Wireless Communications And Quantum-inspired Nanotechnology With Applications Of Ml/ai
Lecturer II
Electronics and Computer Engineering
At the Electronics And Computer Engineering department office
Appointment on Visitation important
Topic: Photonics, Plasmonics And Materials Engineering Nanoscience ;Radio Wave Propagation Quantum Electronics ;Microwave Engineering And Antennas Electromagnetic Waves And Fields Wireless Communications ; Numerical Analysis And Application Digital Signal Processing ;Privacy, Security And Authentication
Description: Using ML/AI/DL/RL applications to solve RRAM in Wireless Communications and Quantum-Inspired Nano-based Metallic Materials using Numerical Techniques such as SFDM, FVM, FEM, CFDM and BEM
# | Certificate | School | Year |
---|---|---|---|
1. | Ph.D (Wireless Communications with Quantum-Based Modelling and Simulations of Plasmonic and Photonic small-scaled devices) | Electrical Engineering, University of Cape Town, South Africa | 2021 |
Enhancing Scalable and Secure Traffic Engineering in Hybrid Switching Software Defined Networking using Deep Deterministic Policy Gradient
Aim:
This study aims to improve the scalability and security of traffic engineering in Hybrid Switching Software Defined Networking (SDN) environments using Deep Deterministic Policy Gradient (DDPG), an advanced reinforcement learning algorithm.
Objectives:
To develop a DDPG-based traffic engineering model capable of handling dynamic and large-scale network environments.
To enhance decision-making for path selection and bandwidth allocation in hybrid switching architectures.
To integrate security constraints into the learning process to prevent traffic hijacking, congestion, and denial-of-service vulnerabilities.
Methodology:
The research will simulate a hybrid SDN architecture combining circuit and packet switching paradigms. A custom traffic environment will be created in simulation platforms such as Mininet and Open AI Gym. The DDPG agent will be trained to learn optimal routing strategies by interacting with network conditions. Key performance metrics—such as throughput, packet loss, latency, and security violations—will be tracked and compared against traditional algorithms like OSPF and Dijkstra-based TE models.
Expected Results:
The DDPG-based model is expected to significantly outperform baseline techniques in adapting to fluctuating traffic loads while maintaining low latency and high throughput. It is also anticipated to respond effectively to simulated security threats, reducing risk exposure in dynamic routing decisions.
Contribution to Knowledge:
This research contributes a novel application of deep reinforcement learning to hybrid SDN traffic engineering, combining real-time adaptability with security-aware policy learning. It bridges AI with modern network architecture, setting the groundwork for more autonomous, intelligent, and resilient SDN infrastructures.
AKINYEMI LATEEF is a Lecturer II at the Department of Electronics and Computer Engineering
AKINYEMI has a Ph.D in Wireless Communications with Quantum-Based Modelling and Simulations of Plasmonic and Photonic small-scaled devices from Electrical Engineering, University of Cape Town, South Africa