Istvan Z. Kovacs’s research while affiliated with Nokia and other places

div>Uplink power control plays a significant role in maintaining a good signal quality at the serving cell while minimizing interference to neighboring cells, thus maximizing the system performance. Traditionally, a single value open-loop power control (OLPC) parameter, P0, is configured for all the user equipments (UEs) in a cell, and often same setting is used for similar cells. Recent studies have demonstrated that optimal P0 depends on many factors, which yields a complex multidimensional optimization problem and there are no efficient approaches known to solve it under practical system-level settings. In this paper, we propose a solution based on reinforcement learning (RL) where each BS autonomously adjusts its P0 setting to maximize its throughput performance. As compared to conventional sub-optimal approach, our solution encompasses a smart clustering of UEs, where each cluster specifies its own P0. The proposed solution is evaluated by extensive system level simulations, where our results demonstrate a potential performance enhancement as compared to the baseline proposals.</div

5G communication systems are one of the major enabling technologies to meet the needs of Industry 4.0. This paper focuses on the use case of automated guided vehicles (AGVs) in an outdoor industrial scenario. To meet the communication requirements in these type of use cases, dual connectivity (DC) with resource aggregation in the uplink (UL) is generally proposed. However, uncontrolled use of DC schemes may negatively affect the network causing effects such as reduced network capacity, increased signaling, and increased interference. To overcome these issues, this paper proposes and evaluates the use of a proactive DC activation algorithm based on the instantaneous quality of service (QoS) and network conditions. The proposed algorithm has two phases, a first phase in which the QoS prediction is performed, and a second phase in which the DC activation decision is made. The performance evaluation of the algorithm has been carried out in two different scenarios: a single-frequency (SF) network and a dual-frequency (DF) network; and compared to two baselines. Our results show that our predictive DC algorithm is sufficiently robust and can offer benefits in terms of reduced signaling and increased UL performance, especially in scenarios with low to medium traffic load.

An accurate estimation of the service quality that the user will experience along a route can be extremely useful for mission-critical services. It can provide the network with in-advance information on the potential critical areas along the route based on availability and reliability estimations. If such estimation is based on empirical/statistical or site-specific estimations, both of which are typically used for cellular network planning, it will lead to significant uncertainty in the estimation as we demonstrate in this paper. Instead, if estimations are based on previously collected measurements, the uncertainty can be significantly reduced. In this paper, we analyze the achievable accuracy of such a data-driven estimation which aggregates measurements from multiple user equipment (UEs) moving along the same route by averaging the measured signal levels over a route segment. We evaluate the estimation error for both empirical/statistical, site-specific and data-driven estimations for measurements collected in urban areas. Based on the demonstrated advantage of data-driven estimation, and the relevance of including context information that we proved in a previous paper, we discuss and analyze how the estimation error can be reduced even further by predicting the Mean Individual Offset (MIO) that each specific UE will observe with respect to the average. To this end, we propose and evaluate a technique for MIO correction that relies on observing a time series of signal level samples when the UE starts a mission-critical service. By observing 100-300 m of real-time samples along the route results show that the overall estimation error can be reduced from 5-6 dB to 4 dB using MIO correction. Finally, using the obtained results, we illustrate how the signal level estimations can be used to estimate the outage probability along the planned route.

Artificial intelligence and data-driven networks will be integral part of 6G systems. In this article, we comprehensively discuss implementation challenges and need for architectural changes in 5G radio access networks for integrating machine learning solutions. As an example use case, we investigate user equipment positioning assisted by deep learning (DL) in 5G and beyond networks. As compared to state of the art positioning algorithms used in today's networks, radio signal fingerprinting and machine learning (ML) assisted positioning requires smaller additional feedback overhead; and the positioning estimates are made directly inside the radio access network (RAN), thereby assisting in radio resource management. In this regard, we study ML-assisted positioning methods and evaluate their performance using system level simulations for an outdoor scenario. The study is based on the use of raytracing tool, a 3GPP 5G NR compliant system level simulator and DL framework to estimate positioning accuracy of the UE. We evaluate and compare performance of various DL models and show mean positioning error in the range of 1-1.5m for a 2-hidden layer DL architecture with appropriate feature-modeling. Building on our performance analysis, we discuss pros and cons of various architectures to implement ML solutions for future networks and draw conclusions on the most suitable architecture.

