Salvador Luna-Ramírez
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Salvador Luna-Ramírez has a total of 57 co-authors. Here are the top co-authors by number of shared publications:
- Matias Toril
Instituto de Telecomunicacin (TELMA), CEI Andaluca TECH, E.T.S. Ingeniera de Telecomunicacin, Universidad de Mlaga, Malaga, Spain
53 shared publications - Carolina Gijon
Instituto de Telecomunicacin (TELMA), CEI Andaluca TECH, E.T.S. Ingeniera de Telecomunicacin, Universidad de Mlaga, Malaga, Spain
14 shared publications - Isabel De-La-Bandera
Department of Communications Engineering, Telecommunication Research Institute, University of Malaga, Malaga, Spain
8 shared publications - Victor Buenestado
Ericsson
8 shared publications F. Ruiz
7 shared publications- Jose Maria Ruiz-Aviles
Department of Communications Engineering, University of Malaga, Malaga, Spain
7 shared publications
Latest Publications
- Open Access
- 27 August 2026
- Published by MDPI AG in Machine Learning and Knowledge Extraction
- Vol. 8 (9) , 260
GRIMCELL: A Graph Neural Network to Predict the Impact of New Cells in Mobile Networks
Network densification will play a vital role in next-generation cellular networks, as it addresses coverage and capacity issues in the radio access domain. However, new cell deployments can negatively impact neighbor cells, requiring careful planning. Unfortunately, the complexity of advanced radio resource management schemes implemented in Beyond 5G and future 6G networks makes it challenging to predict the impact of radio deployments in the existing cells accurately. This work presents GRIMCELL, a novel data-driven method to estimate the impact of radio deployments (i.e., new carriers, sectors, or sites) on cell performance. For this purpose, GRIMCELL combines radio planning, configuration management and performance management data gathered in commercial vendor equipment. The core of GRIMCELL is a proven message passing graph neural network (XENet) to forecast the performance of both new and existing cells after the deployment. Assessment over data from a commercial LTE-A Pro network has shown that GRIMCELL provides high accuracy and flexibility across diverse deployment scenarios.
- 27 August 2026
- Published by MDPI AG in Machine Learning and Knowledge Extraction
- Vol. 8 (9) , 260
Open AccessGRIMCELL: A Graph Neural Network to Predict the Impact of New Cells in Mobile Networks
Network densification will play a vital role in next-generation cellular networks, as it addresses coverage and capacity issues in the radio access domain. However, new cell deployments can negatively impact neighbor cells, requiring careful planning. Unfortunately, the complexity of advanced radio resource management schemes implemented in Beyond 5G and future 6G networks makes it challenging to predict the impact of radio deployments in the existing cells accurately. This work presents GRIMCELL, a novel data-driven method to estimate the impact of radio deployments (i.e., new carriers, sectors, or sites) on cell performance. For this purpose, GRIMCELL combines radio planning, configuration management and performance management data gathered in commercial vendor equipment. The core of GRIMCELL is a proven message passing graph neural network (XENet) to forecast the performance of both new and existing cells after the deployment. Assessment over data from a commercial LTE-A Pro network has shown that GRIMCELL provides high accuracy and flexibility across diverse deployment scenarios.
- Open Access
- 3 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Vehicular Technology
- Vol. 6, 2876-2890
Remote Radio Head Clustering in 5G HetNets by Graph Partitioning
In 5G cellular systems, network densification is a key technique to cope with the strong increase of traffic volume in mobile communications. The deployment of indoor small cells offloads macrocells at the cost of increasing network complexity. In this work, a methodology for planning Centralized-Radio Access Networks (C-RANs) comprising macrocells and small cells is proposed. The aim is to group Radio Remote Heads (RRH) into Base Band Unit (BBU) pools and coordination sets (a.k.a. BBU planning) to maximize user throughput. To this end, the above assignment problem is formulated as a graph partitioning problem, which is solved by graph theory algorithms. Method assessment is carried out by using a radio planning tool that implements a novel analytical system model to check spectral efficiency and resource allocation. Different BBU planning strategies are first compared, and the impact of Inter-Cell Interference Coordination (ICIC), Coordinated Multi-Point Transmission/Reception (CoMP) and Multi-Connectivity (MC) on network performance with the best BBU plan is then assessed under different system loads and coordination constraints. Results show that the selection of a proper graph partitioning scheme for RRH clustering is key to ensure that the above schemes improve system capacity in heterogeneous environments.
