http://hdl.handle.net/10993/86 Search this channel search https://orbilu.uni.lu/simple-search http://hdl.handle.net/10993/56047 Title: Impact of Disentanglement on Pruning Neural Networks <br/> <br/>Author, co-author: Shneider, Carl; Rostami Abendansari, Peyman; Kacem, Anis; Sinha, Nilotpal; Shabayek, Abd El Rahman; Aouada, Djamila <br/> <br/>Abstract: Efficient model compression techniques are required to deploy deep neural networks (DNNs) on edge devices for task specific objectives. A variational autoencoder (VAE) framework is combined with a pruning criterion to investigate the impact of having the network learn disentangled representations on the pruning process for the classification task. Mon, 02 Oct 2023 03:30:40 GMT http://hdl.handle.net/10993/56046 Title: Impact of Disentanglement on Pruning Neural Networks <br/> <br/>Author, co-author: Shneider, Carl; Rostami Abendansari, Peyman; Kacem, Anis; Sinha, Nilotpal; Shabayek, Abd El Rahman; Aouada, Djamila <br/> <br/>Abstract: Deploying deep learning neural networks on edge devices, to accomplish task specific objectives in the real-world, requires a reduction in their memory footprint, power consumption, and latency. This can be realized via efficient model compression. Disentangled latent representations produced by variational autoencoder (VAE) networks are a promising approach for achieving model compression because they mainly retain task-specific information, discarding useless information for the task at hand. We make use of the Beta-VAE framework combined with a standard criterion for pruning to investigate the impact of forcing the network to learn disentangled representations on the pruning process for the task of classification. In particular, we perform experiments on MNIST and CIFAR10 datasets, examine disentanglement challenges, and propose a path forward for future works. Mon, 02 Oct 2023 03:30:38 GMT http://hdl.handle.net/10993/56019 Title: Distributed cohesive radio systems for spaceborne applications <br/> <br/>Author, co-author: Martinez Marrero, Liz <br/> <br/>Abstract: During the last decade, the use of small satellites has revolutionized the field of space exploration and communications. This has opened up new possibilities, such as the feasibility of grouping them into Cohesive Distributed Satellite Systems (CDSSs). A CDSS is a multi-satellite configuration that appears as a single solid entity from an external perspective, which includes data reception, processing, and transmission operations. This entails improving several DSS applications such as Earth observation, geolocation, navigation, imaging, and communications. The synchronization of CDSSs involves precisely aligning time, frequency, and phase among multiple satellites, which is a significant challenge due to the inherent characteristics of space-based environments. For instance, the spacecraft mobility, the round-trip delay, and the resource constraints make the synchronization of DSSs more challenging than its equivalent in wireless terrestrial networks. However, it is simultaneously an unavoidable challenge for future space communications. This requirement does not only apply to small satellites DSS but also to avoid interference in crowded orbits and enable the federated satellites’ system paradigm. This thesis aims to identify the technical synchronization requirements and design the synchronization and coordination techniques to perform cohesive transmission in a DSS. First, we studied the state-of-the-art synchronization techniques and analyzed their feasibility for DSS. Additionally, we summarized other methods related to the synchronization of DSS, such as inter-satellite ranging and positioning. Then, we considered a first approximation to the problem, assuming accurate time synchronization and relative positioning among the satellites in a DSS. This problem is equivalent to synchronizing the local oscillators’ phase in a precoding-enabled multi-beam satellite system. One of the most significant synchronization impairments for implementing CDSSs is the phase noise of the LOs in different spacecraft. In this regard, the two-state phase noise model was implemented and integrated into the channel emulator of the MIMO end-to-end satellite emulator, which allowed us to validate the results included in this thesis. Next, we analyzed the impact of the phase errors and uncertainties in operating a precoded forward link satellite communication system. We formally demonstrated that the uplink phase variations affect precoding performance even when all the LOs share a single frequency reference. Additionally, we identified the individual contributions of each system element to the overall synchronization uncertainties in practical precoding implementations. Besides, for