arXiv subjects

Michele Rossi

Publications and source records attributed to Michele Rossi.

At least 55 records · Page 3Linked to original sources

arXiv2023

Divide and Save: Splitting Workload Among Containers in an Edge Device to Save Energy and Time

The increasing demand for edge computing is leading to a rise in energy consumption from edge devices, which can have significant environmental and financial implications. To address this, in this paper we present a novel method to enhance the energy efficiency while speeding up computations by distributing the workload among multiple containers in an edge device. Experiments are conducted on two Nvidia Jetson edge boards, the TX2 and the AGX Orin, exploring how using a different number of containers can affect the energy consumption and the computational time for an inference task. To demonstrate the effectiveness of our splitting approach, a video object detection task is conducted using an embedded version of the state-of-the-art YOLO algorithm, quantifying the energy and the time savings achieved compared to doing the computations on a single container. The proposed method can help mitigate the environmental and economic consequences of high energy consumption in edge computing, by providing a more sustainable approach to managing the workload of edge devices.

cs.DC↗

arXiv2022

SHARP: Environment and Person Independent Activity Recognition with Commodity IEEE 802.11 Access Points

In this article we present SHARP, an original approach for obtaining human activity recognition (HAR) through the use of commercial IEEE 802.11 (Wi-Fi) devices. SHARP grants the possibility to discern the activities of different persons, across different time-spans and environments. To achieve this, we devise a new technique to clean and process the channel frequency response (CFR) phase of the Wi-Fi channel, obtaining an estimate of the Doppler shift at a radio monitor device. The Doppler shift reveals the presence of moving scatterers in the environment, while not being affected by (environment-specific) static objects. SHARP is trained on data collected as a person performs seven different activities in a single environment. It is then tested on different setups, to assess its performance as the person, the day and/or the environment change with respect to those considered at training time. In the worst-case scenario, it reaches an average accuracy higher than 95%, validating the effectiveness of the extracted Doppler information, used in conjunction with a learning algorithm based on a neural network, in recognizing human activities in a subject and environment independent way. The collected CFR dataset and the code are publicly available for replicability and benchmarking purposes.

eess.SP↗

arXiv2022

Energy Consumption of Neural Networks on NVIDIA Edge Boards: an Empirical Model

Recently, there has been a trend of shifting the execution of deep learning inference tasks toward the edge of the network, closer to the user, to reduce latency and preserve data privacy. At the same time, growing interest is being devoted to the energetic sustainability of machine learning. At the intersection of these trends, we hence find the energetic characterization of machine learning at the edge, which is attracting increasing attention. Unfortunately, calculating the energy consumption of a given neural network during inference is complicated by the heterogeneity of the possible underlying hardware implementation. In this work, we hence aim at profiling the energetic consumption of inference tasks for some modern edge nodes and deriving simple but realistic models. To this end, we performed a large number of experiments to collect the energy consumption of convolutional and fully connected layers on two well-known edge boards by NVIDIA, namely Jetson TX2 and Xavier. From the measurements, we have then distilled a simple, practical model that can provide an estimate of the energy consumption of a certain inference task on the considered boards. We believe that this model can be used in many contexts as, for instance, to guide the search for efficient architectures in Neural Architecture Search, as a heuristic in Neural Network pruning, or to find energy-efficient offloading strategies in a Split computing context, or simply to evaluate the energetic performance of Deep Neural Network architectures.

cs.LG↗

arXiv2022

Energy Efficient Deployment and Orchestration of Computing Resources at the Network Edge: a Survey on Algorithms, Trends and Open Challenges

Mobile networks are becoming energy hungry, and this trend is expected to continue due to a surge in communication and computation demand. Multi-access Edge Computing (MEC), will entail energy-consuming services and applications, with non-negligible impact in terms of ecological sustainability. In this paper, we provide a comprehensive review of existing approaches to make edge computing networks greener, including but not limited to the exploitation of renewable energy resources, and context-awareness. We hence provide an updated account of recent developments on MEC from an energetic sustainability perspective, addressing the initial deployment of computing resources, their dynamic (re)allocation, as well as distributed and federated learning designs. In doing so, we highlight the energy aspects of these algorithms, advocating the need for energy-sustainable edge computing systems that are aligned with Sustainable Development Goals (SDGs) and the Paris agreement. To the best of our knowledge, this is the first work providing a systematic literature review on the efficient deployment and management of energy harvesting MEC, with special focus on the deployment, provisioning, and scheduling of computing tasks, including federated learning for distributed edge intelligence, toward making edge networks greener and more sustainable. At the end of the paper, open research avenues and challenges are identified for all the surveyed topics.

