Several upcoming 5G and 6G services will rely on unmanned aerial vehicles (UAV) sending live information to remote terminals. Thus, understanding the traffic flows that might influence end-user experience in these services is key for cellular network operators. One of these UAV-based services is first person view (FPV) drone control, consisting on the remote control of the UAV in Beyond Visual Line of Sight scenarios using only the live video visualized in a ground control station. This work focuses on the networking aspects of this service by presenting the assembly, integration and evaluation methodology of an UAV quadrotor teleoperated via FPV through a Long Term Evolution (LTE) network or WiFi radio access link. To assess system performance, three different connectivity schemes between UAV and ground control station are tested, namely server-based connection via LTE, direct LTE, and peer-to-peer WiFi connection. Then, several experiments are carried out in the testbed to characterize telemetry, control and video traffic for FPV service in the above schemes. Later, a methodology is defined to estimate Quality of Experience (QoE) for FPV service based on image quality and video latency measurements collected at network and application level. Results show that the QoE model for live video introduced in this work can be the basis of more sophisticated models for cellular FPV services.

Cellular networks are constantly evolving, driven by changes in user behavior and device capabilities. To ensure that networks adapt to these changes, it is of vital importance for mobile operators to have a good understanding of how well their network meets subscriber needs. For this purpose, the Minimization of Drive Test (MDT) feature has been standardized, allowing operators the cost-effective provision of geolocated network performance statistics and radio events. However, in practice, positioning errors severely limit the potential of MDT measurements. In this paper, an in-depth analysis of a large MDT dataset taken from a commercial Long-Term Evolution (LTE) network shows for the first time several sources of positioning errors in MDT measurements not previously reported in the literature. To address these, a novel heuristic filtering algorithm is proposed to discard samples with inaccurate location data. Method assessment is done by checking the impact of filtering on the coverage map built with a real MDT dataset. Results show that the proposed filtering method significantly improves the accuracy of coverage maps by eliminating unreliable measurements.

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 %.

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.

Network dimensioning is a critical task for cellular operators to avoid degraded user experience and unnecessary upgrades of network resources with changing mobile traffic patterns. For this purpose, smart network planning tools require accurate cell and user capacity estimates. In these tools, throughput is often used as a capacity metric due to its close relationship with user satisfaction. In this work, a comprehensive analysis is carried out to compare different well-known Supervised Learning (SL) algorithms for estimating cell and user throughput in the DownLink in busy hours from radio measurements collected on a cell basis in the Operation Support System (OSS). The considered SL approaches include random forest, shallow multi-layer perceptron, support vector regression and k-nearest neighbors. Such algorithms are compared with classical multiple linear regression and deep learning approaches considered in previous works. All these algorithms are tested in two radio access technologies: High Speed DownLink Packet Access (HSDPA) and Long Term Evolution (LTE). To this end, two datasets with the most relevant performance indicators per technology are collected from live cellular networks. Results show that non-deep SL algorithms are the most appropriate option for applications with storage constraints, such as network planning tools, since they provide a higher accuracy with reduced datasets.

An accurate estimation of key performance indicators (KPI) in mobile communication networks is an important issue, especially for the planning stage or optimization purposes. Traditional approaches use simulation tools, including thorough models. However, computational costs strongly increase with network complexity and number of cells. In this paper, several low- complexity and scalable estimation approaches for cell throughput indicators are presented, for both downlink and uplink. Simplifications are focused on geometrical and propagation aspects. Results show how the estimation of data bit rates is accurate enough, specially for uplink, and calculation time is reduced by up to ten times compared to classical estimation approaches.

Mobility robustness optimization has been identified as an important use case of Self-Organizing Network (SON). In this paper, a self-optimization algorithm for tuning handover (HO) parameters of a Long Term Evolution (LTE) femtocell network in an office scenario is presented. The algorithm is implemented by a fuzzy logic controller, which jointly tunes HO margin and Time-to-Trigger parameters. The aim of the algorithm is to improve the overall handover performance, given by the average number of HOs per call and the call dropping ratio. Simulation results show that, unlike existing algorithms, the proposed algorithm improves network performance for any situation of average network load.

In cellular networks, traffic demand is unevenly distributed both in time and space. This paper investigates the problem of re-distributing traffic demand between Long-Term Evolution (LTE) femtocells in an enterprise scenario. Several traffic sharing algorithms based on automatic tuning of femtocell parameters are considered. The proposed algorithms are implemented by fuzzy logic controllers. Performance assessment is carried out in a dynamic system-level simulator. Results show that tuning handover margins and transmit power can be an effective means to solve localized congestion problems in these scenarios.

In Self-Organizing Networks (SON), load balancing has been recognized as an effective means to increase network performance. In cellular networks, cell load balancing can be achieved by tuning handover parameters, for which a Fuzzy Logic Controller (FLC) usually provides good performance and usability. Operator experience can be used to define the behavior of the FLCs. However, such a knowledge is not always available and hence optimization techniques must be applied in the controller design. In this work, a fuzzy $Q$-Learning algorithm is proposed to find the optimal set of fuzzy rules in an FLC for traffic balancing in GSM-EDGE Radio Access Network (GERAN). Load balancing is performed by modifying handover margins. Simulation results show that the optimized FLC provides a significant reduction in call blocking.

Hierarchical network structuring will be a central issue in future cellular networks. In the literature, many graph partitioning methods have been proposed to assign network elements to controllers to minimize network signaling. This paper presents two refinements to such methods, namely the connectedness and site constraints, to improve the spatial consistency of solutions, which is key for easy checking by the operator. Assessment is based on real problem instances of the assignment of base stations to packet control units in GSM-EDGE Radio Access Network (GERAN). Results show that fragmentation and overlapping in the final solution can be greatly reduced, while degrading the network performance only marginally.

Mobile network operators often use handover statistics to improve the structure of their networks. However, such statistics only reflect the movement of connected users. In this paper, a thorough investigation of the correlation between idle and connected user mobility statistics is performed based on measurements from a live GSM network. A multiple regression model is proposed to estimate the number of location updates due to mobility and non-mobility reasons on a per-cell basis from measurements in the network management system. Results shown that, although the number of location updates can normally be predicted accurately from handover statistics, especially when aggregated over large geographical areas, large deviations are observed in cells of lower layers in multi-tier networks.

In this paper, an auto-tuning scheme based on a fuzzy logic controller (FLC) is proposed for a standard inter-system handover (IS-HO) algorithm. A heterogeneous network scenario is considered, comprising GSM and UMTS radio access technologies. FLC modifies IS-HO parameters to perform load sharing between technologies based on network congestion statistics. To validate the proposed scheme, a system-level simulator with a joint radio resource management (JRRM) module has been developed. FLC adaptation capabilities are checked through changes in the time-space traffic distribution in a realistic scenario. Results show that call blocking rates can be significantly reduced, while keeping connection quality almost unaltered, at the expense of increasing network signaling load.

The flexibility emerging from the large set of radio network parameters defined on a cell basis is not currently fully utilized because of the complexity related to their tuning processes. An automatic optimization algorithm for signal level thresholds is proposed, which is based on statistical information derived from collected measurement reports on a cell basis. Simulation results and real data are analyzed to reinforce the conclusions.