IMDEA Networks Institute Digital Repository
Not a member yet
1915 research outputs found
Sort by
Modulating LiFi for dual operation in the visible and infrared spectra
Get PDFLight-Fidelity (LiFi) has emerged in the last few years as a promising technology for alleviating the stringent demand for wireless data services. Prior works have considered LiFi operating either in the visible light or infrared spectrum. Each spectrum band has its own advantages: visible light allows leveraging existing infrastructure for communication, while infrared is not affected by light dimming. In this work, we propose a modulation scheme that retains the benefits of both bands, introducing a simple, low-cost, yet efficient dimming solution for LiFi networks. We compare the performance of the proposed dimming scheme with both the digital and analog dimming techniques traditionally used in LiFi systems. Simulation results show that our dimming solution offers better communication and illumination performance than previous proposals, providing larger signal-to-noise ratio, spectral efficiency, and a full and fine-grained dimming range. Finally, we prototype our solution by designing an extended version of the OpenVLC 1.3 platform, and we experimentally show its robust communication performance under different dimming conditions. We make the implemented system publicly available to the research community.European UnionMinisterio de Asuntos Económicos y Transformación DigitalTRUEpu
Reversing the Virtual Maze: An Overview of the Technical and Methodological Challenges for Metaverse App Analysis
Get PDFThe Metaverse is a virtual world that is becoming increasingly popular.
Recent technological advances, such as head-mounted displays and novel sensors, have allowed for the harmonious integration of the virtual world into our reality.
This integration is transforming how we interact with our environment and with each other and offers endless entertainment, social, and business opportunities.
Since the technology is in its early stages and far from market consolidation, there has been a notable proliferation of new platforms and devices resulting in a highly heterogeneous ecosystem.
Much like with Smartphones, popular platforms allow for the installation of third-party applications, accessible through online marketplaces.
However, the Metaverse aims to enable seamless coordination between virtual environments.
As a result, developers face the pressure of being present on a wide range of platforms and rely on cross-platform development frameworks that add yet another layer of complexity to the stack.
This complexity, together with the fact that XR headsets are equipped with an arsenal of disrupting new sensors, has the potential to pose new risks to the security and privacy of their users.
Although progress has been made in identifying risks in this ecosystem, there is still a significant gap in methodologies and techniques for studying actionable risks in popular headsets.
In this work, we present a vision --- based on experience --- for application analysis that considers various ecosystem components and highlights challenges to address emerging threats effectively.MCIN/AEI/ 10.13039/501100011033ERDF, EUMITES/SEPEESF Investing in your futureEU NextGeneration-EU/PRTRMCIN/AEI/ 10.13039/50110001103TRUEinpres
Unleashing Flexibility and Interoperability in QKD Networks: The Power of Softwarized Architectures
Get PDFSoftwarization and disaggregated architectures have transformed traditional communication networks by separating network control functions from the hardware, catalyzing innovation, and dramatically reducing deployment and operation costs. This paper extends these concepts to Quantum Key Distribution (QKD) networks, emphasizing the importance of virtualizing key management functions with advanced cloud-native technologies for enhanced global interoperability and flexibility. Our approach advocates for a fully disaggregated key management plane supported by softwarization and virtualization technologies, without losing sight of the established standards. The proposed model unlocks multiple benefits, including the facilitation of tailored access control mechanisms for QKD network administrators, the ability to apply real-time performance monitoring and base the control and management of the network on the gathered data, as well as conducting different key generation Quality of Service (QoS) admission control mechanisms on the QKD network. Moreover, a disaggregated model of the key management functions facilitates the implementation of different strategies for cross-domain collaboration between QKD network providers, as well as continuous key provisioning even in the midst of quantum infrastructure outages. The model has been validated through experiments in a virtualized environment, leveraging software tools to create customizable digital twins of QKD networks.TRUEpu
