Selim Ickin- Ph.D.
- Senior Specialist AI at Ericsson
Selim Ickin
- Ph.D.
- Senior Specialist AI at Ericsson
About
39
Publications
7,595
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
562
Citations
Introduction
Skills and Expertise
Current institution
Publications
Publications (39)
Optimization of radio hardware and AI-based network management software yield significant energy savings in radio access networks. The execution of underlying Machine Learning (ML) models, which enable energy savings through recommended actions, may require additional compute and energy, highlighting the opportunity to explore and adopt accurate an...
With emerging ML techniques showing great potential in prediction performance, there is an inclination towards ML for network management. Although modern mobile network deployments have access to essential data storage and computational resources, they would benefit remarkably from more sustainable and cost-effective approaches for ML applications....
In telecommunication networks, many factors that influence QoE are inherently distributed in the network; related to creation, delivery, and presentation of the content at the end-user terminal. Split Learning (SL) is a scalable distributed machine learning (ML) technique that enables joint training and inference on decentralized datasets. These de...
In telecommunications, information delivery is performed over inherently distributed elements such as application servers, packet core network equipment, radio base stations, and mobile user equipment (UE). Predicting key performance indicators (KPIs) for services is important for mobile operators in preventing customer churn. In addition, it is im...
Reliable forecast of COVID-19 hospital admissions in near-term horizons can help enable effective resource management which is vital in reducing pressure from healthcare services. The use of mobile network data has come to attention in response to COVID-19 pandemic leveraged on their ability in capturing people social behavior. Crucially, we show t...
Reliable near-time forecast of COVID-19 hospital admissions can help enable effective resource management which is vital in reducing pressure from healthcare services. The use of mobile network data has come to attention in response to COVID-19 pandemic leveraged on their ability in capturing people social behaviour. Crucially, we show that there a...
The development of Quality of Experience (QoE) models using Machine Learning (ML) is challenging, since it can be difficult to share datasets between research entities to protect the intellectual property of the ML model and the confidentiality of user studies in compliance with data protection regulations such as General Data Protection Regulation...
Increasing complexity in management of immense number of network elements and their dynamically changing environment necessitates machine learning based recommendation models to guide human experts in setting appropriate network configurations to sustain end-user Quality of Experience (QoE). In this paper, we present and demonstrate a generative Co...
Rapid change in sensitive behaviour and profile of distributed mobile network elements necessitates privacy preserving distributed learning mechanism such as Federated Learning. Moreover, this mechanism needs to be robust that seamlessly sustains the jointly trained model accuracy. In order to provide a automated management of the learning process...
Our estimates indicate that the cost of the energy required to power networks represents between 10-30% of the network operating expenses of a communication service provider (CSP), depending on the specificities of its local energy market. In total, this expenditure adds up to approximately 25 billion USD per year [1].
The importance of cellular networks continuously increases as we assume ubiquitous connectivity in our daily lives. As a result, the underlying core telecom systems have very high reliability and availability requirements, that are sometimes hard to meet. This study presents a proactive approach that could aid satisfying these high requirements on...
Quality of Experience (QoE) models need good generalization that necessitates sufficient amount of user-labeled datasets associated with measurements related to underlying QoE factors. However, obtaining QoE datasets is often costly, since they are preferably collected from many subjects with diverse background, and eventually dataset sizes and rep...
The development of QoE models by means of Machine Learning (ML) is challenging, amongst others due to small-size datasets, lack of diversity in user profiles in the source domain, and too much diversity in the target domains of QoE models. Furthermore, datasets can be hard to share between research entities, as the machine learning models and the c...
Streaming data services, such as video-on-demand, are getting increasingly more popular, and they are expected to account for more than 80% of all Internet traffic in 2020. In this context, it is important for streaming service providers to detect deviations in service requests due to issues or changing end-user behaviors in order to ensure that en...
Machine Learning (ML) based Quality of Experience (QoE) models potentially suffer from over-fitting due to limitations including low data volume, and limited participant profiles. This prevents models from becoming generic. Consequently, these trained models may under-perform when tested outside the experimented population. One reason for the limit...