- 3 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Vehicular Technology
- Vol. 6, 2876-2890
Open AccessRemote Radio Head Clustering in 5G HetNets by Graph Partitioning
In 5G cellular systems, network densification is a key technique to cope with the strong increase of traffic volume in mobile communications. The deployment of indoor small cells offloads macrocells at the cost of increasing network complexity. In this work, a methodology for planning Centralized-Radio Access Networks (C-RANs) comprising macrocells and small cells is proposed. The aim is to group Radio Remote Heads (RRH) into Base Band Unit (BBU) pools and coordination sets (a.k.a. BBU planning) to maximize user throughput. To this end, the above assignment problem is formulated as a graph partitioning problem, which is solved by graph theory algorithms. Method assessment is carried out by using a radio planning tool that implements a novel analytical system model to check spectral efficiency and resource allocation. Different BBU planning strategies are first compared, and the impact of Inter-Cell Interference Coordination (ICIC), Coordinated Multi-Point Transmission/Reception (CoMP) and Multi-Connectivity (MC) on network performance with the best BBU plan is then assessed under different system loads and coordination constraints. Results show that the selection of a proper graph partitioning scheme for RRH clustering is key to ensure that the above schemes improve system capacity in heterogeneous environments.
- Open Access
- 30 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Vehicular Technology
- Vol. 6, 396-411
Anomaly Detection in High Mobility MDT Traces Through Self-Supervised Learning
Radio access network optimization is a critical task in current cellular systems. For this purpose, Minimization of Drive Test (MDT) functionality provides mobile operators with georeferenced network performance statistics to tune radio propagation models in re-planning tools. However, some samples in MDT traces contain critical location errors due to the user equipment's energy-saving, thus making MDT data filtering vital to guarantee an adequate performance of MDT-driven algorithms. Supervised Learning (SL) allows to train automatic systems for detecting abnormal MDT measurements by using a labeled dataset. Unfortunately, labeling MDT data is a labor-intensive task, that can be alleviated by using Self-Supervised Learning (SSL). This work presents a novel SSL method to detect MDT measurements with abnormal position information in road scenarios. For this purpose, a dataset is first labeled by combining unlabeled MDT traces from high-mobility users and freely available land use maps, and then an SL classifier is trained. Model assessment is carried out using MDT data collected in a live Long-Term Evolution (LTE) network. Performance analysis includes the comparison of six well-known SL algorithms and 3 different sets of input features aiming to improve model accuracy, generalizability, and explainability, respectively. Results show that considering predictors regarding positioning error increases model accuracy, whereas omitting this information allows to cover a wider range of terminals. Likewise, Shapley Additive exPlanations (SHAP) analysis proves that the use of high-level predictors significantly improves model explainability.
- 30 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Vehicular Technology
- Vol. 6, 396-411
Open AccessAnomaly Detection in High Mobility MDT Traces Through Self-Supervised Learning
Radio access network optimization is a critical task in current cellular systems. For this purpose, Minimization of Drive Test (MDT) functionality provides mobile operators with georeferenced network performance statistics to tune radio propagation models in re-planning tools. However, some samples in MDT traces contain critical location errors due to the user equipment's energy-saving, thus making MDT data filtering vital to guarantee an adequate performance of MDT-driven algorithms. Supervised Learning (SL) allows to train automatic systems for detecting abnormal MDT measurements by using a labeled dataset. Unfortunately, labeling MDT data is a labor-intensive task, that can be alleviated by using Self-Supervised Learning (SSL). This work presents a novel SSL method to detect MDT measurements with abnormal position information in road scenarios. For this purpose, a dataset is first labeled by combining unlabeled MDT traces from high-mobility users and freely available land use maps, and then an SL classifier is trained. Model assessment is carried out using MDT data collected in a live Long-Term Evolution (LTE) network. Performance analysis includes the comparison of six well-known SL algorithms and 3 different sets of input features aiming to improve model accuracy, generalizability, and explainability, respectively. Results show that considering predictors regarding positioning error increases model accuracy, whereas omitting this information allows to cover a wider range of terminals. Likewise, Shapley Additive exPlanations (SHAP) analysis proves that the use of high-level predictors significantly improves model explainability.
