Machine Learning in Mobile Computing

The scope of this module is to impart a comprehensive perception of the integrated use and management of computing, telecommunications, storage and other resources in a mobile computing environment to the students. Emphasis is given on advanced algorithmic management methods, based on distributed processing and machine learning. Mobile Computing (communication-devices-software). Constraints (cost, mobility, security, power consumption, wireless medium). Advanced Technologies for Mobile Computing Infrastructure Management: Overview of Artificial Intelligence, Machine Learning, Neural Networks, Deep Learning, Federated Learning. Applications of Advanced Technologies in Large-Scale Distributed Architectures and Intelligent Mobile Terminals. Radio resource management optimization using supervised and
unsupervised learning techniques. Radio resource management in intelligent transport systems (including 5G-ITS Standards). Cognitive Radio Networks: spectrum sensing and management. Computing (features, architectures, software as a service, routing, security, data management), grid, cloud, fog computing. Resource Orchestration in 5G Systems: Virtual Network Functions (VNF), Resource Orchestration Algorithms. Allocation of computing and network resources to continuous cloud and/or edge infrastructures. Secure distributed storage in cloud and/or edge infrastructures.
Network tomography. Machine Learning at the Edge: Near-User and at the Networks’ Edge. Modern mobile terminals’ hardware: sensors, processing units (CPU, GPU, DSP, NPU), memory, battery. Operating system, power consumption and data storage issues. Laboratory exercises in the above subjects: Lab exercises using Python, related libraries (Keras / Tensor Flow) and Intelligent Mobile Applications (Android, iOS, TFLite).