Modern mobile devices and sensors demand low latency, mobility support, energy efficiency, and high bandwidth from back-end cloud-based data centers and content delivery networks (CDN). The long geographical distance between mobile users and cloud data centers/CDNs leads to high round trip times for end-users, sensors, and applications. Therefore, the computing facilities are moving from centralized data centers/CDNs to edge computing with the emergence of 5G/6G networks. The technologies providing storage and computing services in end-user and device proximity are collectively known as edge network/computing. As a result of the migration of computing resources towards the edge, the end-user Quality of Experience (QoE) has increased. Edge computing is being supported by the federated, collaborative, and distributed intelligence hosted over end-user devices, sensors, and network equipment. The emergence of Edge Intelligence (EI) because of the amalgam of edge computing and Artificial Intelligence (AI) demands novel research proposals to facilitate the users and IoTs in achieving the required QoE.

Many research domains need investigation in EI, such as smart homes/cities, intelligent traffic and surveillance systems, voice and speaker recognition, content caching on the edge, edge and geo-specific recommendation systems, and end-user privacy. Federated learning techniques lower the burden of Machine Learning (ML) inference by distributing the learning process over multiple resource-constrained edge devices. The large amount of data in the edge network is generated by the Internet of Things (IoT) and end-users. The data can be utilized in intelligent resource management and intelligent applications for intelligent traffic management systems, the Internet of Medical Things, and Industrial IoT. However, data processing requires distributed validation and inference models for intelligent decision-making. AI and ML algorithms have solved many end-user problems in edge computing ranging from privacy, resource management, service placement, automation, etc. The exponential increase in the computational power of edge and end-user devices has made the application of compute-intensive AI/ML algorithms feasible. The challenges of designing federated and distributed intelligence models in edge and IoT networks that are cost and resource-efficient while ensuring data privacy and security are yet to be addressed.

This Special Issue focuses on recent advances and innovations in AI and federated learning for IoT and edge networks. We welcome original research and review articles that address the field of edge and IoT networks while procuring AI, machine, and federated learning techniques.

Potential topics include but are not limited to the following:

• Edge intelligence for IoT ecosystems to enable QoE and intelligent applications
• Machine learning-based edge caching
• Edge community-focused recommendation systems
• Blockchain applications for the user and data privacy and security in the edge
• Innovative internet of medical things for health informatics
• IoT integration with edge computing
• Intelligent traffic signaling, monitoring, and surveillance systems on the edge
• Federated intelligence for edge
• Voice/speaker recognition for smart homes on the edge
• Edge intelligence-empowered intelligent sensors and data analytics
• Edge intelligence for industrial automation
• Simulation tools and models for edge intelligence performance evaluation
• Novel collaboration of edge computing with AI/ML