5G Networks Supported by UAVs, RESs, and RISs

This paper presents the examination of the 5G cellular network aware of Renewable Energy Sources (RESs) and supported by Reconfigurable Intelligent Surfaces (RISs) and Unmanned Aerial Vehicles working as mobile access nodes. The investigations have been focused on the energy side of the Radio Access Network (RAN) placed within the area of the city of Poznan (Poland). The gain related to enabling RES generators, i.e., photovoltaic (PV) panels, for base stations (BSs) was presented in the form of two factors -- the average number of UAV replacements (ANUR) with a fully charged one to ensure continuous access to mobile services for currently served user equipment (UE) terminals, and the average reduction in energy consumption (AREC) within the whole network.

Wykorzystanie rekonfigurowalnych matryc antenowych wraz z informacją kontekstową

Reconfigurable intelligent surfaces can be successfully used to control the radio environment. Simple control of the reflection angle of the signal from the surface allows maximization or minimization of the received power in specific places. The paper presents simulations where it is possible to receive a signal in a place where it was not possible, to detect the occupancy of the spectrum in a place where the sensor was unable to make correct detection or to minimize interference in a specific receiver.

Machine Learning-based xApp for Dynamic Resource Allocation in O-RAN Networks

The disaggregated, distributed and virtualised implementation of radio access networks allows for dynamic resource allocation. These attributes can be realised by virtue of the Open Radio Access Networks (O-RAN) architecture. In this article, we tackle the issue of dynamic resource allocation using a data-driven approach by employing Machine Learning (ML). We present an xApp-based implementation for the proposed ML algorithm. The core aim of this work is to optimise resource allocation and fulfil Service Level Specifications (SLS). This is accomplished by dynamically adjusting the allocation of Physical Resource Blocks (PRBs) based on traffic demand and Quality of Service (QoS) requirements. The proposed ML model effectively selects the best allocation policy for each base station and enhances the performance of scheduler functionality in O-RAN - Distributed Unit (O-DU). We show that an xApp implementing the Random Forest Classifier can yield high (85\%) performance accuracy for optimal policy selection. This can be attained using the O-RAN instance state input parameters over a short training duration.

Federated Learning-Based Interference Modeling for Vehicular Dynamic Spectrum Access

A platoon-based driving is a technology allowing vehicles to follow each other at close distances to, e.g., save fuel. However, it requires reliable wireless communications to adjust their speeds. Recent studies have shown that the frequency band dedicated for vehicle-to-vehicle communications can be too busy for intra-platoon communications. Thus it is reasonable to use additional spectrum resources, of low occupancy, i.e., secondary spectrum channels. The challenge is to model the interference in those channels to enable proper channel selection. In this paper, we propose a two-layered Radio Environment Map (REM) that aims at providing platoons with accurate location-dependent interference models by using the Federated Learning approach. Each platoon is equipped with a Local REM that is updated on the basis of raw interference samples and previous interference model stored in the Global REM. The model in global REM is obtained by merging models reported by platoons. The nodes exchange only parameters of interference models, reducing the required control channel capacity. Moreover, in the proposed architecture platoon can utilize Local REM to predict channel occupancy, even when the connection to the Global REM is temporarily unavailable. The proposed system is validated via computer simulations considering non-trivial interference patterns.

Neural Networks for Path Planning

The scientific community is able to present a new set of solutions to practical problems that substantially improve the performance of modern technology in terms of efficiency and speed of computation due to the advancement in neural networks architectures. We present the latest works considering the utilization of neural networks in robot path planning. Our survey shows the contrast between different formulations of the problems that consider different inputs, outputs, and environments and how different neural networks architectures are able to provide solutions to all of the presented problems.

Energy-efficient User Clustering for UAV-enabled Wireless Networks Using EM Algorithm

Unmanned Aerial Vehicles (UAVs) can be used to provide wireless connectivity to support the existing infrastructure in hot-spots or replace it in cases of destruction. UAV-enabled wireless provides several advantages in network performance due to drone small cells (DSCs) mobility despite the limited onboard energy. However, the problem of resource allocation has added complexity. In this paper, we propose an energy-efficient user clustering mechanism based on Gaussian mixture models (GMM) using a modified Expected-Maximization (EM) algorithm. The algorithm is intended to provide the initial user clustering and drone deployment upon which additional mechanisms can be employed to further enhance the system performance. The proposed algorithm improves the energy efficiency of the system by 25% and link reliability by 18.3% compared to other baseline methods.

On the Placement and Sustainability of Drone FSO Backhaul Relays

We consider free-space optical (FSO) communication links for the backhaul connectivity of small cells (SCs) where a UAV with an FSO apparatus can serve as a backhaul relay node. We demonstrate how such drone relay stations (DRSs) can be deployed in a high-rise urban area in order to provide FSO line-of-sight (LOS) links that are unobstructed by buildings. Also, in our solution we consider the case where solar panels are mounted on DRSs such that placing the DRS in a sunny location is prioritized, and we show the gain in terms of number of required trips to recharge the UAV.

Distributed Vehicular Dynamic Spectrum Access for Platooning Environments

In this paper, we propose a distributed Vehicular Dynamic Spectrum Access (VDSA) framework for vehicles operating in platoon formations. Given the potential for significant congestion in licensed frequency bands for vehicular applications such as 5.9 GHz. Our approach proposes to offload part of the intra-platoon data traffic to spectral white-spaces in order to enhance vehicular connectivity in support of on-road operations. To enable VDSA, a Bumblebee-based decision making process is employed which is based on the behavioral models of animals, is employed to provide a means of distributed transmission band selection. Simulation results show the distributed VDSA framework improves the leader packets reception ratio by 5%, thus indicating its potential to increase in reliability of intra-platoon communications.

Frequency Selection for Platoon Communications in Secondary Spectrum Using Radio Environment Maps

Platoon-based driving is an idea that vehicles follow each other at a close distance, in order to increase road throughput and fuel savings. This requires reliable wireless communications to adjust the speeds of vehicles. Although there is a dedicated frequency band for vehicle-to-vehicle (V2V) communications, studies have shown that it is too congested to provide reliable transmission for the platoons. Additional spectrum resources, i.e., secondary spectrum channels, can be utilized when these are not occupied by other users. Characteristics of interference in these channels are usually location-dependent and can be stored in the so-called Radio Environment Maps (REMs). This paper aims to design REM, in order to support the selection of secondary spectrum channel for intra-platoon communications. We propose to assess the channel's quality in terms of outage probability computed, with the use of estimated interference distributions stored in REM. A frequency selection algorithm that minimizes the number of channel switches along the planned platoon route is proposed. Additionally, the REM creation procedure is shown that reduces the number of database entries using (Density-Based Spatial Clustering of Applications with Noise) DBSCAN algorithm. The proposals are tested using real IQ samples captured on a real road. Application of the DBSCAN clustering to the constructed REM provided 7% reduction in its size. Utilization of the proposed channel selection algorithm resulted in a 35 times reduction of channel switches concerning channel assignment performed independently in every location.

Reinforcement Learning Algorithm for Traffic Steering in Heterogeneous Network

Heterogeneous radio access networks require efficient traffic steering methods to reach near-optimal results in order to maximize network capacity. This paper aims to propose a novel traffic steering algorithm for usage in HetNets, which utilizes a reinforcement learning algorithm in combination with an artificial neural network to maximize total user satisfaction in the simulated cellular network. The novel algorithm was compared with two reference algorithms using network simulation results. The results prove that the novel algorithm provides noticeably better efficiency in comparison with reference algorithms, especially in terms of the number of served users with limited frequency resources of the radio access network.