- 8 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 73 (8) , 11909-11922
SLA-Driven Traffic Steering in B5G Systems With Network Slicing
In 5G and beyond wireless systems, Network Slicing (NS) feature will enable the coexistence of extremely different services by splitting the physical infrastructure into several logical slices tailored for a specific tenant or application. In sliced Radio Access Networks (RANs), an optimal traffic sharing among cells is key to guarantee Service Level Agreement (SLA) compliance while minimizing operation costs. The configuration of network functions leading to that optimal point may depend on the slice, claiming for slice-aware traffic steering strategies. This work presents the first data-driven algorithm for slice-aware traffic steering by tuning handover margins (a.k.a. mobility load balancing) per slice and adjacency with a proportional controller based on SLA criteria. The tuning process is driven by a novel indicator, derived from connection traces, showing the imbalance of SLA compliance among neighbor cells per slice. Performance assessment is carried out with a system-level simulator implementing a realistic sliced RAN offering services with different throughput, latency and reliability requirements. Results show that the proposed algorithm improves the overall SLA compliance by 9% in only 15 minutes of network activity compared to the case of not steering traffic, outperforming two legacy mobility load balancing approaches not driven by SLA.
- 8 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 73 (8) , 11909-11922
SLA-Driven Traffic Steering in B5G Systems With Network Slicing
In 5G and beyond wireless systems, Network Slicing (NS) feature will enable the coexistence of extremely different services by splitting the physical infrastructure into several logical slices tailored for a specific tenant or application. In sliced Radio Access Networks (RANs), an optimal traffic sharing among cells is key to guarantee Service Level Agreement (SLA) compliance while minimizing operation costs. The configuration of network functions leading to that optimal point may depend on the slice, claiming for slice-aware traffic steering strategies. This work presents the first data-driven algorithm for slice-aware traffic steering by tuning handover margins (a.k.a. mobility load balancing) per slice and adjacency with a proportional controller based on SLA criteria. The tuning process is driven by a novel indicator, derived from connection traces, showing the imbalance of SLA compliance among neighbor cells per slice. Performance assessment is carried out with a system-level simulator implementing a realistic sliced RAN offering services with different throughput, latency and reliability requirements. Results show that the proposed algorithm improves the overall SLA compliance by 9% in only 15 minutes of network activity compared to the case of not steering traffic, outperforming two legacy mobility load balancing approaches not driven by SLA.
- 21 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network and Service Management
- Vol. 20 (2) , 1008-1023
Data-Driven Estimation of Throughput Performance in Sliced Radio Access Networks via Supervised Learning
In 5G systems, Network Slicing (NS) feature allows to deploy several logical networks customized for specific verticals over a common physical infrastructure. To make the most of this feature, cellular operators need models reflecting cell and slice performance for re-dimensioning the Radio Access Network (RAN). For enhanced Mobility BroadBand (eMBB) services, throughput is regarded as a key performance metric since it strongly influences user experience. This work presents the first comprehensive analysis tackling cell and slice throughput estimation in the downlink of RAN-sliced networks through Supervised Learning (SL), based on information collected in the operations support system. Different well-known SL algorithms are tested in two NS scenarios with single-service or multi-service slices serving eMBB users. To this end, several synthetic datasets are generated with a system-level simulator emulating the activity of a sliced RAN. Results show that NS alters the correlation between network performance indicators and cell throughput compared to legacy RANs, thus being required a separate analysis for NS scenarios. Moreover, the best model to estimate throughput at cell/slice level may depend on the scenario (single-service vs multi-service slices). In all cases, the best models have shown an estimation error below 10 %.