A meta-learning approach in a cattle weight identification system for anomaly detection
Get PDFWeighing management in cattle farming is important for farmers, as it allows them to accurately monitor the growth and development of their animals. It is also a valuable tool that allows farmers to maximize the production and welfare of their animals. However, it is difficult for the farmer to detect if the herd of animals being weighed is gaining the ideal weight for a given breed and age. In addition, normally, when a new breed of cattle is introduced to a farm, there is very little data. This article proposes a meta-learning framework (MTL) for identification models used in the fattening process of animals to detect anomalies in cattle weight. The proposed MTL framework has a knowledge base of Meta-Models on Identification models based on machine learning techniques, which is used to select the identification model to use when a new breed of cattle arrives on the farm. This knowledge base is updated, either because a previous identification model has been successfully adapted to the new breed, or a new identification model has had to be generated, allowing the framework to continuously improve its performance over time. Particularly, this article presents in detail the process of adaptation of the previous identification models to new breeds carried out by our MTL framework. Besides, to test our approach, a case study is presented, using records of animals raised and fattened at the ”El Rosario” farm, located in the municipality of Monteria (Córdoba-Colombia). The results are very encouraging in terms of the ability of our framework to adapt the identification models to different possible scenarios in the process of detecting anomalous weights. In general, the identification models generated with our proposal had an of 90.8%, which suggests that the models can explain the variability observed in the data.TRUEpu
Angle estimation using mmWave RSS measurements with enhanced multipath information
Get PDFmmWave communication has come up as the unexplored spectrum for 5G services. With new standards for 5G NR positioning, more off-the-shelf platforms and algorithms are needed to perform indoor positioning. An object can be accurately positioned in a room either by using an angle and a delay estimate or two angle estimates or three delay estimates. We propose an algorithm to jointly estimate the angle of arrival (AoA) and angle of departure (AoD), based only on the received signal strength (RSS). We use mm-FLEX, an experimentation platform developed by IMDEA Networks Institute that can perform real-time signal processing for experimental validation of our proposed algorithm. Codebook-based beampatterns are used with a uniquely placed multi-antenna array setup to enhance the reception of multipath components and we obtain an AoA estimate per receiver thereby overcoming the line-of-sight (LoS) limitation of RSS-based localization systems. We further validate the results from measurements by emulating the setup with a simple ray-tracing approach.TRUEpu
AICHRONOLENS: Advancing Explainability for Time Series AI Forecasting in Mobile Networks
Get PDFNext-generation mobile networks will increasingly rely on the ability to forecast traffic patterns for resource
management. Usually, this translates into forecasting diverse objectives like traffic load, bandwidth, or channel spectrum utilization, measured over time. Among the other techniques, Long-Short Term Memory (LSTM) proved very successful for this task. Unfortunately, the inherent complexity of these models
makes them hard to interpret and, thus, hampers their deployment in production networks. To make the problem worsen, EXplainable Artificial Intelligence (XAI) techniques, which are primarily conceived for computer vision and natural language processing, fail to provide useful insights: they are blind to the temporal characteristics of the input and only work well with highly rich semantic data like images or text. In this paper, we take the research on XAI for time series forecasting one step further proposing AICHRONOLENS, a new tool that links legacy XAI explanations with the temporal properties of the input. In such a way, AICHRONOLENS makes it possible to dive deep into the model behavior and spot, among other aspects, the hidden cause of errors. Extensive evaluations with real-world mobile traffic traces pinpoint model behaviors that would not be possible to spot otherwise and model performance can increase by 32 %.Ministerio de Asuntos Económicos y Transformación DigitalMinisterio de Ciencia e InnovaciónMinisterio de Trabajo y Economía SocialTRUEpu
Optimizing QoS in Secure RIS-Assisted mmWave Network With Channel Aging
