George Iosifidis
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
George Iosifidis has a total of 167 co-authors. Here are the top co-authors by number of shared publications:
Leandros Tassiulas
Electrical Engineering and Institute for Network Science, Yale University, New Haven, CT, USA
39 shared publicationsKonstantinos Poularakis
Yale Univ, New Haven, CT 06520 USA
14 shared publicationsIordanis Koutsopoulos
Athens University of Economics Business
10 shared publicationsLin Gao
9 shared publications- Jianwei Huang
The Chinese University of Hong Kong
8 shared publications - Leandros Tassiulas
Yale University, New Haven, CT, USA
8 shared publications
Latest Publications
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 44 (07338716) , 3251-3267
Meta-Learning-Based Handover Management in NextG O-RAN
While traditional handovers (THOs) have served as a backbone for mobile connectivity, they increasingly suffer from failures and delays, especially in dense deployments and high-frequency bands. To address these limitations, 3GPP introduced Conditional Handovers (CHOs) that enable proactive cell reservations and user-driven execution. However, both handover (HO) types present intricate trade-offs in signaling, resource usage, and reliability. This paper presents unique, countrywide mobility management datasets from a top-tier mobile network operator (MNO) that offer fresh insights into these issues and call for adaptive and robust HO control in next-generation networks. Motivated by these findings, we propose CONTRA, a framework that, for the first time, jointly optimizes THOs and CHOs within the O-RAN architecture. We study two variants of CONTRA: one where users are a priori assigned to one of the HO types, reflecting distinct service or user-specific requirements, as well as a more dynamic formulation where the controller decides on-the-fly the HO type, based on system conditions and needs. To this end, it relies on a practical meta-learning algorithm that adapts to runtime observations and guarantees performance comparable to an oracle with perfect future information (universal no-regret). CONTRA is specifically designed for near-real-time deployment as an O-RAN xApp and aligns with the 6G goals of flexible and intelligent control. Extensive evaluations leveraging crowdsourced datasets show that CONTRA improves user throughput and reduces both THO and CHO switching costs, outperforming 3GPP-compliant and Reinforcement Learning (RL) baselines in dynamic and real-world scenarios.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 44 (07338716) , 3251-3267
Meta-Learning-Based Handover Management in NextG O-RAN
While traditional handovers (THOs) have served as a backbone for mobile connectivity, they increasingly suffer from failures and delays, especially in dense deployments and high-frequency bands. To address these limitations, 3GPP introduced Conditional Handovers (CHOs) that enable proactive cell reservations and user-driven execution. However, both handover (HO) types present intricate trade-offs in signaling, resource usage, and reliability. This paper presents unique, countrywide mobility management datasets from a top-tier mobile network operator (MNO) that offer fresh insights into these issues and call for adaptive and robust HO control in next-generation networks. Motivated by these findings, we propose CONTRA, a framework that, for the first time, jointly optimizes THOs and CHOs within the O-RAN architecture. We study two variants of CONTRA: one where users are a priori assigned to one of the HO types, reflecting distinct service or user-specific requirements, as well as a more dynamic formulation where the controller decides on-the-fly the HO type, based on system conditions and needs. To this end, it relies on a practical meta-learning algorithm that adapts to runtime observations and guarantees performance comparable to an oracle with perfect future information (universal no-regret). CONTRA is specifically designed for near-real-time deployment as an O-RAN xApp and aligns with the 6G goals of flexible and intelligent control. Extensive evaluations leveraging crowdsourced datasets show that CONTRA improves user throughput and reduces both THO and CHO switching costs, outperforming 3GPP-compliant and Reinforcement Learning (RL) baselines in dynamic and real-world scenarios.
- 16 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (22) , 46611-46625
Cooperative Edge Inferences With Online Learning
The efficient execution of inferences at the edge is becoming increasingly critical for communication systems that are expected to provide users with fast and accurate mobile data analytics. These inference tasks are inherently latency-sensitive and computationally demanding, whereas edge nodes are limited by energy budgets and heterogeneous resources. This article studies how a set of edge nodes can collaborate in executing demanding streaming inference tasks to optimize their aggregate performance. Such collaborative task exchange schemes enable the sharing of scarce computing resources and machine learning (ML) models (which perform the inferences) and constitute a scalable approach to this intricate problem. We formulate this exchange process as an online convex optimization (OCO) problem and design a dynamic task assignment algorithm, which is proven to have optimality guarantees even when the network and service parameters (resources and task properties) are unknown and vary arbitrarily over time. The algorithm aims to maximize inference accuracy while minimizing overall task latency and energy (including for data transfers) and simultaneously ensures that collaborating nodes do not suffer imbalanced energy costs. Through a series of data-driven experiments, we quantify the cooperation benefits under different weight combinations and validate the convergence and adaptability of the proposed learning algorithm across diverse conditions, including variations of system parameters, as well as heterogeneity across nodes and tasks.
