The field of edge computing is moving towards developing more efficient and scalable architectures to support real-time applications. Researchers are exploring innovative approaches to reduce computational complexity, improve model accuracy, and enhance privacy. One notable direction is the integration of compressed sensing theory into split computing frameworks, enabling efficient data transmission and processing. Another area of focus is the design of transceivers for ambient internet of things (A-IoT) devices, aiming to achieve low-power consumption and low-cost deployment. Additionally, novel computing architectures are being proposed to overcome energy constraints in deep learning inference on edge devices. Noteworthy papers include:

• A novel cloud segmentation model that achieves state-of-the-art performance while reducing computational complexity.
• A proposed receiver architecture for A-IoT devices that eliminates external crystals and achieves high sensitivity.
• A split computing architecture inspired by compressed sensing theory that reduces bandwidth utilization and improves computational efficiency.
• A novel computing architecture for wireless edge networks that achieves energy-efficient deep learning inference via in-physics computation at radio frequency.