Get PDFReconfigurable Intelligent Surfaces (RISs) have demonstrated significant potential in securing mmWave communication from potential eavesdropping by configuring reflecting elements to enhance signal strength at desired locations and create nulls at potential eavesdropping locations. Acquiring perfect channel information is crucial for optimizing RIS configuration; however, obtaining such information is costly and, as a result, should be performed sparingly. This work explores the impact of the age of channel information on secrecy performance when a RIS-assisted mmWave network operates under statistical qualityof-service (QoS) constraints. Specifically, we optimize the QoS performance of a RIS-assisted mmWave network given only outdated channel estimates. To this end, we propose a technique for the joint optimization of transmit beamforming and RIS configuration, along with a closed-form solution for the optimal transmit power control policy. We investigate the impact of channel aging on the performance of these techniques. In our Monte-Carlo simulations, we first identify the factors influencing the aging process of a RIS-assisted mmWave channel in both the near and far fields of the RIS. Subsequently, we examine the impact of channel aging on secrecy capacity and demonstrate that adequate secrecy capacity can still be achieved even when channel information is outdated, reducing the need for frequent RIS configuration. Moreover, our optimal power control policy results reveal that operating in a high SNR regime does not
necessarily increase the achievable effective secrecy capacity when the system operates under stricter QoS constraints. This finding allows system designers to adopt a more pragmatic system design approach that consumes less energy while maintaining the required QoS and secrecy performance.European UnionTRUEpu
Efficient network control for large and highly dense millimeter wave deployments
Get PDFWireless networks have become an integral part of modern society, providing ubiquitous connectivity to a growing number of connected devices. Concepts like Augmented Reality (AR)/Virtual Reality (VR), remote surgery and Industry 4.0 will further increase the number of users and the volume of data being transferred. Satisfying the demands for higher throughput, lower latency and higher reliability necessitates novel technologies and innovative designs. Operation in the high frequency Millimeter-Wave (mmWave) band is foreseen as a crucial part of the design of future wireless networks. The extremely large signal bandwidth offered at mmWave frequencies enables multi-Gbps, low-latency wireless connectivity for a peak performance that far exceeds what can be achieved in the currently used sub-6 GHz bands.
Realising the potential of mmWave technology requires adaptation to the challenging propagation environment at all levels of the protocol stack. Significant work has already been done to enable single-link communication through the use of phased antenna arrays to generate narrow directional beampatterns. Network aspects and interactions in large and dense networks, however, remain largely unexplored. The goal of this thesis is to study the performance of mmWave protocols in dense deployments with many Access Points (APs) and Station (STA) near each other. Such deployments are required for sufficient coverage in real-world implementations, however, they come with unique challenges due to the complex nature of interference in mmWave networks. The thesis studies different proposed architectures for mmWave to gain insight into sources of inefficiency and performance degradation. We then propose solutions to address these challenges and enhance the operation of mmWave networks.
To enable research into dense mmWave networks we implemented the latest mmWave WiFi standard, IEEE 802.11ay, in the network simulator ns-3. We used as a basis for our implementation the existing IEEE 802.11ad model, expanding it to cover advanced features like Multiple-Input and Multiple-Output (MIMO), channel bonding and novel Beamforming Training (BFT) protocols introduced in IEEE 802.11ay. Using our model were able to get in-depth insights regarding the performance of various protocol features of the state-of-the-art mmWave WiFi, including, channel access, BFT, interference management and spatial sharing.
We first focus on BFT scalability in dense deployments, looking at how the accuracy of the training can degrade in high-interference environments, as well as how the growing overhead can limit communication throughput. We propose the use of the novel Group Beamforming protocol introduced in IEEE 802.11ay as it enables simultaneous training of all STA with a Basic Service Set (BSS), rather than relying on a per AP-STA training like legacy BFT from IEEE 802.11ad. We additionally propose performance enhancements for the Group Beamforming protocol that can increase the accuracy to ensure correct beampattern selection even under significant interference. Our analysis demonstrates that the modified Group Beamforming protocol has higher accuracy than legacy BFT and enables higher network throughput due to the reductions in overhead.