- 16 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (22) , 46611-46625
Cooperative Edge Inferences With Online Learning
The efficient execution of inferences at the edge is becoming increasingly critical for communication systems that are expected to provide users with fast and accurate mobile data analytics. These inference tasks are inherently latency-sensitive and computationally demanding, whereas edge nodes are limited by energy budgets and heterogeneous resources. This article studies how a set of edge nodes can collaborate in executing demanding streaming inference tasks to optimize their aggregate performance. Such collaborative task exchange schemes enable the sharing of scarce computing resources and machine learning (ML) models (which perform the inferences) and constitute a scalable approach to this intricate problem. We formulate this exchange process as an online convex optimization (OCO) problem and design a dynamic task assignment algorithm, which is proven to have optimality guarantees even when the network and service parameters (resources and task properties) are unknown and vary arbitrarily over time. The algorithm aims to maximize inference accuracy while minimizing overall task latency and energy (including for data transfers) and simultaneously ensures that collaborating nodes do not suffer imbalanced energy costs. Through a series of data-driven experiments, we quantify the cooperation benefits under different weight combinations and validate the convergence and adaptability of the proposed learning algorithm across diverse conditions, including variations of system parameters, as well as heterogeneity across nodes and tasks.
- Open Access
CHOMET: Conditional Handovers via Meta-Learning
Handovers (HOs) are the cornerstone of modern cellular networks for enabling seamless connectivity to a vast and diverse number of mobile users. However, as mobile networks become more complex with more diverse users and smaller cells, traditional HOs face significant challenges, such as prolonged delays and increased failures. To mitigate these issues, 3GPP introduced conditional handovers (CHOs), a new type of HO that enables the preparation (i.e., resource allocation) of multiple cells for a single user to increase the chance of HO success and decrease the delays in the procedure. Despite its advantages, CHO introduces new challenges that must be addressed, including efficient resource allocation and managing signaling/communication overhead from frequent cell preparations and releases. This paper presents a novel framework aligned with the O-RAN paradigm that leverages meta-learning for CHO optimization, providing robust dynamic regret guarantees and demonstrating at least 180% superior performance than other 3GPP benchmarks in volatile signal conditions.
Open AccessCHOMET: Conditional Handovers via Meta-Learning
Handovers (HOs) are the cornerstone of modern cellular networks for enabling seamless connectivity to a vast and diverse number of mobile users. However, as mobile networks become more complex with more diverse users and smaller cells, traditional HOs face significant challenges, such as prolonged delays and increased failures. To mitigate these issues, 3GPP introduced conditional handovers (CHOs), a new type of HO that enables the preparation (i.e., resource allocation) of multiple cells for a single user to increase the chance of HO success and decrease the delays in the procedure. Despite its advantages, CHO introduces new challenges that must be addressed, including efficient resource allocation and managing signaling/communication overhead from frequent cell preparations and releases. This paper presents a novel framework aligned with the O-RAN paradigm that leverages meta-learning for CHO optimization, providing robust dynamic regret guarantees and demonstrating at least 180% superior performance than other 3GPP benchmarks in volatile signal conditions.
- Open Access
- 28 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Communications Society
- Vol. 6, 2474-2485
Multi-Objective Reverse Offloading in Edge Computing for AI Tasks
Offloading tasks between edge nodes is a subject that has drawn a lot of attention since edge computing first emerged. A large number of edge IoT devices utilizing increased computing resources such as autonomous vehicles and UAVs can be used to execute AI tasks close to users. We present a novel approach that deviates from the conventional edge computing offloading concept namely offloading computationally intensive tasks from cloudlets to nearby end nodes. Specifically, we enhance a scenario where end nodes assist more powerful nodes (like cloudlets) in executing AI inference tasks. In edge computing networks, as end nodes grow in number, they build an idle computing capacity which can solve and provide efficient solutions. Our goal is to solve a defined Multi-Objective optimization problem with three objectives namely the overall execution time (slowest substasks), the execution accuracy, and the total energy consumption. We address this challenging optimization problem using a novel method with our released Multi-Objective Edge AI-Adaptive Reverse Offloading, or MOEAI-ARO, algorithm. Using an edge computing testbed and a representative AI service, we demonstrate the effectiveness of our reverse offloading proposal and method. The results indicate that our method further optimizes the system’s performance compared to baseline algorithms.