linear and non-linear precoding, we formally demonstrated that the UTs can track slow time variations in the channel if they equally affect all the beams. The compensation loop to mitigate these impairments was designed, implemented, and integrated into the GW of the MIMO end-to-end satellite emulator. The solution is a closed-loop algorithm that uses the periodical channel phase measurements sent to the GW by the UTs as part of traditional precoding implementations. The proportional-integral controller included in the GW calculates the compensation phase required to align all the beams to the phase of the designated reference beam. Besides, we compared different approaches to combine the channel phase estimations obtained from the UTs using the amplitude of the estimated channel and the UT’s thermal noise. The compensation loop and the combining estimations hardware implementations were used in real-time experiments to assess the feasibility of the precoding technique for GEO satellite systems. Fri, 22 Sep 2023 08:40:34 GMT http://hdl.handle.net/10993/56014 Title: Propagation Characteristics of a Partially Coherent Gaussian Schell-model Array Vortex Beam in the Joint Turbulence Effect of a Jet Engine and Atmosphere <br/> <br/>Author, co-author: Nabil, Hassan; Balhamri, Adil; Bayraktar, Mert; Belafhal, Abdelmajid <br/> <br/>Abstract: This work investigates the joint effects of jet engine exhaust-induced turbulence and atmospheric turbulence on the propagation of a partially coherent Gaussian Schell-model Array (GSMA) vortex beam. Using the two-process propagation method, analytical formulae are derived for the cross-spectral density, spectral density, degree of coherence, and beam width of the considered beam. The results show that the considered beam takes different shapes; when the spatial coherence is large, the spectral density of the GSMA vortex beam takes an elliptical shape, whereas when the spatial coherence is smaller, the spectral density remains a Gaussian shape. The evolution profile of the degree of coherence weakens gradually when the propagation distance, topological charge, and turbulence strength increase. Moreover, the profile of the degree of coherence takes the Gaussian profile when the propagation distance is longer or turbulence atmospheric is stronger. Furthermore, the results reveal that the corresponding beam spreads faster with a larger propagation distance, lower spatial coherence, and high-strength turbulence. This study also concludes from the results that the beam is affected more when its propagation is near the jet engine exhaust, which means that this latter has a significant impact. Wed, 20 Sep 2023 14:31:12 GMT http://hdl.handle.net/10993/56011 Title: Gateway Station Geographical Planning for Emerging Non-Geostationary Satellites Constellations <br/> <br/>Author, co-author: Monzon Baeza, Victor; Ortiz Gomez, Flor de Guadalupe; Lagunas, Eva; Salih Abdu, Tedros; Chatzinotas, Symeon Wed, 20 Sep 2023 10:09:12 GMT http://hdl.handle.net/10993/55999 Title: Performance of hollow hyperbolic sinusoidal Gaussian beam in weak turbulent optical communication links <br/> <br/>Author, co-author: Secilmis, Gamze; Bayraktar, Mert; Elmabruk, Kholoud <br/> <br/>Abstract: The performance of hollow hyperbolic sinusoidal Gaussian beam (HHsGB) propagating in weak turbulent optical communication link is examined by analysing the beam scintillation indices. System parameters’ effect on point-like and aperture-averaged scintillation index values is investigated. The obtained results show that HHsGBs with small source sizes improve the link performance by reducing the scintillation level. In communication systems with a big receiver aperture radius, HHsGBs with high orders significantly minimize the scintillation level. Thus, the provided results will have significant potential in improving free-space optical communication performance in a wide range of applications. Mon, 18 Sep 2023 07:26:18 GMT http://hdl.handle.net/10993/55836 Title: NOMA-Based Backscatter Communications: Fundamentals, Applications, and Advancements <br/> <br/>Author, co-author: Ahmed, Manzoor; Shahwar, Muhammad; Khan, Feroz; Khan, Wali Ullah; Ihsan, Asim; Khan, Umer Sadiq; Xu, Fang; Chatzinotas, Symeon <br/> <br/>Abstract: Developing wireless communication technologies is an ongoing process to satisfy the requirements of new applications and the increasing proliferation of interconnected devices. Using non-orthogonal multiple access (NOMA) and backscatter communication (BC) has surfaced as an advantageous approach for enhancing energy efficiency (EE), maximizing sum rates, ensuring security, and optimizing resource allocation. NOMA permits multiple users to share time and frequency resources even without the