Year-2024 http://repository.iiitd.edu.in/xmlui/handle/123456789/1724 2026-09-20T13:40:10Z 2026-09-20T13:40:10Z Megha Deb, Sujay (Advisor) http://repository.iiitd.edu.in/xmlui/handle/123456789/2142 2026-09-15T22:00:29Z 2024-11-27T00:00:00Z

Development of a continuous heart rate variability (HRV) monitoring system Megha; Deb, Sujay (Advisor) Heart Rate Variability is a widely studied physiological marker that reflects the autonomic regulation of the heart. HRV represents the autonomic nervous system’s control over the heart and is crucial for cardiovascular health monitoring. Heart Rate Variability is measured using methodologies such as electrocardiography (ECG) for RR interval analysis and spectral decomposition, and photoplethysmography (PPG) for blood volume changes. Some of the existing methods use mobile health (mHealth) apps with smartphone sensors, and time frequency analysis for comprehensive HRV dynamics assessment across various time scales. This paper presents VariScan, a device designed to provide accurate and continuous(24x7) HRV monitoring using the MAX30102 sensor with an IoT controller for data transmission. The suggested module in this paper is a low-cost non-invasive device, which can be customized and easily calibrated based on the user applications. VariScan was tested on 10 subjects to predict the HR at five different time instants, and the marked reading was compared with an available heart rate meter. The device HR readings were within 0.5 per cent of the accuracy.

2024-11-27T00:00:00Z Narnoli, Bhavya Shukla, Jainendra (Advisor) http://repository.iiitd.edu.in/xmlui/handle/123456789/2141 2026-09-15T22:00:39Z 2024-12-09T00:00:00Z

Knowledge graph creation using LLM Narnoli, Bhavya; Shukla, Jainendra (Advisor) Knowledge Graphs are essential for structuring data into machine-interpretable formats, enabling ac- curate information retrieval, improved decision-making, and understanding of complex relationships. While KGs have seen significant progress in many languages, Hindi-specific KGs remain underdeveloped, limiting technological advancements and knowledge accessibility in the language. Large language models, with their extensive training on diverse datasets and deep contextual understanding, offer the potential for constructing Hindi KGs, surpassing traditional models like encoders in identifying and mapping complex relationships effectively. This study evaluates two LLMs — Bard, and GPT for generating RDF triples from Hindi NCERT Bi- ology texts. Other hindi llm’s like Llama Results reveal significant variation in accuracy, with Bard achieving a maximum hits@1 accuracy of 32.2%, and Llama performing poorly at 10.5%. After improvising the prompt, for GPT 4o model, it gave an average of 64.5% accuracy for three different Hindi chapters of Biology. The limitations of improvising the performance demonstrated better contextual understanding but faced challenges in extracting relationships from complex, nested sentence structures. Prompt engineering improved GPT’s performance to some extent, yet limitations in handling intricate information and the scarcity of annotated data in Hindi remain barriers to generating reliable KGs.

2024-12-09T00:00:00Z Dwivedi, Shivam Grover, Anuj (Advisor) http://repository.iiitd.edu.in/xmlui/handle/123456789/2071 2026-09-02T22:00:18Z 2024-11-22T00:00:00Z

The Dot Store - a comprehensive study, UX research and in-depth implementation Dwivedi, Shivam; Grover, Anuj (Advisor) The Dot Store, started in May 2023 as an independent project under the guidance of Dr. Anuj Grover, began as an in-depth business research endeavor involving numerous surveys and interviews to identify needs within the creator economy. This research highlighted significant opportunities, leading to the official launch of The Dot Store as a startup in September 2023. The platform was designed to empower creators by turning their product ideas into reality, han- dling all aspects from design to sales, thus addressing the challenge creators face in monetizing their content. The first creator was onboarded in October 2023, with the first sale marking the beginning of a profitable venture, achieving a max monthly revenue of Rs. 120,000/- and a max net profit of Rs. 34,000/- in August 2024. Throughout this B.Tech project, the focus was on solidifying the business fundamentals, establishing standard operating procedures, and expanding the startup further. The Dot Store operated from 532 in the IIITD Innovation and Incubation Centre at IIITD with a core team of three and an extended network of 12 volunteers till July 2024, afterwards it shifted its operations to working partially out of student startup working space in E-Cell IIITD and partially remote. Furthermore, during this BTP, the venture’s innovative approach and business potential were recognized during its progression through three stages of the television show Shark Tank India, although it did not reach the final round. This report details the development of The Dot Store, reflecting on both the challenges faced and the achievements made in creating a sustainable and empowering platform for creators of all sizes. The project exemplifies the practical application of academic and entrepreneurial principles to create a viable business solution in the modern digital landscape.

2024-11-22T00:00:00Z Aadya Darak, Sumit Jagdish (Advisor) http://repository.iiitd.edu.in/xmlui/handle/123456789/2049 2026-08-29T22:00:48Z 2024-12-03T00:00:00Z

Pipeline for end-to-end training of DL-augmented LS channel estimation Aadya; Darak, Sumit Jagdish (Advisor) This work presents an enhanced training strategy for LSDNN (Least Squares Deep Neural Net- work) [1], a lightweight deep neural network designed to improve LS channel estimation in pilot-based OFDM systems. The approach offers two key advantages: (1) it enables LSDNN to be trained using practical data sources, such as transmitted codewords and demapper LLRs, and (2) it achieves superior Bit Error Rate (BER) performance through this end-to-end training. By removing the need for perfect CSI during training, the proposed method simplifies implementation in real-world scenarios and aligns with the goal of deploying the system on hardware such as Software Defined Radios (SDRs). Compared to our baseline model, LSDNN strikes a balance between performance and efficiency, with significantly lower computational complexity, fewer parameters, and reduced memory requirements, while delivering competitive results. The model’s ability to generalize to unseen channel models further underscores its potential for real-world applications. This work demonstrates how deep learning can complement traditional signal processing to create efficient, adapt- able, and high-performance wireless communication systems.

2024-12-03T00:00:00Z