Swathi Rangaswamy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Swathi Rangaswamy has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Sivakumar Venu
Vellore Institute of Technology, India
4 shared publications - Atul Anand
Indian Institute of Technology, Patna, India
1 shared publication - K. Venkateswara Rao
School of Computer Science and Engineering (SCOPE), Vellore Institute of Technology (VIT), Vellore, India
1 shared publication - M. Kanishka
Vellore Institute of Technology, India
1 shared publication - Melissa Maria Augustine
Vellore Institute of Technology, India
1 shared publication - Mohammed Mubasheer
Vellore Institute of Technology, India
1 shared publication
Latest Publications
- 1 September 2025
- Published by Bentham Science Publishers Ltd. in International Journal of Sensors, Wireless Communications and Control
- Vol. 15 (3) , 256-267
A Deep Dive into Detecting and Investigating Fileless Malware
Fileless malware is a very advanced threat that has garnered significant attention due to its highly stealthy and secretive nature, as well as its ability to easily evade traditional security measures. Unlike traditional malware, which leaves footprints on disks, fileless malware operates in the shadows of system memory, thereby surpassing detection and analysis. In this paper, we provide a comprehensive review of fileless malware, including its evolution, detection techniques, and mitigation strategies. We also explore the historical context of fileless malware. By examining various methodologies employed by researchers and practitioners worldwide, this analysis aims to shed light on strategies for combating the evolving threat posed by fileless malware. We discuss current research efforts and emerging trends in fighting fileless malware, emphasizing the importance of proactive defense strategies in mitigating this evolving threat landscape. Our analysis delves into a comparative study of traditional malware and fileless malware, specifically focusing on Kovter. Leveraging advanced tools like Any.run and VirusTotal, we examine the unique challenges that traditional antivirus solutions encounter when attempting to detect fileless malware. This study underscores the limitations of conventional detection methods in addressing the stealthy nature of these advanced threats.
- 1 September 2025
- Published by Bentham Science Publishers Ltd. in International Journal of Sensors, Wireless Communications and Control
- Vol. 15 (3) , 256-267
A Deep Dive into Detecting and Investigating Fileless Malware
Fileless malware is a very advanced threat that has garnered significant attention due to its highly stealthy and secretive nature, as well as its ability to easily evade traditional security measures. Unlike traditional malware, which leaves footprints on disks, fileless malware operates in the shadows of system memory, thereby surpassing detection and analysis. In this paper, we provide a comprehensive review of fileless malware, including its evolution, detection techniques, and mitigation strategies. We also explore the historical context of fileless malware. By examining various methodologies employed by researchers and practitioners worldwide, this analysis aims to shed light on strategies for combating the evolving threat posed by fileless malware. We discuss current research efforts and emerging trends in fighting fileless malware, emphasizing the importance of proactive defense strategies in mitigating this evolving threat landscape. Our analysis delves into a comparative study of traditional malware and fileless malware, specifically focusing on Kovter. Leveraging advanced tools like Any.run and VirusTotal, we examine the unique challenges that traditional antivirus solutions encounter when attempting to detect fileless malware. This study underscores the limitations of conventional detection methods in addressing the stealthy nature of these advanced threats.
- 22 August 2025
- Published by IGI Global Scientific Publishing
- p. 127-156
A Smart Farmland for Crop Protection Using Sensors and IoT
Effective farm management faces challenges, such as the need for continuous monitoring and timely intervention to protect crops from improper irrigation and fire risks. Current solutions often rely on manual monitoring and traditional irrigation methods, which are inefficient and prone to human error. Additionally, fire prevention is typically reactive, leading to significant crop loss. To address these issues, combines IoT-based soil moisture sensors and advanced fire detection sensors. Soil moisture sensors provide real-time data to optimize irrigation schedules, avoiding both over-watering and under-watering. Simultaneously, fire sensors detect early signs of fire, allowing for immediate intervention to mitigate potential damage. This dual-layered system enhances irrigation efficiency and fire protection, promoting sustainable farming, crop health and safety. By automating key aspects of farm management, the system reduces the need for manual labour, conserves resources, and supports resilience against adverse conditions, making it an effective solution for modern agriculture.