cs.NI↗

arXiv2022

SHED: A Newton-type algorithm for federated learning based on incremental Hessian eigenvector sharing

There is a growing interest in the distributed optimization framework that goes under the name of Federated Learning (FL). In particular, much attention is being turned to FL scenarios where the network is strongly heterogeneous in terms of communication resources (e.g., bandwidth) and data distribution. In these cases, communication between local machines (agents) and the central server (Master) is a main consideration. In this work, we present SHED, an original communication-constrained Newton-type (NT) algorithm designed to accelerate FL in such heterogeneous scenarios. SHED is by design robust to non i.i.d. data distributions, handles heterogeneity of agents' communication resources (CRs), only requires sporadic Hessian computations, and achieves super-linear convergence. This is possible thanks to an incremental strategy, based on eigendecomposition of the local Hessian matrices, which exploits (possibly) outdated second-order information. The proposed solution is thoroughly validated on real datasets by assessing (i) the number of communication rounds required for convergence, (ii) the overall amount of data transmitted and (iii) the number of local Hessian computations. For all these metrics, the proposed approach shows superior performance against state-of-the art techniques like GIANT and FedNL.

cs.LG↗

arXiv2022

EASE: Energy-Aware Job Scheduling for Vehicular Edge Networks With Renewable Energy Resources

The energy sustainability of multi-access edge computing (MEC) platforms is here addressed by developing Energy-Aware job Scheduling at the Edge (EASE), a computing resource scheduler for edge servers co-powered by renewable energy resources and the power grid. The scenario under study involves the optimal allocation and migration of time-sensitive computing tasks in a resource-constrained internet of vehicles (IoV) context. This is achieved by tackling, as the main objective, the minimization of the carbon footprint of the edge network, whilst delivering adequate quality of service (QoS) to the end users (e.g., meeting task execution deadlines). EASE integrates i) a centralized optimization step, solved through model predictive control (MPC), to manage the renewable energy that is locally collected at the edge servers and their local computing resources, estimating their future availability, and ii) a distributed consensus step, solved via dual ascent in closed form, to reach agreement on service migrations. EASE is compared with four existing migration strategies. Quantitative results demonstrate its greater energy efficiency, which often gets close to complete carbon neutrality, while also improving the QoS.

cs.NI↗

arXiv2022

A Review of Indoor Millimeter Wave Device-based Localization and Device-free Sensing Technologies and Applications

The commercial availability of low-cost millimeter wave (mmWave) communication and radar devices is starting to improve the penetration of such technologies in consumer markets, paving the way for large-scale and dense deployments in fifth-generation (5G)-and-beyond as well as 6G networks. At the same time, pervasive mmWave access will enable device localization and device-free sensing with unprecedented accuracy, especially with respect to sub-6 GHz commercial-grade devices. This paper surveys the state of the art in device-based localization and device-free sensing using mmWave communication and radar devices, with a focus on indoor deployments. We first overview key concepts about mmWave signal propagation and system design. Then, we provide a detailed account of approaches and algorithms for localization and sensing enabled by mmWaves. We consider several dimensions in our analysis, including the main objectives, techniques, and performance of each work, whether each research reached some degree of implementation, and which hardware platforms were used for this purpose. We conclude by discussing that better algorithms for consumer-grade devices, data fusion methods for dense deployments, as well as an educated application of machine learning methods are promising, relevant and timely research directions.

cs.NI↗

arXiv2022

An extension of polar duality of toric varieties and its consequences in Mirror Symmetry

The present paper is dedicated to illustrating an extension of polar duality between Fano toric varieties to a more general duality, called \emph{framed} duality, so giving rise to a powerful and unified method of producing mirror partners of hypersurfaces and complete intersections in toric varieties, of any Kodaira dimension. In particular, the class of projective hypersurfaces and their mirror partners are studied in detail. Moreover, many connections with known Landau-Ginzburg mirror models, Homological Mirror Symmetry and Intrinsic Mirror Symmetry, are discussed.