- 21 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network and Service Management
- Vol. 20 (2) , 1008-1023
Data-Driven Estimation of Throughput Performance in Sliced Radio Access Networks via Supervised Learning
In 5G systems, Network Slicing (NS) feature allows to deploy several logical networks customized for specific verticals over a common physical infrastructure. To make the most of this feature, cellular operators need models reflecting cell and slice performance for re-dimensioning the Radio Access Network (RAN). For enhanced Mobility BroadBand (eMBB) services, throughput is regarded as a key performance metric since it strongly influences user experience. This work presents the first comprehensive analysis tackling cell and slice throughput estimation in the downlink of RAN-sliced networks through Supervised Learning (SL), based on information collected in the operations support system. Different well-known SL algorithms are tested in two NS scenarios with single-service or multi-service slices serving eMBB users. To this end, several synthetic datasets are generated with a system-level simulator emulating the activity of a sliced RAN. Results show that NS alters the correlation between network performance indicators and cell throughput compared to legacy RANs, thus being required a separate analysis for NS scenarios. Moreover, the best model to estimate throughput at cell/slice level may depend on the scenario (single-service vs multi-service slices). In all cases, the best models have shown an estimation error below 10 %.
- Open Access
- 3 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 123245-123259
The Upstream Matters: Impact of Uplink Performance on YouTube 360° Live Video Streaming in LTE
Live video streaming services are gaining momentum as network and terminal capabilities improve. However, 360° live video streaming services pose new challenges due to its high bandwidth and computational requirements both on the user and service provider. In this paper, a study of the impact of the uplink of a cellular network on the performance of 360° live video streaming in YouTube is presented. Unlike previous works, the analysis focuses on the upstream between the video source and the server, not on the downstream between the server and viewers. To this end, a measurement campaign is conducted where a live video feed is transmitted and received through YouTube 360° platform in a pilot Long Term Evolution (LTE) system. During the tests, a large dataset of real traces is collected at different protocol layers, both in upstream and downstream, to check the correlation between TCP/IP metrics and key service performance indicators (e.g., video segment quality and end-to-end latency). Results show that uplink performance has a strong impact on the latency perceived by the user, which is critical for the considered live services.
- 3 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 123245-123259
Open AccessThe Upstream Matters: Impact of Uplink Performance on YouTube 360° Live Video Streaming in LTE
Live video streaming services are gaining momentum as network and terminal capabilities improve. However, 360° live video streaming services pose new challenges due to its high bandwidth and computational requirements both on the user and service provider. In this paper, a study of the impact of the uplink of a cellular network on the performance of 360° live video streaming in YouTube is presented. Unlike previous works, the analysis focuses on the upstream between the video source and the server, not on the downstream between the server and viewers. To this end, a measurement campaign is conducted where a live video feed is transmitted and received through YouTube 360° platform in a pilot Long Term Evolution (LTE) system. During the tests, a large dataset of real traces is collected at different protocol layers, both in upstream and downstream, to check the correlation between TCP/IP metrics and key service performance indicators (e.g., video segment quality and end-to-end latency). Results show that uplink performance has a strong impact on the latency perceived by the user, which is critical for the considered live services.
- Open Access
- 26 July 2021
- Published by MDPI AG in Electronics
- Vol. 10 (15) , 1782
Statistical Model for Mobile User Positioning Based on Social Information
In spite of the vast set of measurements provided by current mobile networks, cellular operators have problems pinpointing problematic locations because the origin of such measurements (i.e., user location) is usually not registered. At the same time, social networks generate a huge amount of data that can be used to infer population density. In this work, a data-driven model is proposed to deduce the statistical distribution of connections, exploiting the knowledge of network layout and population density in the scenario. Due to the absence of GPS measurements, the proposed method combines data from radio connection traces stored in the network management system and geolocated posts from social networks. This information is enriched with user context information inferred from data traffic attributes. The method is tested with a large trace dataset from a live Long Term Evolution (LTE) network and a database of geotagged posts from social networks collected in real-time.
