Research Objectives: The objective of this ESR is to study the feasibility of cellular-based communication for UAVs. Accurate path loss and channel models will be developed for the wireless communications between UAVs and ground stations. The characterization of the antenna and propagation is vital in order to take advantage of the full potential of this technology and to achieve reliable link performance and spectrum efficiency especially for environments with huge traffic demands.

Host Institution: University of Huddersfield, United Kingdom

Research Objectives: The goal is to utilize mm-Wave signals for simultaneous mapping and localization. When the bandwidths become larger, one needs to consider the effect of the delays of different paths on the mm-wave channel model. Therefore, the determination of the angle of arrival (AOA) and angle of departure (AOD) become challenging and new algorithms for AOA and AOD will need to be designed. These algorithms will be integrated without a priori knowledge of the geometry and material properties of the surroundings by using AI. 

Host Institution: University of Huddersfield, United Kingdom

Research Objectives: The main objective is to design deep learning techniques in LSA networks to make key decisions and to achieve efficient and smart spectrum coexistence between incumbent users (IUs) and licensee users (LUs) under the supervision of the regulator with rules and conditions to guarantee predictable quality of service (QoS) levels. Deep learning will be utilised to alleviate the impact of imperfect spectrum sensing.

Host Institution: University of Huddersfield, United Kingdom

Research Objectives: The key challenge is to employ network slicing, i.e. divide spectral resources flexibly and dynamically depending on application specific requirements. DSA when combined with network slicing in a distributed manner presents a completely new challenge. Such a scenario is of relevance especially considering the sharing of spectral resources across different operators flexibly, thus enabling spectrum as a service type of business model. This naturally complements the carrier aggregation capabilities of FWNs.

Host Institution: University of Huddersfield, United Kingdom

Research Objectives: The main objective is to develop adaptive BF algorithms, which can be applied on realistic antenna arrays. Real life effects, such as the non-isotropic radiation of the array elements and the mutual coupling between them, will be taken into account. To achieve instant response and high accuracy, these techniques will be implemented by using AI, which means that properly trained neural networks (NNs) are going to be incorporated in the structure of the beamforming system that controls the antenna array. Since the traditional NNs are not expected to have the efficiency required by adaptive BF, new architectures are going to be studied, such as Recurrent or Convolutional neural networks (NNs) or a combination of these two architectures. The possibility to incorporate Self Organizing Maps (a particular type of NN) for signal classification will be investigated. In addition, novel evolutionary and non-evolutionary optimization methods will be studied

Host Institution: Aristotle University of Thessaloniki, Greece

Research Objectives: The objective is to develop optimised realistic antennas or antenna arrays that concurrently satisfy multiple requirements such as maximum antenna forward gain, low side lobe levels, nulls towards the direction of arrival (DOA) of undesired incoming signals, etc. The geometry and excitation of antenna arrays will be optimised in order to shape the produced radiation pattern in a specific way depending on the application. Finally, a proper feeding network that provides the required excitations on the antenna elements as well as matching to a central transmission line will be designed. The optimisation of antennas, antenna arrays and feeding networks will be performed by developing and applying novel evolutionary optimisation techniques in conjunction with full-wave analysis methods like the Finite Difference in Time Domain (FDTD) method, the Finite Integration Technique (FIT), the Finite Element Method (FEM), and the Method of Moments (MoM).

Host Institution: Aristotle University of Thessaloniki, Greece

Research Objectives: This work will focus on the end-to-end service orchestration for FWN applications. Research will be performed on the analysis of monitoring data and traces to characterise network traffic in FWN scenarios and identify potential usage patterns. The goal will be to forecast the network status and application-specific events in order to take proactive decisions on the allocation of network and computing resources.

Host Institution: Maggioli S.p.A., Italy

Research Objectives: Broadcasters are looking for new means to cover crowded events (concerts, football games, etc.) by providing exciting perspectives to viewers while simultaneously reducing production and delivery costs. The reception of live content from cameras located in the playing field, replays, and additional contextual information on mobile devices, is a challenging user case for all the above market actors. The main issues to deal with are, the delay in the delivery of such content in the range of milliseconds rather than seconds, and the considerable strain imposed on the backhaul network. In this framework, the aim shall be to deliver benefits to media producers and mobile operators by “enabling” them to offer a highly interactive experience and by deploying “key functionalities” at the edge, i.e., evolved Multimedia Broadcast Multicast Services (eMBMS) or local network services, such as real-time analytics together with multi-tenancy support by small cells.