math.AG↗

arXiv2022

Human Tracking with mmWave Radars: a Deep Learning Approach with Uncertainty Estimation

mmWave radars have recently gathered significant attention as a means to track human movement within indoor environments. Widely adopted Kalman filter tracking methods experience performance degradation when the underlying movement is highly non-linear or presents long-term temporal dependencies. As a solution, in this article we design a convolutional-recurrent Neural Network (NN) that learns to accurately estimate the position and the velocity of the monitored subjects from high dimensional radar data. The NN is trained as a probabilistic model, utilizing a Gaussian negative log-likelihood loss function, obtaining explicit uncertainty estimates at its output, in the form of time-varying error covariance matrices. A thorough experimental assessment is conducted using a 77 GHz FMCW radar. The proposed architecture, besides allowing one to gauge the uncertainty in the tracking process, also leads to greatly improved performance against the best approaches from the literature, i.e., Kalman filtering, lowering the average error against the ground truth from 32.8 to 7.59 cm and from 56.8 to 14 cm/s in terms of position and velocity tracking, respectively.

eess.SP↗

arXiv2022

SPARCS: A Sparse Recovery Approach for Integrated Communication and Human Sensing in mmWave Systems

A well established method to detect and classify human movements using Millimeter-Wave ( mmWave) devices is the time-frequency analysis of the small-scale Doppler effect (termed micro-Doppler) of the different body parts, which requires a regularly spaced and dense sampling of the Channel Impulse Response ( CIR). This is currently done in the literature either using special-purpose radar sensors, or interrupting communications to transmit dedicated sensing waveforms, entailing high overhead and channel utilization. In this work we present SPARCS, an integrated human sensing and communication solution for mmWave systems. SPARCS is the first method that reconstructs high quality signatures of human movement from irregular and sparse CIR samples, such as the ones obtained during communication traffic patterns. To accomplish this, we formulate the micro-Doppler extraction as a sparse recovery problem, which is critical to enable a smooth integration between communication and sensing. Moreover, if needed, our system can seamlessly inject short CIR estimation fields into the channel whenever communication traffic is absent or insufficient for the micro-Doppler extraction. SPARCS effectively leverages the intrinsic sparsity of the mmWave channel, thus drastically reducing the sensing overhead with respect to available approaches. We implemented SPARCS on an IEEE 802.11ay Software Defined Radio (SDR) platform working in the 60 GHz band, collecting standard-compliant CIR traces matching the traffic patterns of real WiFi access points. Our results show that the micro-Doppler signatures obtained by SPARCS enable a typical downstream application such as human activity recognition with more than 7 times lower overhead with respect to existing methods, while achieving better recognition performance.

eess.SP↗

arXiv2022

DeepCSI: Rethinking Wi-Fi Radio Fingerprinting Through MU-MIMO CSI Feedback Deep Learning

We present DeepCSI, a novel approach to Wi-Fi radio fingerprinting (RFP) which leverages standard-compliant beamforming feedback matrices to authenticate MU-MIMO Wi-Fi devices on the move. By capturing unique imperfections in off-the-shelf radio circuitry, RFP techniques can identify wireless devices directly at the physical layer, allowing low-latency low-energy cryptography-free authentication. However, existing Wi-Fi RFP techniques are based on software-defined radio (SDRs), which may ultimately prevent their widespread adoption. Moreover, it is unclear whether existing strategies can work in the presence of MU-MIMO transmitters - a key technology in modern Wi-Fi standards. Conversely from prior work, DeepCSI does not require SDR technologies and can be run on any low-cost Wi-Fi device to authenticate MU-MIMO transmitters. Our key intuition is that imperfections in the transmitter's radio circuitry percolate onto the beamforming feedback matrix, and thus RFP can be performed without explicit channel state information (CSI) computation. DeepCSI is robust to inter-stream and inter-user interference being the beamforming feedback not affected by those phenomena. We extensively evaluate the performance of DeepCSI through a massive data collection campaign performed in the wild with off-the-shelf equipment, where 10 MU-MIMO Wi-Fi radios emit signals in different positions. Experimental results indicate that DeepCSI correctly identifies the transmitter with an accuracy of up to 98%. The identification accuracy remains above 82% when the device moves within the environment. To allow replicability and provide a performance benchmark, we pledge to share the 800 GB datasets - collected in static and, for the first time, dynamic conditions - and the code database with the community.

cs.NI↗

arXiv2021

MilliTRACE-IR: Contact Tracing and Temperature Screening via mm-Wave and Infrared Sensing

