The demand for computational resources is increasing exponentially due to an increasing amount of digital services. Cloud computing is becoming the standard for enterprises to provide these resources. This resulted in hyperscalers which consist of a large number of servers. Data centers consume more than 1% of the world’s electrical energy. Therefore, many techniques are developed that both help to fulfil the needs of digital services and reduce the energy consumption of data center services. Modern-day servers can switch between different operating states which are often integrated in a power configuration mode of servers known as eco-mode. One of these techniques throttles the clock frequency of a central processing unit (CPU) which enables the possibility to lower the power needed for that CPU. This technique is known as dynamic frequency and voltage scaling (DFVS) and the states it switches between are known as performance states (P-states). A different technique that is integrated with eco-mode is the ability to switch between different idle states of the CPU. These states define whether certain caches of the CPU are flushed or not to conserve energy. These states are known as core states (C-states). A different approach that is focused on conserving energy is virtualisation within data centers. Virtualisation enables one physical server to host multiple virtual instances of servers (virtual machines). This reduces resource wastage and energy consumption of a data center. However, this creates the need for a strategic placement that ensures that the demands of the virtual machines are met and that minimises energy consumption and resource wastage. This thesis analyses four of these techniques: the best fit decreasing (BFD) algorithm, the integer linear programming (ILP) algorithm, the particle swarm optimisation (PSO) algorithm and the genetic algorithm (GA). This thesis provides a framework that uses a holistic approach to provide insights into the effects of using eco-mode of servers within the dynamics of virtual machine placement in data centers. This framework serves as a first step in parameterising the dynamics of a data center regarding its energy consumption and performance. The results show a potential energy reduction of up to approximately 20% with negligible impact on a data center’s performance. This result occurs when applying the best fit decreasing algorithm and having a server with an energy-efficient eco-mode. However, this thesis does not cover all parameters that play a role in the data center’s performance and energy consumption, so more research on this area is recommended.

Accurate capacity planning is essential to ensure uninterrupted services and network stability through peak hours for the transport core network of KPN. This involves a trade-off between minimizing the risks of capacity shortages and costs of capacity expansions. High network loads are occurring more frequently and their magnitude is increasing. This necessitates measures to foresee high load situations before network capacity is surpassed. Currently, planning is based on manual predictions that lack substantiation. This research aims to improve network capacity planning by development of a forecast for the next year.
An analysis of the daily maximum traffic data of the transport core is performed, to determine the most suitable models for the prediction of network traffic. The data analysis, employing time series decomposition, revealed non-stationary trends and annual seasonality; traffic decreases throughout the summer and increases in the winter. An upward trend in the frequency and intensity of traffic peaks, highlights the growing demand and shifts in usage behavior. The extreme traffic peaks in the historical data were correlated to F1 race days and other anticipated events.
Two algorithms that integrate exogenous variables were assessed to predict the extreme values. The models either yielded inaccurate traffic predictions or encountered challenges in interpretability and pattern recognition, with the limited amount of data available. In response to these limitations, a decomposed forecast was created that predicts the trend and seasonality. Furthermore, Extreme Value Analysis (EVA) was implemented to address the extreme values in the data.
The final prediction framework combines the decomposed forecast with EVA for the next six quarters and outperforms the other models. The model effectively captures extreme values and provides insights into the maximum expected peaks and risk levels. The substantiated forecasts of the EVA model and the manual predictions yielded comparable results. However, the EVA model provides better insights into the likelihood of exceeding specific traffic values, which enhances capacity calculations and precision.
The prediction framework has been integrated into the business interface of KPN, which marks the initial step in the automatization of short-term capacity planning. The research insights emphasize the intricate nature of accurate prediction of future demand and advocate for scalable solutions beyond building new capacity. These solutions range from short-term mitigation to long-term strategies designed to alleviate high network loads. They underscore the importance of the implementation and integration of
dynamic decision-making within a digital twin of the network to ensure sustained effectiveness.

Network capability exposure (NCE) in 5G allows service providers to make network functionalities—such as data, connectivity services, and traffic management—accessible to developers and enterprises through APIs. This is essential for creating programmable networks that support diverse and complex 5G use cases, including gaming, drones, smart manufacturing, and autonomous vehicles. By leveraging these APIs, developers can access advanced 5G capabilities to design innovative applications, while service providers and enterprises unlock new revenue streams. For instance, APIs can enable mobile devices to dynamically activate high-speed connectivity tiers for specific applications, showcasing the flexibility and potential of NCE in 5G.