requirement of antenna arrays, whereas BC employs ambient RF signals for low-power communication. By integrating the advantages of NOMA and BC, NOMA-based BC provides a solution for future energy-efficient and low-power networks. Despite its potential, there is a lack of a comprehensive overview of NOMA-BC, necessitating a systematic survey that covers its principles, applications, challenges, and future directions. This survey aims to bridge the gap by exploring NOMA-BC within B5G and 6G networks. We delve into its technical aspects, performance optimization techniques, and real-world applications to enhance understanding and knowledge. First, we cover topics such as enhancing EE, maximizing the sum rates, ensuring security, and optimizing resource allocation. Our primary goal is to provide researchers and practitioners with valuable insights that enable them to grasp the capabilities and benefits of NOMA-BC. To achieve this, we comprehensively analyze various schemes' performance by presenting detailed summary tables. These analyses cover a range of scenarios, methods, and objectives, focusing on emerging B5G technologies such as reconfigurable intelligent surfaces (RIS), visible light communication (VLC), and unmanned aerial vehicle (UAV) communication. By examining NOMA-BC's effectiveness within these contexts, we aim to provide a holistic view of its potential and applicability in diverse technological domains. Moreover, our survey identifies and discusses open research challenges and proposes future directions to guide researchers toward unexplored areas and facilitate advancements in NOMA-BC. Wed, 23 Aug 2023 18:56:43 GMT http://hdl.handle.net/10993/55834 Title: Joint optimization of UAV-IRS placement and resource allocation for wireless powered mobile edge computing networks <br/> <br/>Author, co-author: Ahmec, Manzoor; Alshahrani, Haya Mesfer; Alruwais, Nuha; Asiri, Mashael M.; Duhayyim, Mesfer Al; Khan, Wali Ullah; khurshaid, Tahir; Nauman, Ali <br/> <br/>Abstract: The rapid evolution of communication systems towards the next generation has led to an increased deployment of Internet of Things (IoT) devices for various real-time applications. However, these devices often face limitations in terms of processing power and battery life, which can hinder overall system performance. Additionally, applications such as augmented reality and surveillance require intensive computations within tight timeframes. This research focuses on investigating a mobile edge computing (MEC) network empowered by unmanned aerial vehicle intelligent reflecting surfaces (UAV-IRS) to enhance the computational energy efficiency of the system through optimized resource allocation. The MEC infrastructure incorporates the energy transfer circuit (ETC) and edge server (ES), co-located with the intelligent access point (AP). To eliminate interference between energy transfer and data transmission, a time-division multiple access method is utilized. In the first phase, the ETC wirelessly transfers power to low-power IoT devices, which efficiently harvest and store the received energy in their batteries. In the second phase, IoT devices employ the stored energy for local computing or offloading tasks. Furthermore, the presence of tall buildings may obstruct communication routes, impacting system functionality. To address these challenges, we propose an optimization framework that simultaneously considers time, power, phase shift design, and local computational resources. This joint optimization problem is non-convex and non-linear, making it NP-hard. To tackle this complexity, we decompose the problem into subproblems and solve them iteratively using a convex optimization toolbox like CVX. Through simulations, we demonstrate that our proposed optimization framework significantly improves 40.7% system performance compared to alternative approaches. Wed, 23 Aug 2023 18:28:12 GMT http://hdl.handle.net/10993/55833 Title: Reconfigurable Intelligent Surfaces Enhanced NOMA D2D Communications Underlaying UAV Networks <br/> <br/>Author, co-author: Khan, Wali Ullah <br/> <br/>Abstract: Device-to-device (D2D) communications offers high spectral efficiency, low energy consumption and transmission latency. However, one of the main limitations of D2D communications is co-channel interference from underlaying wireless system. Reconfigurable intelligent surfaces (RIS) is a promising technology because it can manipulate the electromagnetic waves in their environment to overcome interference and enhance wireless communications. This paper considers RIS enhanced D2D communications underlaying unmanned aerial vehicle (UAV) networks with non-orthogonal multiple access (NOMA). The objective is to maximize the sum rate of NOMA D2D communications by