- 28 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Communications Society
- Vol. 6, 2474-2485
Open AccessMulti-Objective Reverse Offloading in Edge Computing for AI Tasks
Offloading tasks between edge nodes is a subject that has drawn a lot of attention since edge computing first emerged. A large number of edge IoT devices utilizing increased computing resources such as autonomous vehicles and UAVs can be used to execute AI tasks close to users. We present a novel approach that deviates from the conventional edge computing offloading concept namely offloading computationally intensive tasks from cloudlets to nearby end nodes. Specifically, we enhance a scenario where end nodes assist more powerful nodes (like cloudlets) in executing AI inference tasks. In edge computing networks, as end nodes grow in number, they build an idle computing capacity which can solve and provide efficient solutions. Our goal is to solve a defined Multi-Objective optimization problem with three objectives namely the overall execution time (slowest substasks), the execution accuracy, and the total energy consumption. We address this challenging optimization problem using a novel method with our released Multi-Objective Edge AI-Adaptive Reverse Offloading, or MOEAI-ARO, algorithm. Using an edge computing testbed and a representative AI service, we demonstrate the effectiveness of our reverse offloading proposal and method. The results indicate that our method further optimizes the system’s performance compared to baseline algorithms.
- Open Access
- 23 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 24 (6) , 5400-5417
Minimization of the Training Makespan in Hybrid Federated Split Learning
Parallel Split Learning (SL) allows resource-constrained devices that cannot participate in Federated Learning (FL) to train deep neural networks (NNs) by splitting the NN model into parts. In particular, such devices (clients) may offload the processing task of the largest model part to a computationally powerful helper, and multiple helpers may be employed and work in parallel. In hybrid federated and split learning (HFSL), on the other hand, devices can participate in the training process through any of the two protocols (SL and FL), depending on the system's characteristics. This could considerably reduce the maximum training time over all clients (makespan), especially in highly heterogeneous scenarios. In this paper, we study the joint problem of the training protocol selection, client-helper assignments, and scheduling decisions, to minimize the training makespan. We prove this problem is NP-hard and propose two solution methods: one based on the decomposition of the problem by leveraging its inherent symmetry, and a second fully scalable one. Through numerical evaluations using our testbed's measurements, we build a solution strategy comprising these methods. Moreover, this strategy finds a near-optimal solution and achieves a shorter makespan than the baseline schemes by up to 71%.
- 23 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 24 (6) , 5400-5417
Open AccessMinimization of the Training Makespan in Hybrid Federated Split Learning
Parallel Split Learning (SL) allows resource-constrained devices that cannot participate in Federated Learning (FL) to train deep neural networks (NNs) by splitting the NN model into parts. In particular, such devices (clients) may offload the processing task of the largest model part to a computationally powerful helper, and multiple helpers may be employed and work in parallel. In hybrid federated and split learning (HFSL), on the other hand, devices can participate in the training process through any of the two protocols (SL and FL), depending on the system's characteristics. This could considerably reduce the maximum training time over all clients (makespan), especially in highly heterogeneous scenarios. In this paper, we study the joint problem of the training protocol selection, client-helper assignments, and scheduling decisions, to minimize the training makespan. We prove this problem is NP-hard and propose two solution methods: one based on the decomposition of the problem by leveraging its inherent symmetry, and a second fully scalable one. Through numerical evaluations using our testbed's measurements, we build a solution strategy comprising these methods. Moreover, this strategy finds a near-optimal solution and achieves a shorter makespan than the baseline schemes by up to 71%.
- Open Access
Smooth Handovers via Smoothed Online Learning
With users demanding seamless connectivity, handovers (HOs) have become a fundamental element of cellular networks. However, optimizing HOs is a challenging problem, further exacerbated by the growing complexity of mobile networks. This paper presents the first countrywide study of HO optimization, through the prism of Smoothed Online Learning (SOL). We first analyze an extensive dataset from a commercial mobile network operator (MNO) in Europe with more than 40M users, to understand and reveal important features and performance impacts on HOs. Our findings highlight a correlation between HO failures/delays, and the characteristics of radio cells and end-user devices, showcasing the impact of heterogeneity in mobile networks nowadays. We subsequently model UE-cell associations as dynamic decisions and propose a realistic system model for smooth and accurate HOs that extends existing approaches by (i) incorporating device and cell features on HO optimization, and (ii) eliminating (prior) strong assumptions about requiring future signal measurements and knowledge of end-user mobility. Our algorithm, aligned with the O-RAN paradigm, provides robust dynamic regret guarantees, even in challenging environments, and shows superior performance in multiple scenarios with real-world and synthetic data.