Artificial intelligence and data-driven networks will be integral part of 6G systems. In this article, we comprehensively discuss implementation challenges and need for architectural changes in 5G radio access networks for integrating machine learning (ML) solutions. As an example use case, we investigate user equipment (UE) positioning assisted by deep learning (DL) in 5G and beyond networks. As compared to state of the art positioning algorithms used in today's networks, radio signal fingerprinting and machine learning (ML) assisted positioning requires smaller additional feedback overhead; and the positioning estimates are made directly inside the radio access network (RAN), thereby assisting in radio resource management. In this regard, we study ML-assisted positioning methods and evaluate their performance using system level simulations for an outdoor scenario. The study is based on the use of raytracing tool, a 3GPP 5G NR compliant system level simulator and DL framework to estimate positioning accuracy of the UE. We evaluate and compare performance of various DL models and show mean positioning error in the range of 1-1.5m for a 2-hidden layer DL architecture with appropriate feature-modeling. Building on our performance analysis, we discuss pros and cons of various architectures to implement ML solutions for future networks and draw conclusions on the most suitable architecture.

Cellular connected unmanned aerial vehicle (UAV) has been identified as a promising paradigm and attracted a surge of research interest recently. Although the nearly line-of-sight (LoS) channels are favorable to receive higher powers, UAV can in turn cause severe interference to each other and to any other users in the same frequency band. In this contribution, we focus on the uplink communications of cellular-connected UAV. To cope with the severe interference among UAV-UEs, several different scheduling and power control algorithms are proposed to optimize the spectrum efficiency (SE) based on the geometrical programming (GP) principle together with the successive convex approximation (SCA) technique. The proposed schemes include maximizing the sum SE of UAVs, maximizing the minimum SE of UAVs, etc., applied in the frequency domain and/or the time domain. Moreover, the quality of service (QoS) constraint and the uplink single-carrier (SC) constraint are also considered. The performances of these power and resource allocation algorithms are evaluated via extensive simulations in both full buffer transmission mode and bursty traffic mode. Numerical results show that the proposed algorithms can effectively enhance the uplink SEs of cellular-connected UAVs.

Power control is becoming increasingly essential for the fifth-generation (5G) and beyond systems. An example use-case, among others, is the unmanned-aerial-vehicle (UAV) communications where the nearly line-of-sight (LoS) radio channels may result in very low signal-to-interference-plus-noise ratios (SINRs). Investigations in [1] proposed to efficiently and reliably solve this kind of non-convex problem via a series of geometrical programmings (GPs) using condensation approximation. However, it is only applicable for a small-scale network with several communication pairs and practically infeasible with more (e.g. tens of) nodes to be jointly optimized. We therefore in this paper aim to provide new insights into this problem. By properly introducing auxiliary variables, the problem is transformed to an equivalent form which is simpler and more intuitive for condensation. A novel condensation method with linear complexity is also proposed based on the form. The enhancements make the GP-based power control feasible for both small-and especially large-scale networks that are common in 5G and beyond. The algorithm is verified via simulations. A preliminary case study of uplink UAV communications also shows the potential of the algorithm.

Cellular-connected unmanned aerial vehicles (UAVs) have recently attracted a surge of interest in both academia and industry. Understanding the air-to-ground (A2G) propagation channels is essential to enable reliable and/or high-throughput communications for UAVs and protect the ground user equipments (UEs). In this contribution, a recently conducted measurement campaign for the A2G channels is introduced. A uniform circular array (UCA) with 16 antenna elements was employed to collect the downlink signals of two different Long Term Evolution (LTE) networks, at the heights of 0-40 m in three different, namely rural, urban and industrial scenarios. The channel impulse responses (CIRs) have been extracted from the received data, and the spatial, including angular, parameters of the multipath components in individual channels were estimated according to a high-resolution-parameter-estimation (HRPE) principle. Based on the HRPE results, clusters of multi-path components were further identified. Finally, comprehensive spatial channel characteristics were investigated in the composite and cluster levels at different heights in the three scenarios.

This chapter explains that existing long‐term evolution networks and future 5G networks are able to ensure reliable command and control (C2) communication to drones and play an important role in the provisioning of end‐to‐end reliability for drone communications, similar to other use cases requiring high reliability. In unmanned aerial vehicles (UAVs) scenarios, especially for C2 link performance in cellular networks, it is extremely valuable to determine how the cellular radio modem on the UAV would perceive the radio channel in the existing cellular networks. The chapter focuses on the C2 link and the possibility of providing reliable communication over cellular networks, while also having a look at the ability of cellular networks to provide high uplink throughputs. To understand how existing cellular networks work when users are in the air, the chapter discusses the propagation characteristics for drones.