IMDEA Networks Institute Digital Repository
Not a member yet
1915 research outputs found
Sort by
An Experimental Prototype for Multistatic Asynchronous ISAC
Get PDFWe prototype and validate a multistatic millimeter-wave (mmWave) Integrated Sensing And Communication (ISAC) system based on IEEE 802.11ay. Compensation of the clock asynchrony between each transmitter (TX) and receiver (RX) pair is performed using the sole line-of-sight (LoS) wireless signal propagation. As a result, our system provides concurrent target tracking and micro-Doppler (μD) estimation from multiple points of view, paving the way for practical multistatic data fusion. Our results on human movement sensing, complemented with precise, quantitative ground-truth (GT) data, demonstrate the enhanced sensing capabilities of multistatic ISAC, due to the spatial diversity of the RX nodes.TRUEpu
Improved Decision Module Selection for Hierarchical Inference in Resource-Constrained Edge Devices
Get PDFThe Hierarchical Inference (HI) paradigm has recently emerged as an effective method for balancing inference accuracy, data processing, transmission throughput, and offloading cost. This approach proves particularly efficient in scenarios involving resource-constrained edge devices like micro controller units (MCUs), tasked with executing tinyML inference. Notably, it outperforms strategies such as local inference execution, inference offloading, and split inference (i.e., inference execution distributed between two endpoints). Building upon the HI paradigm, this work explores different techniques aimed at further optimizing inference task execution. We propose three distinct HI approaches and evaluate their utility for image classification.Ministry of Economic Affairs and Digital Transformation, European Union Next Generation-EU, project TSI-063000- 2021-59, and through MSCA-PF projectTRUEpu
Flowrest: Practical Flow-Level Inference in Programmable Switches with Random Forests
Get PDFUser-plane machine learning facilitates low-latency, high-throughput inference at line rate. Yet, user planes are highly constrained environments, and restrictions are especially marked in programmable switches with limited memory and minimum support for mathematical operations or data types. Thus, current solutions for in-switch inference that are compatible with production-level hardware lack support for complex features or suffer from limited scalability, and hit performance barriers in complex tasks involving large decision spaces. To address this limitation, we present Flowrest, a first complete Random Forest (RF) model implementation that operates at the level of individual flows in commercial switches. Our solution builds on (i) an original framework to embed flow-level machine learning models into programmable switch ASICs, and (ii) novel guidelines for tailoring RF models to operations in programmable switches already at the design stage. We implement Flowrest as an open-source software using the P4 language, and assess its performance in an experimental platform based on Intel Tofino switches. Tests with tasks of unprecedented complexity show how our model can improve accuracy by up to 39% over previous approaches to implement RF models in real-world equipment.European Union Horizon 2020 research and innovation program under Marie Skłodowska-Curie grant agreement no. 860239 “BANYAN”European Union Horizon 2020 research and innovation program under grant agreement no. 101017109 “DAEMON”TRUEpu
Energy-Aware Adaptive Scaling of Server Farms for NFV with Reliability Requirements
Get PDFAuto-scaling techniques aim to keep the right number of active servers for the current load: if this number is too small we risk service disruption, but if it is too large we waste resources. Despite the interest in the efficient operation of this type of systems, no prior work has addressed auto-scaling techniques for Network Function Virtualization (NFV) with stringent reliability requirements such as those envisioned in 5G (5 or 6 nines). To achieve such levels of reliability, we need to account for both the activation delay until servers become available (i.e., the wake-up or activation time) and the fallible nature of servers (which may fail with some probability). In this paper, we build on control theory to design an auto-scaling technique for a server farm for NFV that guarantees certain reliability while minimizing the number of active resources. We show that the use of well-established tools from control theory results in convergence times much shorter than those obtained with state-of-the-art reinforcement learning techniques. This shows that, despite the current trend to apply machine learning to all sorts of networking problems, there may be some cases where other techniques (such as control theory) can be more suitable.TRUEpu
Demonstrating Flow-Level In-Switch Inference
Get PDFExisting approaches for in-switch inference with Random Forest (RF) models that can run on production-level hardware do not support flow-level features and have limited scalability to the task size. This leads to performance barriers when tackling complex inference problems with sizable decision spaces. Flowrest is a complete RF model framework that fills existing gaps in the existing literature and enables practical flow-level inference in commercial programmable switches. In this demonstration, we exhibit how Flowrest can classify individual traffic flows at line rate in an experimental platform based on Intel Tofino switches. To this end, we run experiments with real-world measurement data, and show how Flowrest yields improvements in accuracy with respect to solutions that are limited to packet-level inference in programmable hardware.European Union Horizon 2020 research and innovation program under grant agreement no. 101017109 “DAEMON”European Union Horizon 2020 research and innovation program under Marie Skłodowska-Curie grant agreement no. 860239 “BANYAN”CHIST-ERA grant no. CHIST-ERA-20-SICT- 001 “ECOMOME”, via grant PCI2022-133013 of Agencia Estatal de InvestigaciónTRUEpu
A System Architecture for Battery-free IoT Networks