This thesis is divided into two parts. The first part investigates NCE in 5G mobile networks, focusing on the architecture, functionalities, and applications of the Network Exposure Function (NEF). It examines the NEF's role in securely exposing network services, its integration within the 5G ecosystem, and its implementation. Furthermore, the thesis evaluates capability exposure across industry standards, including 3GPP, O-RAN, the Operator Platform, and the CAMARA Project. The second part explores two use cases: (1) augmented reality (AR)-enhanced communications, where additional network capability exposure can enrich voice calling with AR features, and (2) drone operations, emphasizing collision avoidance for Beyond Visual Line of Sight (BVLOS) scenarios. Based on these analyses, the thesis proposes new exposure capabilities, including call control capability exposure and a Collision Avoidance API, to address identified gaps.

The exponential growth in mobile network traffic, driven by the rapid deployment of 5G technologies and the proliferation of new services, presents significant challenges for telecommunication operators. This thesis addresses these challenges by developing a predictive capacity management solution for 4G and 5G cellular networks. The primary objective is to forecast network traffic and identify potential congestion points up to one year in advance, enabling proactive network management and optimizing resource allocation, particularly through the use of spectral efficiency as a key predictive measure.

This study utilizes data from KPN’s Operations Support System (OSS), comprising 67 days of hourly data across the entire network, with a focus on predicting future traffic and network performance up to one year ahead. The methodology integrates historical data analysis, time series forecasting, and machine learning techniques. The approach combines Cumulative Distribution Function (CDF) modeling for traffic volume prediction with supervised machine learning algorithms, including Linear Regression, Lasso Regression, Random Forest, and CatBoost, to forecast Physical Resource Block (PRB) utilization and spectral efficiency at the sector level.

The detailed analysis identifies Lasso Regression as the most effective model for predicting spectral efficiency, with the lowest Mean Absolute Percentage Error (MAPE). Lasso’s ability to handle extrapolation beyond observed data ranges makes it particularly well-suited for long-term capacity management when combined with CDF-based traffic prediction. The findings demonstrate significant improvements in the accuracy of congestion predictions and the efficiency of resource utilization.

The study also revealed that, without additional resources, the number of congested sectors is expected to increase as traffic demand continues to grow. This highlights the critical need for new spectrum allocation to maintain service quality. Additionally, the research evaluated the impact of deploying new spectrum resources, such as the 3.5 GHz band, in specific sectors. The results showed that the deployment of the 3.5 GHz band significantly reduced congestion and improved network performance and user experience during the forecast period.

In modern society, critical operations, such as emergency response and public safety, rely on communication systems, in this context also referred to as mission critical systems. These systems must meet strict availability requirements, since any failure can lead to severe consequences, including loss of life. Traditionally, dedicated private communication systems, like local Push-to-Talk systems, were used for such critical operations, but there is a growing shift towards utilizing public 4G and 5G mobile networks to achieve better coverage, higher data speeds and more innovative features at a lower cost.

With the increasing complexity and vast amount of data from the network, automation and artificial intelligence (AI) are now becoming essential tools in these communication systems to efficiently manage data, configure networks in real-time, and effectively handle alarms. The use of AI can improve the end-to-end availability of mission critical systems, ensuring communication during critical situations.

The main goal of this research is to investigate whether and how the use of AI can improve the end-to-end availability of mission critical systems, with a specific focus on the Mission Critical Push-to-Talk (MCPTT) system of KPN, which is using the public 4G and 5G network. Currently, the KPN MCPTT system is being used with a relatively limited number of users. However, the vision for MCPTT extends beyond its current implementation, aiming to scale up this service. With an increasing number of users, using automation and AI is essential for optimizing and managing the complexities of this mission critical communication system.