We then designed a physical (PHY) layer signalling solution that enhances packet reception in mmWave WiFi devices. We focused on two sources of inefficiency caused by the contention-based random channel access - the use of omnidirectional receiver beampatterns, and the overhearing of unwanted packets. Both of these problems limit the performance in the network and affect spatial re-use. SIGNaling in the PHY Preamble (SIGNiPHY) embeds the user identifier (ID) in the PHY packet preamble, allowing for early user identification. This enables the receiver to use the correct directional beampattern to receive the packet payload, as well as to filter any packets for which it is not the recipient. Thus, SIGNiPHY increases the resilience to interference, enabling packet decoding under challenging conditions and increasing spatial sharing. We evaluated SIGNiPHY in ns-3, as well as an FPGA testbed, revealing significant gains in throughput, latency and fairness.
The next work in the thesis presented our mmWave MIMO implementation with standard compliant MIMO BFT protocols and channel access. We demonstrate how our analog BFT protocol was able to train multiple transmit and receive antennas to find spatially separated, independent streams. Challenges with mobility, the sparsity of the mmWave channel and complex BFT protocols require further research into mmWave MIMO. However, we found promising results regarding the viability of mmWave MIMO even with a fully analog architecture.
We further investigate an alternative architecture for devices with multiple Radio Frequency (RF) chains by introducing multi-connectivity. In multi-connectivity networks, users maintain several simultaneous links with spatially distributed APs. Unlike MIMO networks, multi-connectivity designs aim to not only increase throughput but also enhance resilience and robustness. This makes them extremely suitable for mmWave networks which suffer from frequent outages and service interruptions. Furthermore, mmWave multi-connectivity networks can have reduced implementation complexity by exploiting the spatial separation of the directional links. Therefore, we propose a distributed multi-connectivity design that relies solely on local analog beamforming for interference management. Our architecture was able to enhance resilience and maintain connectivity at all times even under high interference, as well as exploit the spatial diversity of the multiple links to achieve gains in throughput.
Finally, we study the novel IEEE 802.11bf protocol which aims to standardize sensing operation in WiFi. As a topic of significant interest from both academia and industry, environmental sensing using communication signals opens new possibilities for mmWave networks. We present a first initial system-level study that looks at joint communication and sensing in a mmWave WiFi network. We look at resource allocation and study how the sensing and data traffic interact with each other. We further analyse how the sensing parameters affect the performance and identify network configurations where both sensing and communication can coexist, enabling successful integration of sensing and communication in a single system.
To conclude, in this thesis we present a comprehensive analysis of dense mmWave networks, proposing performance enhancements to enhance scalability and efficiency. We then look at future possibilities for mmWave, by analysing the possibilities of advanced devices with multiple RF chains, as well as novel paradigms that integrate environmental sensing into mmWave WiFi operation.Telematics EngineeringUniversidad Carlos III de Madrid, Spai
Study of Explainability Analysis Methods for the LAMDA Family Algorithms in Classification and Clustering Tasks
Get PDFExplainability analysis is a very relevant topic today, due to the interest of allowing the interpretability of machine learning models. In this work, we carry out an in-depth study of explainability analysis for the algorithms of the LAMDA (Learning Algorithm for Multivariate Data Analysis) family that have been used in the context of supervised and unsupervised learning. In particular, for the case of classification the LAMDA-HAD algorithm, and for the case of clustering the LAMDA-RD algorithm. For the explainability analysis, two classic methods from the explainability area were considered, LIME (Local Interpretable Model-Agnostic Explanation) and Feature Importance, and another one developed by us for the LAMDA family. In particular, our explainability method for LAMDA allows measuring the importance of each characteristic in a general way, and for each cluster. In general, the results obtained in both cases (classification and clustering) are satisfactory, especially because our explainability method for LAMDA gives an explainability similar to the traditional ones, but in addition, it can be given by cluster.TRUEpu
Privacy Perceptions and Behaviors of LGBTQ+ Community in Türkiye
Get PDFThis research delves into the distinctive privacy challenges faced by the LGBTQ+ community, arising from a toxic environment and potential discrimination. By studying the privacy perceptions and behaviors in online social networks and dating applications, the study aims to inform the design of more inclusive technological solutions, with a particular focus on the LGBTQ+ community in Türkiye.TRUEpu