- 22 August 2025
- Published by IGI Global Scientific Publishing
- p. 127-156
A Smart Farmland for Crop Protection Using Sensors and IoT
Effective farm management faces challenges, such as the need for continuous monitoring and timely intervention to protect crops from improper irrigation and fire risks. Current solutions often rely on manual monitoring and traditional irrigation methods, which are inefficient and prone to human error. Additionally, fire prevention is typically reactive, leading to significant crop loss. To address these issues, combines IoT-based soil moisture sensors and advanced fire detection sensors. Soil moisture sensors provide real-time data to optimize irrigation schedules, avoiding both over-watering and under-watering. Simultaneously, fire sensors detect early signs of fire, allowing for immediate intervention to mitigate potential damage. This dual-layered system enhances irrigation efficiency and fire protection, promoting sustainable farming, crop health and safety. By automating key aspects of farm management, the system reduces the need for manual labour, conserves resources, and supports resilience against adverse conditions, making it an effective solution for modern agriculture.
- 21 March 2025
- Published by IGI Global Scientific Publishing
- p. 227-242
Precision Fitness Instruction System Using Vector Database
Nowadays regular exercises are needed for preventing and maintaining the health from diseases. Analyze and give the best exercises based on the input to the user which then presents them the concatenated video. This paper integrates an LLM, in this case GPT-3.5 for NLP, which enables to interpret user inputs for the Weaviate vector database. The results were stitched video which is a combination of videos taken from the S3 bucket. The stitched video didn't result in any major bugs or quality issues. It is relevant to the user inputs such as preference, weaknesses and more. This application is focused to helps people doesn't know how and where to begin get a basic idea of exercises they can achieve their goals regarding their fitness. The videos selected and stitched together to make it easier for person viewing it to understand how the exercises work as well as do them correctly. The final output video is personalized to the user's preferences and skill level and so it will meet their objectives making the workout experience, better fitness, enjoyable and effective.
- 21 March 2025
- Published by IGI Global Scientific Publishing
- p. 227-242
Precision Fitness Instruction System Using Vector Database
Nowadays regular exercises are needed for preventing and maintaining the health from diseases. Analyze and give the best exercises based on the input to the user which then presents them the concatenated video. This paper integrates an LLM, in this case GPT-3.5 for NLP, which enables to interpret user inputs for the Weaviate vector database. The results were stitched video which is a combination of videos taken from the S3 bucket. The stitched video didn't result in any major bugs or quality issues. It is relevant to the user inputs such as preference, weaknesses and more. This application is focused to helps people doesn't know how and where to begin get a basic idea of exercises they can achieve their goals regarding their fitness. The videos selected and stitched together to make it easier for person viewing it to understand how the exercises work as well as do them correctly. The final output video is personalized to the user's preferences and skill level and so it will meet their objectives making the workout experience, better fitness, enjoyable and effective.
- 14 February 2025
- Published by IGI Global Scientific Publishing
- p. 469-502
Generative AI and Secure IoMT Systems for Enhanced Real-Time Health Monitoring and Security
AI is going to change the face of healthcare through the bridge with the Internet of Medical Things (IoMT) by diagnosing problems, guaranteeing information security, and providing personalized care. This framework applies hybrid cloud-edge architecture for the realization of real-time diagnostics while reducing latency and leveraging AI, generative AI, and IoMT. Deep learning models enable better cancer detection and mainstream advancements in imaging, drug discovery, and treatment options enabled by generative AI. Federated learning guarantees privacy of data as data is retained at the edge while ensuring efficiency and has improved security, covered with a chaotic map-based encryption algorithm for safe data even in resource-constrained IoMT devices. The system offers real-time patient monitoring by adopting personal adaptable treatments and is scalable in healthcare environments, providing high diagnostic precision, enhanced security, and responsive medical care, making it a breakthrough in digital health in pursuit of a more tailored patient solution.
- 14 February 2025
- Published by IGI Global Scientific Publishing
- p. 469-502
Generative AI and Secure IoMT Systems for Enhanced Real-Time Health Monitoring and Security
AI is going to change the face of healthcare through the bridge with the Internet of Medical Things (IoMT) by diagnosing problems, guaranteeing information security, and providing personalized care. This framework applies hybrid cloud-edge architecture for the realization of real-time diagnostics while reducing latency and leveraging AI, generative AI, and IoMT. Deep learning models enable better cancer detection and mainstream advancements in imaging, drug discovery, and treatment options enabled by generative AI. Federated learning guarantees privacy of data as data is retained at the edge while ensuring efficiency and has improved security, covered with a chaotic map-based encryption algorithm for safe data even in resource-constrained IoMT devices. The system offers real-time patient monitoring by adopting personal adaptable treatments and is scalable in healthcare environments, providing high diagnostic precision, enhanced security, and responsive medical care, making it a breakthrough in digital health in pursuit of a more tailored patient solution.