Description of Research Interests:
Description of Research Interests:
General Area: Signal Processing and Machine Learning.
Hyperparameter free machine learning using Random Fourier features.
Analytical results for hyperparameter free RFF based deep learning.
Applications: Localization, Tactical Communications, massive MIMO, Signal Processing for Optical Wireless Communications.
Hyperparameter free criterion learning.
Significant research outputs:
•R. Mitra, G. Kaddoum, G. Dahman and G. Poitau, “Hyperparameter Free MEE-FP Based Localization," in IEEE Signal Processing Letters, vol. 28, pp. 1938-1942, 2021, doi:10.1109/LSP.2021.3111596.
•R. Mitra, G. Kaddoum and G. Poitau, “Analytical Guarantees for Hyperparameter Free RFF based Deep Learning in the Low-DataRegime," in IEEE Transactions on Circuits and Systems II: Express Briefs, doi: 10.1109/TCSII.2021.3096805.
•R. Mitra, V. Bhatia, S. Jain and K. Choi, “Performance Analysis of Random Fourier Features-Based Unsupervised Multistage-Clusteringfor VLC," in IEEE Communications Letters, vol. 25, no. 8, pp. 2659-2663, Aug. 2021, doi: 10.1109/LCOMM.2021.3089933.
•R. Mitra, G. Kaddoum and V. Bhatia, “Hyperparameter-Free Transmit-Nonlinearity Mitigation Using a Kernel-Width Sampling Technique," in IEEE Transactions on Communications, vol. 69, no. 4, pp. 2613-2627, April 2021, doi: 10.1109/TCOMM.2020.3048045.
•R. Mitra, G. Kaddoum, D. B. da Costa, “Hyperparameter Free MEEF based Learning for Next Generation Communication Systems",IEEE Transactions on Communications, vol. 70, no. 3, pp. 1682-1696, Jan. 2022.
•R. Mitra, S. Jain, G. Kaddoum and K. Choi, "Recursive Hyperparameter-Free Criterion Learning," in IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 69, no. 11, pp. 4618-4621, Nov. 2022.