Biography
Dr. Electa Alice Jayarani A. is an accomplished academician, researcher, and invited speaker with over 20 years of teaching and research experience in Electronics and Communication Engineering. She is currently serving as an Associate Professor at K.S. Institute of Technology, Bengaluru, and is a UGC-NET qualified faculty member with a Ph.D. from REVA University. Her research expertise spans Compressive Sensing, Audio Signal Processing, Digital Watermarking, Antenna Design, IoT, AI/ML applications, and Data Science, with 30+ publications in reputed Scopus, Springer, IEEE, and international journals and conferences. She is the winner of the BITES Best Ph.D. Thesis Award (2023), Venus International Outstanding Woman researcher in signal processing award 2026 and a recognized research supervisor under VTU. Dr. Jayarani holds 7 granted and published patents across India, the UK, and Australia, reflecting strong innovation and industry relevance. She has actively coordinated and served as session chair, resource person, and reviewer for several national and international conferences, and has led multiple funded and submitted research proposals with agencies such as VGST, SERB, DST, and KSCST.
Conference contributions 1
FEDERATED LEARNING IN EMBEDDED DEVICES:
Review of privacy-preserving approaches for behavioral biometrics to ensure data confidentiality
Federated Learning has attracted significant interest in recent years for its ability to keep data localized, thereby enhancing privacy and security while still enabling the development of robust, generalized models. Ensuring the confidentiality of sensitive health information, such as ECG recordings, is critically important. In this context, Federated Learning serves as a promising approach to safeguard privacy during data classification. This study explores this potential by comparing centralized learning models with Federated Learning models trained on both Independently and Identically Distributed (IID) data and Non-IID data. The result of this study would open path for conduct further experiments on various observable behaviors, where pedometric data obtained through federated learning can play a more significant role in gait analysis for financial fraud detection.