- 26 July 2021
- Published by MDPI AG in Electronics
- Vol. 10 (15) , 1782
Open AccessStatistical Model for Mobile User Positioning Based on Social Information
In spite of the vast set of measurements provided by current mobile networks, cellular operators have problems pinpointing problematic locations because the origin of such measurements (i.e., user location) is usually not registered. At the same time, social networks generate a huge amount of data that can be used to infer population density. In this work, a data-driven model is proposed to deduce the statistical distribution of connections, exploiting the knowledge of network layout and population density in the scenario. Due to the absence of GPS measurements, the proposed method combines data from radio connection traces stored in the network management system and geolocated posts from social networks. This information is enriched with user context information inferred from data traffic attributes. The method is tested with a large trace dataset from a live Long Term Evolution (LTE) network and a database of geotagged posts from social networks collected in real-time.
- Open Access
- 12 May 2021
- Published by MDPI AG in Electronics
- Vol. 10 (10) , 1151
Long-Term Data Traffic Forecasting for Network Dimensioning in LTE with Short Time Series
Network dimensioning is a critical task in current mobile networks, as any failure in this process leads to degraded user experience or unnecessary upgrades of network resources. For this purpose, radio planning tools often predict monthly busy-hour data traffic to detect capacity bottlenecks in advance. Supervised Learning (SL) arises as a promising solution to improve predictions obtained with legacy approaches. Previous works have shown that deep learning outperforms classical time series analysis when predicting data traffic in cellular networks in the short term (seconds/minutes) and medium term (hours/days) from long historical data series. However, long-term forecasting (several months horizon) performed in radio planning tools relies on short and noisy time series, thus requiring a separate analysis. In this work, we present the first study comparing SL and time series analysis approaches to predict monthly busy-hour data traffic on a cell basis in a live LTE network. To this end, an extensive dataset is collected, comprising data traffic per cell for a whole country during 30 months. The considered methods include Random Forest, different Neural Networks, Support Vector Regression, Seasonal Auto Regressive Integrated Moving Average and Additive Holt–Winters. Results show that SL models outperform time series approaches, while reducing data storage capacity requirements. More importantly, unlike in short-term and medium-term traffic forecasting, non-deep SL approaches are competitive with deep learning while being more computationally efficient.
- 12 May 2021
- Published by MDPI AG in Electronics
- Vol. 10 (10) , 1151
Open AccessLong-Term Data Traffic Forecasting for Network Dimensioning in LTE with Short Time Series
Network dimensioning is a critical task in current mobile networks, as any failure in this process leads to degraded user experience or unnecessary upgrades of network resources. For this purpose, radio planning tools often predict monthly busy-hour data traffic to detect capacity bottlenecks in advance. Supervised Learning (SL) arises as a promising solution to improve predictions obtained with legacy approaches. Previous works have shown that deep learning outperforms classical time series analysis when predicting data traffic in cellular networks in the short term (seconds/minutes) and medium term (hours/days) from long historical data series. However, long-term forecasting (several months horizon) performed in radio planning tools relies on short and noisy time series, thus requiring a separate analysis. In this work, we present the first study comparing SL and time series analysis approaches to predict monthly busy-hour data traffic on a cell basis in a live LTE network. To this end, an extensive dataset is collected, comprising data traffic per cell for a whole country during 30 months. The considered methods include Random Forest, different Neural Networks, Support Vector Regression, Seasonal Auto Regressive Integrated Moving Average and Additive Holt–Winters. Results show that SL models outperform time series approaches, while reducing data storage capacity requirements. More importantly, unlike in short-term and medium-term traffic forecasting, non-deep SL approaches are competitive with deep learning while being more computationally efficient.