Open AccessSmooth Handovers via Smoothed Online Learning
With users demanding seamless connectivity, handovers (HOs) have become a fundamental element of cellular networks. However, optimizing HOs is a challenging problem, further exacerbated by the growing complexity of mobile networks. This paper presents the first countrywide study of HO optimization, through the prism of Smoothed Online Learning (SOL). We first analyze an extensive dataset from a commercial mobile network operator (MNO) in Europe with more than 40M users, to understand and reveal important features and performance impacts on HOs. Our findings highlight a correlation between HO failures/delays, and the characteristics of radio cells and end-user devices, showcasing the impact of heterogeneity in mobile networks nowadays. We subsequently model UE-cell associations as dynamic decisions and propose a realistic system model for smooth and accurate HOs that extends existing approaches by (i) incorporating device and cell features on HO optimization, and (ii) eliminating (prior) strong assumptions about requiring future signal measurements and knowledge of end-user mobility. Our algorithm, aligned with the O-RAN paradigm, provides robust dynamic regret guarantees, even in challenging environments, and shows superior performance in multiple scenarios with real-world and synthetic data.
- Open Access
- 4 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 51-67
Through the Telco Lens: A Countrywide Empirical Study of Cellular Handovers
Cellular networks rely on handovers (HOs) as a fundamental element to enable seamless connectivity for mobile users. A comprehensive analysis of HOs can be achieved through data from Mobile Network Operators (MNOs); however, the vast majority of studies employ data from measurement campaigns within confined areas and with limited end-user devices, thereby providing only a partial view of HOs. This paper presents the first countrywide analysis of HO performance, from the perspective of a top-tier MNO in a European country. We collect traffic from approximately 40M users for 4 weeks and study the impact of the radio access technologies (RATs), device types, and manufacturers on HOs across the country. We characterize the geo-temporal dynamics of horizontal (intra-RAT) and vertical (inter-RATs) HOs, at the district level and at millisecond granularity, and leverage open datasets from the country's official census office to associate our findings with the population. We further delve into the frequency, duration, and causes of HO failures, and model them using statistical tools. Our study offers unique insights into mobility management, highlighting the heterogeneity of the network and devices, and their effect on HOs.
- 4 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 51-67
Open AccessThrough the Telco Lens: A Countrywide Empirical Study of Cellular Handovers
Cellular networks rely on handovers (HOs) as a fundamental element to enable seamless connectivity for mobile users. A comprehensive analysis of HOs can be achieved through data from Mobile Network Operators (MNOs); however, the vast majority of studies employ data from measurement campaigns within confined areas and with limited end-user devices, thereby providing only a partial view of HOs. This paper presents the first countrywide analysis of HO performance, from the perspective of a top-tier MNO in a European country. We collect traffic from approximately 40M users for 4 weeks and study the impact of the radio access technologies (RATs), device types, and manufacturers on HOs across the country. We characterize the geo-temporal dynamics of horizontal (intra-RAT) and vertical (inter-RATs) HOs, at the district level and at millisecond granularity, and leverage open datasets from the country's official census office to associate our findings with the population. We further delve into the frequency, duration, and causes of HO failures, and model them using statistical tools. Our study offers unique insights into mobility management, highlighting the heterogeneity of the network and devices, and their effect on HOs.
- Open Access
- 9 October 2024
- Published by Elsevier in Computer Networks
- Vol. 255
Adaptive reverse task offloading in edge computing for AI processes
- 9 October 2024
- Published by Elsevier in Computer Networks
- Vol. 255
Open AccessAdaptive reverse task offloading in edge computing for AI processes
- 3 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (22) , 38352-38364
Balancing Energy Preservation and Performance in Energy-Harvesting Sensor Networks
The development of environmentally friendly, green communications is at the forefront of designing future Internet of Things (IoT) networks, although many opportunities to improve energy conservation from Energy Harvesting (EH) sensors remain unexplored. Ubiquitous computing power, available in the form of cloudlets, enables processing of the collected observations at the network edge. Often, the information that the Artificial Intelligence of Things (AIoT) application obtains by processing observations from one sensor can also be obtained by processing observations from another sensor. Consequently, a sensor can take advantage of the correlation between processed observations to avoid unnecessary transmissions and save energy. For example, when two cameras monitoring the same intersection detect the same vehicles, the system can recognize this overlap and reduce redundant data transmissions. This approach allows the network to conserve energy while still ensuring accurate vehicle detection, thereby maintaining the overall performance of the AIoT task. In this paper, we consider such a system and develop a novel solution named Balancing Energy Efficiency in Sensor Networks with Multi-Agent Reinforcement Learning (BEES-MARL). Our proposed solution is capable of taking advantage of correlations in a system with multiple EH-powered sensors observing the same scene and transmitting their observations to a cloudlet. We evaluate the proposed solution in two data-driven use cases to verify its benefits and in a general setting to demonstrate scalability. Our solution improves task performance, measured by recall, by up to 16% over a heuristic approach, while minimizing latency and preventing outages.