optimizing the power budget of D2D transmitter, NOMA power allocation coefficients of D2D receivers and passive beamforming of RIS while guaranteeing the quality of services of UAV user. Due to non-convexity, the optimization problem is intractable and challenging to handle. Therefore, it is solved in two parts using alternating optimization. Simulation results unviel the performance of the proposed RIS enhanced D2D communications scheme. Results demonstrate that the proposed scheme ach Wed, 23 Aug 2023 18:10:28 GMT http://hdl.handle.net/10993/55832 Title: Resource Optimization for Integrated Terrestrial Non-Terrestrial Networks Involving IRS <br/> <br/>Author, co-author: Khan, Wali Ullah; Mahmood, Asad; Lagunas, Eva; Jamshed, Muhammad Ali; Chatzinotas, Symeon; Ottersten, Björn <br/> <br/>Abstract: Intelligent reconfigurable surfaces (RIS) have emerged as one of the most promising and cost-effective technologies due to their high energy efficiency, extended wireless coverage, enhanced signal strength, and interference mitigation capability. This paper provides a new framework of cognitive radio-based integrated terrestrial non-terrestrial networks (ITNTNs) involving IRS. The objective is to maximize the achievable sum rate of the secondary network by simultaneously optimizing the transmission power, user association, phase shift design of IRS and 2D placement of UAVs while controlling the co-channel interference temperature to the primary network. The problem is formulated as non-convex/non-linear due to interference and decision variables which makes it NP-hard and intractable. To reduce the complexity and make the problem tractable, we first decouple it into subproblems and iteratively obtain an efficient solution. Numerical results demonstrate that the proposed optimization scheme converges within a few iterations and achieves high sum rate than the benchmark suboptimal schemes. Wed, 23 Aug 2023 18:04:24 GMT http://hdl.handle.net/10993/55791 Title: New Horizons in Tactical Communications: An Overview of Emerging Technologies Possibilities <br/> <br/>Author, co-author: Monzon Baeza, Victor; Concha Salor, Laura Tue, 15 Aug 2023 17:53:03 GMT http://hdl.handle.net/10993/55784 Title: Towards a peer-to-peer residential short-term load forecasting with federated learning <br/> <br/>Author, co-author: Delgado Fernandez, Joaquin; Potenciano Menci, Sergio; Pavić, Ivan <br/> <br/>Abstract: The inclusion of intermittent and renewable energy sources has increased the importance of demand forecasting in the power systems. Smart meters play a critical role in modern load forecasting due to the high granularity of the measurement data. Federated Learning can enable accurate residential load forecasting in a distributed manner. In this regard, to compensate for the variability of households, clustering them in groups with similar patterns can lead to more accurate forecasts. Usually, clustering requires a central server that has access to the entire dataset, which collides with the decentralized nature of federated learning. In order to complement federated learning, this study proposes a decentralized Peer-to-Peer strategy that employs agent-based modeling. We evaluate it in comparison to a typical centralized k-means clustering. To create clusters, we compare Euclidian and Dynamic time warping distances. We employ these clusters to build short-term load forecasting models using federated learning. Our results reveal the possibility of using Peer-to-Peer clustering along with simple Euclidean distances and Federated Learning to obtain highly performant load forecasting models in a fully decentralized setting. Mon, 14 Aug 2023 08:21:26 GMT http://hdl.handle.net/10993/55582 Title: Early-Stopped Technique for BCH Decoding Algorithm Under Tolerant Fault Probability <br/> <br/>Author, co-author: Chou, Hung-Pu <br/> <br/>Abstract: In this paper, a technique for the Berlekamp Massey(BM) algorithm is provided to reduce the latency of decoding and save decoding power by early termination or early stopped checking. We investigate the consecutive zero discrepan cies during the decoding iteration and decide to early stop the decoding process. This technique is subject to decoding failure in exchange for the decoding latency. We analyze our propose d technique by considering the weight distribution of BCH code and estimating the bounds of undetected error probability as the event of enormous stop checking. The proposed method is effective in numerical results and the probability of decoding failure is lower than 10 −119 for decoding 16383 code length of BCH codes. Furthermore, the complexity compared the conventional early termination method with the proposed approach for decoding the long BCH code. The proposed approach reduces