A5002919366- Advanced MIMO Systems Optimization
- Expert finding and Q&A systems
- Wireless Networks and Protocols
- Knowledge Management and Sharing
- Software System Performance and Reliability
- Complex Network Analysis Techniques
- IoT and Edge/Fog Computing
- Reinforcement Learning in Robotics
- Energy Harvesting in Wireless Networks
- Recommender Systems and Techniques
- Satellite Communication Systems
- Evolutionary Algorithms and Applications
- Wikis in Education and Collaboration
- Web Data Mining and Analysis
- Digital Marketing and Social Media
- ICT Impact and Policies
- IoT-based Smart Home Systems
- Green IT and Sustainability
- Opinion Dynamics and Social Influence
- Power Line Communications and Noise
- Anomaly Detection Techniques and Applications
- IoT Networks and Protocols
- Social Media and Politics
- Security and Verification in Computing
- Cooperative Communication and Network Coding
Ericsson (Ireland)
2019-2022
Ollscoil na Gaillimhe – University of Galway
2014-2021
Online discussion boards, or Internet forums, are a significant part of the Internet. People use forums to post questions, provide advice and participate in discussions. These online conversations represented as threads, conversation trees within these threads important understanding behaviour users. Unfortunately, reply structures generally not publicly accessible maintained. Hence, this paper, we introduce an efficient simple approach reconstruct structure threaded conversations. We contrast...
The future of the manufacturing industry highly depends on digital systems that transform existing production and monitoring into autonomous fulfilling stringent requirements in terms availability, reliability, security, low latency, positioning with high accuracy.In order to meet such requirements, private 5G networks are considered a key enabling technology.In this paper, we introduce 5G-CLARITY system integrates new radio (5GNR), Wi-Fi light fidelity (LiFi) access networks, develops novel...
Remote Electrical Tilt (RET) optimization is an efficient method for adjusting the vertical tilt angle of Base Stations (BSs) antennas in order to optimize Key Performance Indicators (KPIs) network. Reinforcement Learning (RL) provides a powerful framework RET because its self-learning capabilities and adaptivity environmental changes. However, RL agent may execute unsafe actions during course interaction, i.e., resulting undesired network performance degradation. Since reliability services...
The move towards more complex and dynamic telecommunication networks renders the need for automation in management system stringent. This is especially true context of 5G networks, where understanding providing right recommendations crucial. Our work focuses on gathering from current status network correlating it with useful information existing documents provider operator's domain. We propose an architecture that, based collected interlinked information, can provide automatic addressing...
Safe interaction with the environment is one of most challenging aspects Reinforcement Learning (RL) when applied to real-world problems. This particularly important unsafe actions have a high or irreversible negative impact on environment. In context network management operations, Remote Electrical Tilt (RET) optimisation safety-critical application in which exploratory modifications antenna tilt angles base stations can cause significant performance degradation network. this paper, we...
As telecom networks generate high-dimensional data, it becomes important to support large numbers of co-existing network attributes and provide an interpretable eXplainable Artificial Intelligence (XAI) anomaly detection system. Most state-of-the-art techniques tackle the problem detecting anomalies with high precision but models don't solution. This makes hard for operators adopt given solutions. The proposed Cluster Characterized Autoencoder (CCA) architecture improves model...
Today we see more and systems support intent languages, frameworks interfaces. The incorporation of these approaches provide elements abstraction between the request execution. However mapping is still required to get from abstract real. This often introduces rigidity into system. work presents a flexible approach driven through an interpreter for realization. mechanism demonstrated interactions with mock slice manager functions running on open source, as service platform. accepts messages...
The Internet of Things (IoT) has seen expeditious growth in recent times with 7 billion connected devices 2020, thus leading to the vital importance real-time monitoring IoT devices. Through this paper, we demonstrate idea building a cloud-native application monitor smart home intends provide valuable performance metrics from perspective end-users and react anomalies real-time. In demo conduct demonstration using Autoencoder (an unsupervised technique) based Deep Neural Networks (DNNs) learn...
The adoption of 5G has recently picked up pace and across the globe commercial deployments are more numerous. In addition to better performance, technology brings, among others, management operation flexibility, thus allowing industry verticals exploit features telecommunication networks that go far beyond new mobile access capabilities. As with any technology, integration, testing validation vertical applications pose great challenges at both operational levels. this context, there is an...
This research has been conducted with the financial support of Science Foundation Ireland (SFI) under Grant Number SFI/12/RC/2289.
Maintaining online communities is vital in order to increase and retain their economic social value. That why community managers look gauge the success of by measuring a variety user behaviour, such as member activity, turnover interaction. However, vary widely purpose, implementation demographics, although many indicators have been proposed literature, we will show that there no one-fits-all approach success: Different criteria depend on different behaviour. To demonstrate this, put...
Researchers and community managers try to measure the success of online communities using a variety determinants, such as member activity, turnover interaction. Although many determinants have been proposed in literature, it remains largely unclear which are applicable kind communities. In this work, we focus on use case question answering (Q&A) communities, where can asker satisfaction an aspect towards goal solving questions. We evaluate how well correlate satisfaction, find that less...
Remote Electrical Tilt (RET) optimization is an efficient method for adjusting the vertical tilt angle of Base Stations (BSs) antennas in order to optimize Key Performance Indicators (KPIs) network. Reinforcement Learning (RL) provides a powerful framework RET because its self-learning capabilities and adaptivity environmental changes. However, RL agent may execute unsafe actions during course interaction, i.e., resulting undesired network performance degradation. Since reliability services...
In the advancement of a better understanding user behaviour on Internet, clustering approaches are used as means to group and categorise online communities. Although indicate similarity between communities (e.g. through an priori defined distance function based features) they do not show what causes differences types this paper, we study effects that certain topics may have behaviour. We cluster 29 Stack Exchange platform, correlation clustered three latent topic properties Accessibility,...
Safe interaction with the environment is one of most challenging aspects Reinforcement Learning (RL) when applied to real-world problems. This particularly important unsafe actions have a high or irreversible negative impact on environment. In context network management operations, Remote Electrical Tilt (RET) optimisation safety-critical application in which exploratory modifications antenna tilt angles base stations can cause significant performance degradation network. this paper, we...