- 3 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (22) , 38352-38364
Balancing Energy Preservation and Performance in Energy-Harvesting Sensor Networks
The development of environmentally friendly, green communications is at the forefront of designing future Internet of Things (IoT) networks, although many opportunities to improve energy conservation from Energy Harvesting (EH) sensors remain unexplored. Ubiquitous computing power, available in the form of cloudlets, enables processing of the collected observations at the network edge. Often, the information that the Artificial Intelligence of Things (AIoT) application obtains by processing observations from one sensor can also be obtained by processing observations from another sensor. Consequently, a sensor can take advantage of the correlation between processed observations to avoid unnecessary transmissions and save energy. For example, when two cameras monitoring the same intersection detect the same vehicles, the system can recognize this overlap and reduce redundant data transmissions. This approach allows the network to conserve energy while still ensuring accurate vehicle detection, thereby maintaining the overall performance of the AIoT task. In this paper, we consider such a system and develop a novel solution named Balancing Energy Efficiency in Sensor Networks with Multi-Agent Reinforcement Learning (BEES-MARL). Our proposed solution is capable of taking advantage of correlations in a system with multiple EH-powered sensors observing the same scene and transmitting their observations to a cloudlet. We evaluate the proposed solution in two data-driven use cases to verify its benefits and in a general setting to demonstrate scalability. Our solution improves task performance, measured by recall, by up to 16% over a heuristic approach, while minimizing latency and preventing outages.
- 16 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 73 (3) , 1787-1800
Adaptive Resource Allocation for Virtualized Base Stations in O-RAN With Online Learning
Open RAN systems, with their virtualized base stations (vBSs), offer increased flexibility and reduced costs, vendor diversity, and interoperability. However, optimizing the allocation of radio resources in such systems raises new challenges due to the volatile vBSs operation, and the dynamic network conditions and user demands they are called to support. Leveraging the novel O-RAN multi-tier control architecture, we propose a new set of resource allocation threshold policies with the aim of balancing the vBSs’ performance and energy consumption in a robust and provably optimal fashion. To that end, we introduce an online learning algorithm that operates under minimal assumptions and without requiring knowledge of the environment, hence being suitable even for “challenging” environments with non-stationary or adversarial demands and conditions. We also develop a meta-learning scheme that utilizes other available algorithmic schemes, e.g., tailored for more “easy” environments, by choosing dynamically the best-performing algorithm; thus enhancing the system’s effectiveness. We prove that the proposed solutions achieve sub-linear regret (zero optimality gap), and characterize their dependence on the main system parameters. The performance of the algorithms is evaluated with real-world data from a testbed, in stationary and adversarial conditions, indicating energy savings of up to 64.5% compared with several state-of-the-art benchmarks.
- 16 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 73 (3) , 1787-1800
Adaptive Resource Allocation for Virtualized Base Stations in O-RAN With Online Learning
Open RAN systems, with their virtualized base stations (vBSs), offer increased flexibility and reduced costs, vendor diversity, and interoperability. However, optimizing the allocation of radio resources in such systems raises new challenges due to the volatile vBSs operation, and the dynamic network conditions and user demands they are called to support. Leveraging the novel O-RAN multi-tier control architecture, we propose a new set of resource allocation threshold policies with the aim of balancing the vBSs’ performance and energy consumption in a robust and provably optimal fashion. To that end, we introduce an online learning algorithm that operates under minimal assumptions and without requiring knowledge of the environment, hence being suitable even for “challenging” environments with non-stationary or adversarial demands and conditions. We also develop a meta-learning scheme that utilizes other available algorithmic schemes, e.g., tailored for more “easy” environments, by choosing dynamically the best-performing algorithm; thus enhancing the system’s effectiveness. We prove that the proposed solutions achieve sub-linear regret (zero optimality gap), and characterize their dependence on the main system parameters. The performance of the algorithms is evaluated with real-world data from a testbed, in stationary and adversarial conditions, indicating energy savings of up to 64.5% compared with several state-of-the-art benchmarks.