Host Institution: Singular Logic S.A., Greece

Research Objectives: The main objective of this ESR is to design and implement machine intelligence algorithms for the purpose of ensuring good quality of experience (QoE) in wireless ecosystems. In particular, the study will focus on the modeling of QoE with machine learning for user specific scenarios. Finally, a complete methodology for enhancing the QoE in FWNs will be proposed.

Host Institution: Technical University Sofia, Bulgaria

Research Objectives: The main objective is to extract monetary value from novel technologies. FWNs will enrich the mobile internet experience and will therefore open new possibilities, applications and services. The blending of the physical, digital and virtual worlds is pushing towards the rapid growth of new FWN-supported business model ecosystems. A modern innovative approach to the formation of new Licensed Shared Access (LSA) algorithms is considered using network slicing. Two different approaches using deep learning and blockchain-based licensing will be developed for LSA. These algorithms will then contribute to the construction of novel business models.

Host Institution: Aarhus University, Denmark

Research Objectives: In 5G, higher data rates have already been achieved; however, reliability in terms of connectivity remains a challenge. According to the International Telecommunication Union (ITU) report published in 2021, around 2.9 billion people have never used the internet, and non-terrestrial networks (NTN) in integration with terrestrial networks could be a potential solution for enabling wireless connectivity in remote areas. In release 15 of the 3rd Generation Partnership Project (3GPP), NTN was first considered, and further standardisation was given with optimisation in release 18. Considering the high mobility of drones in NTN and the broadcast nature of wireless communication, the information of the legitimate node is constantly exposed to the eavesdropper. Contrary to the terrestrial networks, providing seamless and infrastructure-less security is inevitable due to the Size, Weight and Power (SWaP) constraints on the drones in NTN, which can only be possible with the physical layer security (PLS). Multiple PLS techniques are used to ensure security; however, secret key generation is the best approach in NTN due to the power efficiency over the rest of the techniques. The secret key generation by exploiting the wireless random channel followed by information reconciliation is one of the physical layer-based techniques used in data encryption. role is to improve the encryption mechanism by improving the quantisation and reconciliation in the secret key generation with the help of wireless random channel sampling without compromising the Quality-of-Service (QoS) in NTN.

Host Institution: Aarhus University, Denmark

Research Objectives: With the number of IoT devices booming and the number of connected devices per human increasing rapidly, the demand for directed communication goes beyond advanced base stations in a network and into the wearable terminals. Both in rural and urban environments the radio spectrum is becoming increasingly crowded, so the need for directed communication is there to increase capacity, range and accuracy. Multiple simultaneous beams can increase link redundancy, diversity, and reliability. Specifically, reliability will be important when the commercial use of drones in urban areas will start to increase. The main objective is to create a new generation of miniaturized adaptive beamforming antenna arrays based on metamaterials for beyond 5G air-borne and wearable devices for tactical and commercial use. These will enable the creation of pocket-sized 3D radars, 2D direction-height finders, high accuracy geolocation of signals, and jamming-resistant highly reliable communication antennas.

Host Institution: MyDefence Communication Aps, Denmark

Research Objectives: The main objective is to demonstrate how ultra-reliable low-latency communications (URLLC) services can be used in the IoT context. Specifically, for Industry 4.0 solutions such as industrial automation, reliability and low latency are critical. Similar requirements arise in the context of connected vehicles. While low power wide area networking (LPWAN) technologies have triggered significant proliferation of IoT devices, these cannot scale to provide low latency interfaces. Cellular IoT solutions, specifically NB (Narrow Band)-IoT can provide much better coverage and latency performance. Ideally future IoT networks should be agile enough to incorporate dynamic reconfiguration of protocol parameters in response to network conditions and application requirements. The work carried out by ESR13 will focus on the development of such capabilities.

Host Institution: Politehnica University of Bucharest, Romania

Research Objectives: Many operators consider propagation prediction software (such as ICS Telecom, LS Telcom, etc.) as an efficient tool to design their access network and verify its coverage. The new 5G standard and the diversity of the generic access points /eNodeBs (for outdoor/indoor access, for “classical” services or for D2D/M2M devices) is coming with new challenges for the access network design according to the requirements of the operators, network providers, and simple users.

Host Institution: Politehnica University of Bucharest, Romania

Research Objectives: Network Function Virtualisation (NFV) and Multi-access Edge Computing (MEC) are considered to be a vital technology for FWNs. FWNs will support dynamic end-to-end service provisioning with a fully programmable underlying infrastructure, which is operated by an intelligent MANO framework. The objective of this ESR is to develop a convergence between Virtual Network Functions (VNFs) and Container Network Functions (CNFs) with unified MEC and Cloud application programme interfaces (APIs).

Host Institution: IQUADRAT, Spain