The implementation of AI in the MCPTT system follows a systematic approach, starting with the independent analysis and monitoring of system specific elements. By focusing on these elements and using data such as Call Detail Records (CDR) and log data, insights into the system's behavior can be obtained. Through collaboration with system experts, AI algorithms can be trained to effectively detect anomalies, thereby enhancing the overall availability of the MCPTT system. Looking ahead, the integration of real-time data becomes crucial for proactive monitoring. Establishing a streamlined data pipeline facilitates the flow of real-time information, offering a comprehensive overview of system performance and enabling swift anomaly detection. It is concluded that the monitoring of individual system elements with the use of AI is a first step towards improving the end-to-end availability.

In order to ensure the correct use of AI throughout the complete cycle, it is crucial to look at explainability, safety, and data quality. These points should be included at each stage of the AI process. By ensuring explainability, system experts can gain insights into the decision-making process of the AI algorithms. By using safety mechanisms, potential risks and vulnerabilities can be mitigated. Maintaining data quality is essential to achieve accurate outcomes.

Multi-access Edge Computing (MEC) is a concept brought up by ETSI and it places computing, storage, processing and network resources into MEC hosts and places these MEC hosts as close as needed to the telecom network edge in order to reduce service latency and bandwidth usage. For self-driving vehicles, streaming video and real-time gaming, the devices involved (e.g. vehicles, cellphones, etc.) might not have enough capabilities to perform all the computations and might not have sufficient storage capacity; MEC can be used here for offloading data computations and content caching. To enhance service quality and user experience, MEC hosts and MEC applications should be located close(r) to the end-users, which increases the number of handovers between MEC hosts to maintain MEC service continuity for mobile end-users as well as the costs for the telecom operators. Therefore, a balance needs to be found. Consider the fact that mobile UEs need MEC service handovers to maintain service continuity and handovers may cause service interruptions which can cause severe degradation to MEC service qualities and user experience, hence the number of handovers between MEC hosts experienced by end-users should be minimized. To find a suitable deployment of MEC hosts and MEC applications in order to minimize the number of handovers, three greedy algorithms and two heuristic algorithms are introduced, implemented, tested, compared and analyzed in this thesis to see which identifies the deployment mechanism that has the smallest number of handovers. When it is time for a mobile UE to connect to a new MEC host and there are multiple potential choices of the new MEC host, the most suitable one for the UE needs to be determined dynamically according to the real-time condition of each possible MEC host. To achieve this, reinforcement learning is considered. Three different reinforcement learning algorithms based on SARSA learning and Deep Q Network are introduced, implemented, tested, compared and analyzed in this thesis. Furthermore, a decision-making mechanism is designed to cope with exceptional situations where the required service quality cannot be guaranteed.

The main goal of the thesis is to investigate how to optimize Quality of Experience (QoE) of users using applications over satellite links by application aware load balancing capabilities of SD-WAN. SES (Commercial satellite operator) customers want to use applications over satellite links that have high latency and are often more congested than terrestrial networks which results in lower Quality of Experience (QoE) of users. The applications have been designed and optimized for terrestrial networks, not for satellite networks. Thus, SES wants to use its hybrid (MEO/GEO) satellite network and application aware routing capabilities of SD-WAN to prioritize and steer traffic at the application layer based on intent and business rules and enforced via policy for appropriate QoE.
In the thesis, work is carried out in two parts: Firstly, experiments in lab to perform performance measurement of selected widely used applications over the different satellite links (GEO, MEO and LEO). Then performance of video applications over MEO link in different congestion scenarios (Unidirectional and Bidirectional Congestion) was measured. In order to improve the performance of video applications load balancing mechanism was defined to optimize QoE of the user. Secondly, a simulation model emulating a future SD-WAN scenario on Simulink, which is used to measure QoE of multiple users is designed. A load balancing mechanism which not only optimizes the QoE for multiple users but is also a cost effective alternative to manage the QoE is proposed.
It was concluded that applications belonging to the same category have varied performances in different congestion scenarios on satellite links. Hence, each application has its performance, variation and should be dealt with accordingly. Identifying performance thresholds in different scenarios is essential to derive load balancing mechanisms to improve QoE and optimize the cost. Key applications that drive the behaviour of experience should be identified (which differs in each use case and for different customers) and steered accordingly to the best possible link so that overall QoE could be improved. Recommendations on the designing of policies for different use cases and overall development of SD-WAN as a product have also been presented in the thesis.