the complexity of the conventional approach by up to 80%. As a result, the FPGA testing on a USB device validates the reliability of the proposed method Mon, 10 Jul 2023 11:08:34 GMT http://hdl.handle.net/10993/55515 Title: RIS-Assisted Energy-Efficient LEO Satellite Communications with NOMA <br/> <br/>Author, co-author: Khan, Wali Ullah; Lagunas, Eva; Mahmood, Asad; Chatzinotas, Symeon; Ottersten, Björn <br/> <br/>Abstract: This paper proposes an energy-efficient RIS-assisted downlink NOMA communication for LEO satellite networks. The proposed framework simultaneously optimizes the transmit power of ground terminals of the LEO satellite and the passive beamforming of RIS while ensuring the quality of services. Due to the nature of the considered system and optimization variables, the energy efficiency maximization problem is non-convex. In practice, obtaining the optimal solution for such problems is very challenging. Therefore, we adopt alternating optimization methods to handle the joint optimization in two steps. In step 1, for any given phase shift vector, we calculate satellite transmit power towards each ground terminal using the Lagrangian dual method. Then, in step 2, given the transmit power, we design passive beamforming for RIS by solving the semi-definite programming. We also compare our solution with a benchmark framework having a fixed phase shift design and a conventional NOMA framework without involving RIS. Numerical results show that the proposed optimization framework achieves 21.47% and 54.9% higher energy efficiency compared to the benchmark and conventional frameworks. Tue, 04 Jul 2023 15:02:47 GMT http://hdl.handle.net/10993/55506 Title: Trusted Reconfigurable Intelligent Surface for Multi-User Quantum Key Distribution <br/> <br/>Author, co-author: Kisseleff, Steven; Chatzinotas, Symeon <br/> <br/>Abstract: We consider a multi-user quantum key distribution (QKD) system based on free-space optics (FSO) in terrestrial environment. Due to obstacles in the signal path, FSO-QKD is heavily affected by decoherence and other harmful effects. In order to avoid the performance degradation, trusted nodes (TNs) are introduced in the QKD network. However, in absence of alternative signal routes, the classical data and the quantum key are relayed by the same TN, which makes it a promising target for an eavesdropping attack. In order to address this issue, we explore the use of a trusted reconfigurable intelligent surface (RIS) that passively reflects the classical signals and manipulates them to make the encryption consistent with the distributed keys. The concept is supported by numerical simulations with multiple FSO-QKD users and an eavesdropper. We demonstrate that the level of data protection is very high with the proposed concept even in case of a successful attack on the trusted RIS and a complete key acquisition by the eavesdropper. Tue, 04 Jul 2023 06:03:32 GMT http://hdl.handle.net/10993/55483 Title: Numerical investigation of performance of mirrored Bessel beam in turbulence <br/> <br/>Author, co-author: Bayraktar, Mert; Salma, Chib; Belafhal, Abdelmajid <br/> <br/>Abstract: We study scintillation and bit error rate performance of mirrored Bessel beams through turbulent atmosphere in this article. We beneft from numerical method to model the atmosphere. Since scintillation plays a vital role in optical wireless applications, reduction in this term provides better performance in these systems. Related with this, our results indicate that it is possible to decrease scintillation by increasing beam order to three when strong turbulent conditions are satisfed. In addition, we observe that argument of Bessel beam has more dominant role than beam order in moderate turbulence. Lastly, mirroring brings us a slight advantage in case of bit error rate. Results of this study can be benefcial for optical link designers Mon, 03 Jul 2023 11:49:13 GMT http://hdl.handle.net/10993/55297 Title: Efficient resource prediction framework for software-defined heterogeneous radio environmental infrastructures <br/> <br/>Author, co-author: Nawaz, Muhammad Ul Saqlain; Ehsan, Muhammad Khurram; Mahmood, Asad; Mumtaz, Shahid; Sodhro, Ali Hassan; Khan, Wali Ullah <br/> <br/>Abstract: Artificial Intelligence (AI) is defining the future of next-generation infrastructures as proactive and data-driven systems. AI-empowered radio systems are replacing the existing command and control radio networks due to their intelligence and capabilities to adapt to the radio environmental infrastructures that include intelligent networks, smart cities and AV/VR enabled factory premises or localities. An efficient resource prediction framework (ERPF) is proposed to provide proactive knowledge about the availability of radio