasizes on discussing various deep learning algorithms like computational neural network (CNN), K-means algorithm (KNN) and many other algorithms for creating an efficient model to predict the occurrence or non-occurrence of the diseases with high accuracy. The proposed system uses 2 different deep learning algorithms: CNN and KNN for construction of model and prediction. And their efficiencies are compared in accordance with the performance of the models.

Rajan S.J., Radhakrishnan S.

Embedded devices and Internet of Things (IoT) applications are becoming increasingly popular in today's world. The resource-constrained devices prefer to employ light-weight block ciphers for security. When light-weight cryptographic algorithms are used to provide security in these devices, they have a tendency to leak some information. Gain of this information helps the attacker to retrieve the key. One such attack that has gained increased attention is power analysis. Boolean masking is a countermeasure against power analysis attacks. The substitution box (S-Box) of lightweight ciphers PRESENT, LED, GIFT, KLEIN, and RECTANGLE has been considered here. We present a secured compact S- Box that consumes less power and area. The results of implementation on 45nm and 90nm technology application-specific integrated circuits (ASIC) validate our state. This design is suitable for low area applications and devices like smart cards, Radio frequency identification (RFID). Furthermore, Masking algorithms are used to obliterate information making it impossible to recover the original data. The suggested compact S-Box of these low weight ciphers was examined and compared to the protected and unprotected gate equivalents in terms of area, power consumption, and latency.1 to 5% reduction of area and 7% reduction of delay is achieved for protected compact S-Box when compared with un-protected non-compact S-Box.

Aslam M.A., Choudhary R., Ramanathan K., Nisha T.N.

A lot of user speech data is being accumulated as Automatic Speech Recognition (ASR) are integrated into devices to improve these systems. Privacy protection in speech data has drawn increased attention since the General Data Protection Regulation (GDPR) was implemented in the EU. As such voice data contains Personal Information (PI) which may jeopardize users’ right to privacy. These devices also capture voice passively even when the user is not interacting with the application. This poses a serious threat to the entity's sensitive information. This initiated the requirement of safety precautions for the use of voice data. The goal of this study is to discover methods for maintaining voice message/command security without changing the required content to maintain the data utility while safeguarding users’ privacy. We suggest a model, Advanced Automatic Speech Recognition (AASR) that segregates the confidential information of the user from the data that is required by the device/application. This is implemented in a phased manner where Support Vector Machine (SVM) is used in the first phase and adversarial noise in the second phase. The conceptual model emphasizes privacy by SVM removing most of the sensitive data and adversarial noise masking the remaining sensitive information. We also deduce any potential future propositions of the model.

Pradhan K., Patra S.

The success of spectral-spatial hyperspectral image classification techniques are dependent on their ability to consider appropriate spatial information. Structure preserving image filtering technique preserves structures of the objects while removing the noises from the image. To take into account better spectral-spatial information, in this research we have constructed a profile that consists of multiple filter images generated by applying a recently developed semantic-aware image filtering technique. Then, the pixels on the profile represented with spectral-spatial features are used to classify the hyperspectral images. The experiments conducted on three real hyperspectral datasets confirm potentiality of the proposed technique.

Singh R., Gupta A., Mittal P.

Blockchain technology is rapidly emerging with diverse applications across various domains. Notably, smart contracts represent a significant use case for blockchain technology. These self-executing code pieces, deployed on the blockchain, automatically execute when predetermined conditions are met. Smart contracts are gaining popularity due to their distinct advantages over traditional methods. They eliminate the need for intermediaries like banks, lawyers, or government agencies for execution. Furthermore, their deployment on the blockchain enhances security and accessibility. This article delves deeply into smart contracts and their emerging applications, serving as a valuable resource for scholars, researchers, and developers. It offers an extensive examination of smart contracts, covering their platforms, programming languages, and recent developments in fields such as the Internet of Things (IoT), Healthcare, Identity Management and Access Control, Real Estate, and Supply Chain. The article also addresses the challenges associated with these advancements.

Ingole S., Narsale S., Desale D., Rade S., Adhav S.

In complicated contexts like shopping malls, airports, hospitals, and museums Indoor navigation seems extremely difficult. In such circumstances, conventional navigation methods mostly struggle to provide accurate and understandable guidance. It examines the most current-edge AR-based indoor navigation systems and discusses their essential characteristics, advantages, and disadvantages. It emphasizes how well-suited this system is to various interior environments. The survey also focuses on problems and unanswered concerns in the field of augmented reality-based indoor navigation. Accurate positioning and tracking, occlusion handling, scalability, user interface design, and real-time performance are some of these difficulties. Aspects of usability, user feedback, and user experience evaluation of AR-based indoor navigation systems are also covered in the paper. The survey also investigates the combination of augmented reality with complementing technologies including sensor fusion, machine learning, and computer vision to create more effective indoor navigation systems.