Akhila S.

Work place: Department of Electronics and Communication Engineering, B.M.S. College of Engineering, VTU, Bengaluru, Karnataka 560019, India

E-mail: akhilas.ece@bmsce.ac.in


Research Interests:


Dr. S. Akhila is a Professor in the Department of ECE with over 28 years of teaching experience and 14 years of research experience. She completed her Ph. D in Wireless Communication from VTU in the year 2013 and her graduate degree in Electronics in 1995 from UVCE, Bangalore. She has guided several PG and UG level dissertations, has 5 research scholars under her supervision, of which one has been awarded doctorate, has over 48 refereed journal and conference publications and has applied for two patents. E-mail: akhilas.ece@bmsce.ac.in

Author Articles
D2D Communication Using Distributive Deep Learning with Coot Bird Optimization Algorithm

By Nethravathi H. M. Akhila S. Vinayakumar Ravi

DOI: https://doi.org/10.5815/ijcnis.2023.05.01, Pub. Date: 8 Oct. 2023

D2D (Device-to-device) communication has a major role in communication technology with resource and power allocation being a major attribute of the network. The existing method for D2D communication has several problems like slow convergence, low accuracy, etc. To overcome these, a D2D communication using distributed deep learning with a coot bird optimization algorithm has been proposed. In this work, D2D communication is combined with the Coot Bird Optimization algorithm to enhance the performance of distributed deep learning. Reducing the interference of eNB with the use of deep learning can achieve near-optimal throughput. Distributed deep learning trains the devices as a group and it works independently to reduce the training time of the devices. This model confirms the independent resource allocation with optimized power value and the least Bit Error Rate for D2D communication while sustaining the quality of services. The model is finally trained and tested successfully and is found to work for power allocation with an accuracy of 99.34%, giving the best fitness of 80%, the worst fitness value of 46%, mean value of 6.76 and 0.55 STD value showing better performance compared to the existing works.

[...] Read more.
Other Articles