Get PDFWhile much research effort has been invested in long-range and low-power uplink communication for battery-free IoT networks, current deployments lack a scalable bi-directional communication infrastructure for data collection and processing with battery-free devices. To fill this gap, we introduce LoW-Fi, a system architecture specifically designed to meet the requirements of battery-free IoT applications. We show the suitability of LoW-Fi for deploying monitoring systems for precision agriculture indoors. This sector is revolutionizing with the installation of smart greenhouses that require the constant monitoring of ambient parameters to ensure optimal conditions for the crops growth. Our system is implemented using commercial off-the-shelf devices, and it works at the intersection of WiFi and LiFi for downlink and RF backscatter for uplink, retaining the advantages of each technology and solving their practical limitations. We evaluate LoW-Fi performance in a real greenhouse, and the experimental results show that it can achieve an uplink (downlink) range of 45 m (70 m) with 0% BER. The aggregated data rate is up to 4.5 Mb/s.European Union’s Horizon 2020 Marie Sklodowska Curie grant ENLIGHT’EM (814215)MSCA Postdoctoral Fellowship grant RISA-VLC (101061853)Project RISC-6G, reference TSI-063000-2021-59, granted by the Ministry of Economic Affairs and Digital Transformation and the European Union- NextGenerationEU through the UNICO-5G R&D Program of the Spanish Recovery, Transformation and Resilience Plan.TRUEpu
RAPID: Retrofitting IEEE 802.11ay Access Points for Indoor Human Detection and Sensing
Get PDFIn this work we present RAPID, the first joint communication and radar system based on next-generation IEEE 802.11ay WiFi networks operating in the 60 GHz band. Unlike existing approaches for human sensing at millimeter-wave frequencies, which rely on special-purpose radars, RAPID achieves radar-level sensing accuracy with IEEE 802.11ay access points, thus avoiding the burden of installing ad-hoc sensors. RAPID enables contactless human sensing applications, such as people tracking, Human Activity Recognition (HAR), and person identification without requiring modifications to the standard packet structure. Specifically, we leverage IEEE 802.11ay beam training to accurately localize and track multiple individuals within the same environment. Then, we propose a new way of using beam tracking to extract micro-Doppler signatures from the time-varying Channel Impulse Response (CIR) estimated from reflected packets. Such signatures are fed to a deep learning classifier to perform HAR and person identification. RAPID is implemented on a cutting-edge IEEE 802.11ay-compatible FPGA platform with phased antenna arrays, and evaluated on a large dataset of CIR measurements. It is robust across different environments and subjects, and outperforms state-of-the-art sub-6 GHz WiFi sensing techniques. Using two access points, RAPID reliably tracks multiple subjects, reaching HAR and person identification accuracies of 94% and 90%, respectively.TRUEpu
Spotting Deep Neural Network Vulnerabilities in Mobile Traffic Forecasting with an Explainable AI Lens
Get PDFThe ability to forecast mobile traffic patterns is key to resource management for mobile network operators and planning for local authorities. Several Deep Neural Networks (DNN) have been designed to capture the complex spatiotemporal characteristics of mobile traffic patterns at scale. These models are complex black boxes whose decisions are inherently hard to explain. Even worse, they have proven vulnerable to adversarial attacks which undermine their applicability in production networks. In this paper, we conduct a first in-depth study of the vulnerabilities of DNNs for large-scale mobile traffic forecasting. We propose DEEXP, a new tool that leverages EXplainable Artificial Intelligence (XAI) to understand which Base Stations (BSs) are more influential for forecasting from a spatio-temporal perspective. This is challenging as existing XAI techniques are usually applied to computer vision or natural language processing and need to be adapted to the mobile network context. Upon identifying the more influential BSs, we run stateof-the art Adversarial Machine Learning (AML) techniques on those BSs and measure the accuracy degradation of the predictors. Extensive evaluations with real-world mobile traffic traces pinpoint that attacking BSs relevant to the predictor significantly degrades its accuracy across all the scenarios.Comunidad de MadridMinisterio de Ciencia e InnovaciónEuropean UnionTRUEpu
Atomic Appends in Asynchronous Byzantine Distributed Ledgers
Get PDFA Distributed Ledger Object (DLO) is a concurrent object that maintains a
totally ordered sequence of records. In this work we formalize a linearizable
Byzantine-tolerant Distributed Ledger Object (BDLO), which is a linearizable
DLO where clients and servers processes may deviate arbitrarily from their
intended behavior (i.e. they may be Byzantine). The proposed formal definition
is accompanied by algorithms that implement BDLOs on top of an underlying
Byzantine Atomic Broadcast service.
Then we develop a suite of algorithms, based on the previous BDLO implementations, that solve the Atomic Appends problem in the presence of asynchrony, Byzantine clients and Byzantine servers. This problem occurs when
clients have a composite record (set of basic records) to append to different
BDLOs, in such a way that either each basic record is appended to its BDLO
(and this must occur in good circumstances), or no basic record is appended.
Distributed algorithms are presented, which solve the Atomic Appends problem when the clients (involved in the Atomic Appends) and the servers (which maintain the BDLOs) may be Byzantine. Finally we provide proof of concept
implementations and an experimental evaluation of the presented algorithms.Comunidad de MadridSpanish Ministry of Science and InnovationTRUEpu
Lady and the Tramp Nextdoor: Online Manifestations of Real-World Inequalities in the Nextdoor Social Network
Get PDFFrom health to education, income impacts a huge range of life choices. Many papers have leveraged data from online social networks to study precisely this. In this paper, we ask the opposite question: do different levels of income result in different online behaviors? We demonstrate it does. We present the first large-scale study of Nextdoor, a popular location-based social network. We collect 2.6 Million posts from 64,283 neighborhoods in the United States and 3,325 neighborhoods in the United Kingdom, to examine whether online discourse reflects the income and income inequality of a neighborhood. We show that posts from neighborhoods with different income indeed differ, e.g. richer neighborhoods have a more positive sentiment and discuss crimes more, even though their actual crime rates are much lower. We then show that user-generated content can predict both income and inequality. We train multiple machine learning models and predict both income (R2=0.841) and inequality (R2=0.77).TRUEinpres