The emergence of novel cellular network technologies, within 5G, are envisioned as key enablers of a new set of use-cases, including industrial automation, intelligent transportation, and tactile internet. The critical nature of the traffic requirements ranges from ultra-reliable communications, massive connectivity, and enhanced mobile broadband. Thus, the growing research on cellular network monitoring and prediction aims for ensuring a satisfied user-base and fulfillment of service level agreements. The scope of this study is to develop an approach for predicting the cellular link throughput of end-users, with a goal to benchmark the performance of network slices. First, we report and analyze a measurement study involving real-life cases, such as driving in urban, sub-urban, and rural areas, as well as tests in large crowded areas. Second, we develop machine learning models using lower-layer metrics, describing the radio environment, to predict the available throughput. The models are initially validated on the LTE network and then applied to a non-standalone 5G network. Finally, we suggest scaling the proposed model into the future standalone 5G network. We have achieved 93% and 84% R^2 accuracy, with 0.06 and 0.17 mean squared error, in predicting the end-user's throughput in LTE and non-standalone 5G network, respectively.

Quality of Experience (QoE) as an aggregate of Quality of Service (QoS) and human user-related metrics will be the key success factor for current and future mobile computing systems. QoE measurement and prediction are complex tasks as they may involve a large parameter space such as location, delay, jitter, packet loss, and user satisfaction just to name a few. These tasks necessitate the development of practical context-aware QoE models that efficiently determine relationships between user context and QoE parameters. In this paper, we propose, develop, and validate a novel decision-theoretic approach called CaQoEM for QoE modelling, measurement, and prediction. We address the challenge of QoE measurement and prediction where each QoE parameter can be measured on a different scale and may involve different units of measurement. CaQoEM is context-aware and uses Bayesian networks and utility theory to measure and predict users' QoE under uncertainty. We validate CaQoEM using extensive experimentation, user studies and simulations. The results soundly demonstrate that CaQoEM correctly measures range-defined QoE using a bipolar scale. For QoE prediction, an overall accuracy of 98.93% was achieved using 10-fold cross validation in multiple diverse network conditions such as vertical handoffs, wireless signal fading and wireless network congestion.

Developments such as the increasing electrical energy demand, growth of renewable energy sources, cyber–physical security threats, increased penetration of electric vehicles (EVs), and unpredictable behavior of prosumers and EV users pose a range of challenges to the electric power system. To address these challenges, a decentralized system using blockchain technology and machine learning techniques for secure communication, distributed energy management and decentralized energy trading between prosumers is required. Blockchain enables secure distributed trust platforms, addresses optimization and reliability challenges, and allows P2P distributed energy exchange as well as flexibility services between customers. On the other hand, machine learning techniques enable intelligent smart grid operations by using prediction models and big data analysis. Motivated from these facts, in this review, we examine the potential of combining blockchain technology and machine learning techniques in the development of smart grid and investigate the benefits achieved by using both techniques for the future smart grid scenario. Further, we discuss research challenges and future research directions of applying blockchain and machine learning techniques for smart grids both individually as well as combining them together. The identified areas that require significant research are demand management in power grids, improving the security of grids with better consensus mechanisms, electric vehicle charging systems, scheduling of the entire grid system, designing secure microgrids, and the interconnection of different blockchain networks.