Artificial Intelligence (AI) integration in network management has significantly improved Quality of Experience (QoE) and achieved high availability, reaching 99.99% uptime with an aim to meet the five 9s standard in Next Generation Networks. Utilizing Closed Loop Automation (CLA) within the management plane, AI-driven systems dynamically allocate network resources based on real-time user data. However, the shift to automated management introduces privacy concerns, as retained user data may violate the GDPR's “right to be forgotten” requirement. To address this, we propose a machine unlearning framework for precise data removal in Deep Neural Network (DNN) models that is fully integrated into the CLA. This framework ensures compliant data deletion requests while preserving model performance. Our experiments demonstrate the framework's effectiveness by preserving the model's accuracy on validation data by 79.38 % compared to the existing benchmark methods of 57.0048 %, underscoring its potential in secure, privacy-aware network management. The update is tested on model forecasting Internet usage from a telecommunications dataset and achieves a low complexity while maintaining the model's accuracy over a validation dataset. The method provides a reasonable time to run for the 7 million parameter model. Its ease of integration within the network management plane allows operators to reach compliance with the Data Acts.

Existing network paradigms have achieved lower downtime as well as a higher Quality of Experience (QoE) through the use of Artificial Intelligence (AI)-based network management tools. These AI management systems, allow for automatic responses to changes in network conditions, lowering operation costs for operators, and improving overall performance. While adopting AI-based management tools enhance the overall network performance, it also introduce challenges such as removing human supervision, privacy violations, algorithmic bias, and model inaccuracies. Furthermore, AI-based agents that fail to address these challenges should be culpable themselves rather than the network as a whole. To address this accountability gap, a framework consisting of a Deep Reinforcement Learning (DRL) model and a Machine Learning (ML) model is proposed to identify and assign numerical values of responsibility to the AI-based management agents involved in any decision-making regarding the network conditions, which eventually affects the end-user. A simulation environment was created for the framework to be trained using simulated network operation parameters. The DRL model had a 96% accuracy during testing for identifying the AI-based management agents, while the ML model using gradient descent learned the network conditions at an 83% accuracy during testing.

Autonomous Vehicles (AVs), furnished with sensors capable of capturing essential vehicle dynamics such as speed, acceleration, and precise location, possess the capacity to execute intelligent maneuvers, including lane changes, in anticipation of approaching roadblocks. Nevertheless, the sheer volume of sensory data and the processing necessary to derive informed decisions can often overwhelm the vehicles, rendering them unable to handle the task independently. Consequently, a common approach in traffic scenarios involves transmitting the data to servers for processing, a practice that introduces challenges, particularly in situations demanding real-time processing. In response to this challenge, we present a novel DL-based semantic traffic control system that entrusts semantic encoding responsibilities to the vehicles themselves. This system processes driving decisions obtained from a Reinforcement Learning (RL) agent, streamlining the decision-making process. Specifically, our framework envisions scenarios where abrupt roadblocks materialize due to factors such as road maintenance, accidents, or vehicle repairs, necessitating vehicles to make determinations concerning lane-keeping or lane-changing actions to navigate past these obstacles. To formulate this scenario mathematically, we employ a Markov Decision Process (MDP) and harness the Deep Q Learning (DQN) algorithm to unearth viable solutions.

A sudden roadblock on highways due to many reasons such as road maintenance, accidents, and car repair is a common situation we encounter almost daily. Autonomous Vehicles (AVs) equipped with sensors that can acquire vehicle dynamics such as speed, acceleration, and location can make intelligent decisions to change lanes before reaching a roadblock. A number of literature studies have examined car-following models and lane-changing models. However, only a few studies proposed an integrated car-following and lane-changing model, which has the potential to model practical driving maneuvers. Hence, in this paper, we present an integrated car-following and lane-changing decision-control system based on Deep Reinforcement Learning (DRL) to address this issue. Specifically, we consider a scenario where sudden construction work will be carried out along a highway. We model the scenario as a Markov Decision Process (MDP) and employ the well-known DQN algorithm to train the RL agent to make the appropriate decision accordingly (i.e., either stay in the same lane or change lanes). To overcome the delay and computational requirement of DRL algorithms, we adopt an MEC-assisted architecture where the RL agents are trained on MEC servers. We utilize the highly reputable SUMO simulator and OPENAI GYM to evaluate the performance of the proposed model under two policies; E-greedy policy and Boltzmann policy. The results unequivocally demonstrate that the DQN agent trained using the e-greedy policy significantly outperforms the one trained with the Boltzmann policy.

The increasing demand for cellular networks has prompted vendors and service providers to concentrate on establishing a more autonomous network infrastructure. Among numerous emerging paradigms, network service federation has gained prominence. This model facilitates collaboration between Virtual Network Operators (VNO) and Infrastructure Providers in a Multi-Domain Network, enabling resource sharing, enhanced quality of service, and expanded coverage areas. Nevertheless, the varying service level agreements (SLA) among network operators can influence several aspects for customers attempting to offload a task to a network, particularly in roaming scenarios. Factors such as cost and service performance become crucial considerations, potentially creating complexities for end users. Herein, to automate this process, a Deep Reinforcement Learning (DRL) model was devised for efficient MEC selection. The DRL model introduced on the Multi-Access Edge Computing (MEC) finds the optimal policy for the user offloading the task. Through simulation, the DRL model was able to reduce cost by 38.84% and had an accuracy of 97.6% during training.