resources in such software-defined heterogeneous radio environmental infrastructures (SD-HREIs). That prior information enables the coexistence of radio users in SD-HREIs. In a proposed framework, the radio activity is measured in both the unlicensed bands that include 2.4 and 5 GHz, respectively. The clustering algorithms k- means and DBSCAN are implemented to segregate the already measured radioactivity as signal (radio occupancy) and noise (radio opportunity). Machine learning techniques CNN and LRN are then trained and tested using the segregated data to predict the radio occupancy and radio opportunity in SD-HREIs. Finally, the performance of CNN and LRN is validated using the cross-validation metrics. Wed, 07 Jun 2023 09:30:41 GMT http://hdl.handle.net/10993/55296 Title: Energy-Efficient Beamforming and Resource Optimization for STAR-IRS Enabled Hybrid-NOMA 6G Communications <br/> <br/>Author, co-author: Asif, Muhammad; Ihsan, Asim; Khan, Wali Ullah; Ali, Zain; Zhang, Shengli; Xiaoxiao, Sissi <br/> <br/>Abstract: In this manuscript, we propose an energy-efficient optimization framework for a multi-cluster simultaneous transmitting and reflecting intelligent reflecting surfaces (STAR-IRS) enabled time-division multiple-access (TDMA) based hybrid-NOMA system to realize the future sixth-generation (6G) wireless communication systems. Specifically, the energy-efficiency maximization is achieved by optimizing the successive-interference cancellation (SIC) decoding order, time-allocation, and active-beamforming vectors at the transmitter, as well as transmission and reflection coefficients at the STAR-IRS under quality-of-service (QoS), conservation of energy, time-allocation, phase-shifts, and SIC-decoding constraints. Moreover, the proposed alternating optimization algorithm tackles the considered highly non-convex optimization problem in four steps. In first step, for computing the SIC-decoding order of NOMA users, an efficient optimization technique is proposed which maximizes the sum of combined channel gains by optimizing the transmission and reflection beamforming vectors of the considered STAR-IRS assisted hybrid-NOMA system. Further, in second step, an optimal time-allocation for each cluster in transmission and reflection region is computed for given SIC-decoding order. With decoding order and time-allocation in hand, active-beamforming vectors are computed by exploiting the sequential-convex approximation (SCA) and second-order-conic programming (SOCP) in third step. Finally, in the fourth step, the transmission and reflection coefficients of STAR-IRS are computed by transforming the non-convex optimization problem into a semi-definite programming (SDP) problem. Th numerical simulation results demonstrate that the proposed optimization framework exhibits an efficient energy efficiency performance and converges within a few iterations. Wed, 07 Jun 2023 09:21:37 GMT http://hdl.handle.net/10993/55204 Title: SDN-based Testbed for Emerging Use Cases in Beyond 5G NTN-Terrestrial Networks <br/> <br/>Author, co-author: Minardi, Mario; Drif, Youssouf; Vu, Thang Xuan; Maity, Ilora; Politis, Christos; Chatzinotas, Symeon Tue, 23 May 2023 14:10:10 GMT http://hdl.handle.net/10993/55067 Title: Efficient Content Caching for Delivery Time Minimization in the LEO Satellite Networks <br/> <br/>Author, co-author: Bhandari, Sovit; Vu, Thang Xuan; Chatzinotas, Symeon; Ottersten, Björn <br/> <br/>Abstract: Next-generation multi-spot beam satellite systems open a new way to design low earth orbit (LEO) satellite communication systems with full flexibility in managing bandwidth, transmit power, and spot beam coverage, enabling the adoption of spatial multiplexing techniques to meet the unprecedented demand for future mobile traffic. However, conventional spatial multiplexing techniques perform poorly in satellite systems due to high correlation between the satellite channels, resulting in inefficient mitigation of inter-user interference. In this paper, we exploit the flexibility of multi-spot beam LEO satellites and consider the geographic distribution of users to improve the performance of LEO satellite-assisted edge caching systems. Our goal is to jointly optimize the beam coverage, bandwidth and transmit power and minimize the cache delivery time. In particular, the spot beam coverage is optimized by using the K-means algorithm applied to the realistic user demands, followed by a proposed successive convex approximation (SCA)-based iterative algorithm for optimizing the radio resources. Simulations shows that our optimal approach outperforms the conventional precoding-based approach and also shows a significant improvement in the minimization of the maximum content delivery time. Mon, 08 May 2023 12:03:54 GMT