Machine Learning based Quality of Experience (QoE) models potentially suffer from over-fitting due to limitations including low data volume, and limited participant profiles. This prevents models from becoming generic. Consequently, these trained models may under-perform when tested outside the experimented population. One reason for the limited da...
Practitioners on the area of mobile application development usually rely on set of app-related success factors, the majority of which are directly related to their economical/business profit (e.g., number of downloads, or the in-app purchases revenue). However, gathering also the user-related success factors, that explain the reasons why users choo...
The most energy-consuming applications in battery life-constrained smartphones are the ones that comprise data transmission, especially via the 3G interface. Scheduling download activities on smartphones is especially necessary, if there are multiple asynchronous downloads scattered over a long duration. The latter scenario highly increases the ene...
Wireless networks have become more and more popular because of ease of installation, ease of access, and support of smart terminals and gadgets on the move. In the overall life cycle of providing green wireless technology, from production to operation and, finally, removal, this chapter focuses on the operation phase and summarizes insights in ener...
The usage of network-demanding applications is growing rapidly such as video streaming on mobile terminals. However, network and/or service providers might not guarantee the perceived quality for video streaming that demands high packet transmission rate. In order to satisfy the user expectations and to minimize user churn, it is important for netw...
One of the most influencing factors on the overall end-user perceived quality from applications and services, i.e., QoE, running on the smartphones is their limited battery life. Particular cloud-based applications/services on the smartphone with a constrained battery life might consume high energy even when the smartphone is in screen-OFF state. T...
The experimental setting of Human Mobile Computer Interaction (HCI) studies is moving from the controlled laboratory to the user's daily-life environments, while employing the users' own smartphones. These studies are challenging for both new and expert researchers in human subject studies in the HCI field. Within the last three years, we conducted...
Subjective performance of smartphone-based high bandwidth- and energy-demanding applications and services such as video streaming, are highly influenced by the temporal impairments perceived by the user at the user interface; and the application's energy consumption patterns. Therefore, we study the influence of the anomalies detected by objective...
Evaluating video Quality of Experience (QoE) on a mobile phone has not yet been studied much. It is common that the data collected through user studies in mobile platform involves high fluctuation of user ratings without obvious reasons related to variation in network level. User disparity, user's various intermediate or previous experiences, video...
Increasingly, we use mobile applications and services in our daily life activities, to support our needs for information, communication or leisure. However, user acceptance of a mobile application depends on at least two conditions: the application's perceived experience, and the appropriateness of the application to the user¿s context and needs. H...
The smartphone usage nearly tripled in 2011 according to Cisco Virtual Networking Index. There is a high demand of energy for using popular mobile applications, which run on smartphones with limited battery life. Video streaming applications are widely used on mobile devices, and their high power consumption exhibits high variance during a live str...
Mobile applications and services increasingly assist us in our daily life situations, fulfilling our needs for information, communication, entertainment or leisure. However, user acceptance of a mobile application depends on at least two conditions; the application's perceived Quality of Experience (QoE) and the appropriateness of the application t...
Quality of Experience is a parameter used to express the relationship between Quality of Service and the satisfaction of network service subscribers. The modeling of Quality of Experience demands for solving a multidimensional problem. In this paper, we present a Quality of Experience analysis of streaming videos. Related to this, we show that we c...
The satisfaction of end-users is important when evaluating services and products. Visualizing the network behavior in mobile streaming as well as modeling the correlation between Quality of Service (QoS) and Quality of Experience (QoE) is expected to improve user satisfaction of services on the Internet. Given that network and human factors are the...
This demo shows a prototype of a user-centric mobility framework that provides handover for macro-mobility on handheld devices. The framework is designed for mobile- controlled handover and does not require modification of the Internet infrastructure. The end-users are so able to control the roaming process governed by user considerations in additi...
Being connected to a wireless access point is not enough; in addition, adequate per-application performance is expected. To achieve high user perception, it is necessary to adapt the ongoing connection to the frequently changing network conditions. Not only collecting Quality of Service (QoS) metrics but analyzing these on a per-technology basis is...