- Open Access
- 6 May 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 71673-71686
A Deep-Learning Model for Estimating the Impact of Social Events on Traffic Demand on a Cell Basis
In cellular networks, a deep knowledge of the traffic demand pattern in each cell is essential in network planning and optimization tasks. However, a precise forecast of the traffic time series per cell is hard to achieve, due to the noise originated by abnormal local events. In particular, mass social events (e.g., concerts, conventions, sport events⋯) have a strong impact on traffic demand. In this paper, a data-driven model to estimate the impact of local events on cellular traffic is presented. The model is trained with a large dataset of geotagged social events taken from public event databases and hourly traffic data from a live Long Term Evolution (LTE) network. The resulting model is combined with a traffic forecast module based on a multi-task deep-learning architecture to predict the hourly traffic series with scheduled mass events. Model assessment is performed over a real dataset created with geolocated social event information collected from public event directories and hourly cell traffic measurements during two months in a LTE network. Results show that the addition of the proposed model significantly improves traffic forecasts in the presence of massive events.
- 6 May 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 71673-71686
Open AccessA Deep-Learning Model for Estimating the Impact of Social Events on Traffic Demand on a Cell Basis
In cellular networks, a deep knowledge of the traffic demand pattern in each cell is essential in network planning and optimization tasks. However, a precise forecast of the traffic time series per cell is hard to achieve, due to the noise originated by abnormal local events. In particular, mass social events (e.g., concerts, conventions, sport events⋯) have a strong impact on traffic demand. In this paper, a data-driven model to estimate the impact of local events on cellular traffic is presented. The model is trained with a large dataset of geotagged social events taken from public event databases and hourly traffic data from a live Long Term Evolution (LTE) network. The resulting model is combined with a traffic forecast module based on a multi-task deep-learning architecture to predict the hourly traffic series with scheduled mass events. Model assessment is performed over a real dataset created with geolocated social event information collected from public event directories and hourly cell traffic measurements during two months in a LTE network. Results show that the addition of the proposed model significantly improves traffic forecasts in the presence of massive events.
- 15 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network and Service Management
- Vol. 18 (3) , 3541-3555
A Service-Centric Q-Learning Algorithm for Mobility Robustness Optimization in LTE
Due to the diversity of mobile services and rising user expectations, mobile network management has changed its focus from Quality of Service (QoS) to Quality of Experience (QoE). As a consequence, classical network optimization procedures must be updated accordingly. One of these optimization procedures is Mobility Robustness Optimization (MRO), whose aim is to improve HandOver (HO) performance by reducing HO failures. In this work, a novel QoE-aware MRO algorithm is proposed considering a multi-service scenario. Unlike previous approaches, whose aim is to increase successful handover rates, the optimization aim in this work is two-folded: to improve cell edge QoE while improving successful handover rates in the whole network. For this purpose, the handover trigger point, defined by the pair of HO control parameters HO margin and Time to Trigger, are tuned on a per-adjacency basis according to QoE and HO failure measurements. Method assessment is based on a dynamic system-level simulator implementing a realistic LTE scenario with multiple services. Results show that the proposed QoE-aware MRO algorithm improves cell edge QoE throughout the network while increasing the percentage of successful handovers compared to traditional approaches.
- 15 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network and Service Management
- Vol. 18 (3) , 3541-3555
A Service-Centric Q-Learning Algorithm for Mobility Robustness Optimization in LTE
Due to the diversity of mobile services and rising user expectations, mobile network management has changed its focus from Quality of Service (QoS) to Quality of Experience (QoE). As a consequence, classical network optimization procedures must be updated accordingly. One of these optimization procedures is Mobility Robustness Optimization (MRO), whose aim is to improve HandOver (HO) performance by reducing HO failures. In this work, a novel QoE-aware MRO algorithm is proposed considering a multi-service scenario. Unlike previous approaches, whose aim is to increase successful handover rates, the optimization aim in this work is two-folded: to improve cell edge QoE while improving successful handover rates in the whole network. For this purpose, the handover trigger point, defined by the pair of HO control parameters HO margin and Time to Trigger, are tuned on a per-adjacency basis according to QoE and HO failure measurements. Method assessment is based on a dynamic system-level simulator implementing a realistic LTE scenario with multiple services. Results show that the proposed QoE-aware MRO algorithm improves cell edge QoE throughout the network while increasing the percentage of successful handovers compared to traditional approaches.