The field of edge computing and traffic classification is witnessing significant advancements, driven by the need for efficient and scalable solutions for real-time data analysis. Researchers are exploring innovative approaches to optimize deep neural networks for resource-constrained devices, enabling the classification of encrypted traffic and improving cybersecurity for IoT networks. Meanwhile, the integration of edge computing with 5G connectivity is opening up new possibilities for applications such as smart prosthetic hands and drone swarms. Noteworthy papers in this area include Efficient Traffic Classification using HW-NAS, which presents a hardware-efficient deep neural network optimized through hardware-aware neural architecture search, and TraGe, a novel generic packet representation model for traffic classification. Additionally, the development of low-latency surveillance systems using entropy-based adaptive buffering and MobileNetV2 on edge devices is showcasing the potential for real-time video surveillance in resource-constrained environments.