Social distancing and temperature screening have been widely employed to counteract the COVID-19 pandemic, sparking great interest from academia, industry and public administrations worldwide. While most solutions have dealt with these aspects separately, their combination would greatly benefit the continuous monitoring of public spaces and help trigger effective countermeasures. This work presents milliTRACE-IR, a joint mmWave radar and infrared imaging sensing system performing unobtrusive and privacy preserving human body temperature screening and contact tracing in indoor spaces. milliTRACE-IR combines, via a robust sensor fusion approach, mmWave radars and infrared thermal cameras. It achieves fully automated measurement of distancing and body temperature, by jointly tracking the subjects's faces in the thermal camera image plane and the human motion in the radar reference system. Moreover, milliTRACE-IR performs contact tracing: a person with high body temperature is reliably detected by the thermal camera sensor and subsequently traced across a large indoor area in a non-invasive way by the radars. When entering a new room, a subject is re-identified among several other individuals by computing gait-related features from the radar reflections through a deep neural network and using a weighted extreme learning machine as the final re-identification tool. Experimental results, obtained from a real implementation of milliTRACE-IR, demonstrate decimeter-level accuracy in distance/trajectory estimation, inter-personal distance estimation (effective for subjects getting as close as 0.2 m), and accurate temperature monitoring (max. errors of 0.5°C). Furthermore, milliTRACE-IR provides contact tracing through highly accurate (95%) person re-identification, in less than 20 seconds.

eess.SP↗

arXiv2021

Real-time People Tracking and Identification from Sparse mm-Wave Radar Point-clouds

Mm-wave radars have recently gathered significant attention as a means to track human movement and identify subjects from their gait characteristics. A widely adopted method to perform the identification is the extraction of the micro-Doppler signature of the targets, which is computationally demanding in case of co-existing multiple targets within the monitored physical space. Such computational complexity is the main problem of state-of-the-art approaches, and makes them inapt for real-time use. In this work, we present an end-to-end, low-complexity but highly accurate method to track and identify multiple subjects in real-time using the sparse point-cloud sequences obtained from a low-cost mm-wave radar. Our proposed system features an extended object tracking Kalman filter, used to estimate the position, shape and extension of the subjects, which is integrated with a novel deep learning classifier, specifically tailored for effective feature extraction and fast inference on radar point-clouds. The proposed method is thoroughly evaluated on an edge-computing platform from NVIDIA (Jetson series), obtaining greatly reduced execution times (reduced complexity) against the best approaches from the literature. Specifically, it achieves accuracies as high as 91.62%, operating at 15 frames per seconds, in identifying three subjects that concurrently and freely move in an unseen indoor environment, among a group of eight.

eess.SP↗

arXiv2021

Multi-Person Continuous Tracking and Identification from mm-Wave micro-Doppler Signatures

In this work, we investigate the use of backscattered mm-wave radio signals for the joint tracking and recognition of identities of humans as they move within indoor environments. We build a system that effectively works with multiple persons concurrently sharing and freely moving within the same indoor space. This leads to a complicated setting, which requires one to deal with the randomness and complexity of the resulting (composite) backscattered signal. The proposed system combines several processing steps: at first, the signal is filtered to remove artifacts, reflections and random noise that do not originate from humans. Hence, a density-based classification algorithm is executed to separate the Doppler signatures of different users. The final blocks are trajectory tracking and user identification, respectively based on Kalman filters and deep neural networks. Our results demonstrate that the integration of the last-mentioned processing stages is critical towards achieving robustness and accuracy in multi-user settings. Our technique is tested both on a single-target public dataset, for which it outperforms state-of-the-art methods, and on our own measurements, obtained with a 77 GHz radar on multiple subjects simultaneously moving in two different indoor environments. The system works in an online fashion, permitting the continuous identification of multiple subjects with accuracies up to 98%, e.g., with four subjects sharing the same physical space, and with a small accuracy reduction when tested with unseen data from a challenging real-life scenario that was not part of the model learning phase.

eess.SP↗

arXiv2020

Mobility and Blockage-aware Communications in Millimeter-Wave Vehicular Networks

Mobility may degrade the performance of next-generation vehicular networks operating at the millimeter-wave spectrum: frequent mis-alignment and blockages require repeated beam training and handover, and incur enormous overhead. Nevertheless, mobility induces temporal correlations in the communication beams and in blockage events. In this paper, an adaptive design of beam training, data transmission and handover is proposed, that learns and exploits these temporal correlations to reduce the beam training overhead and optimally trade-off throughput and power consumption. At each time-slot, the serving base station (BS) decides to perform either beam training, data communication, or handover when blockage is detected, under uncertainty in the system state. The decision problem is cast as a partially observable Markov decision process, and the goal is to maximize the throughput delivered to the UE, under an average power constraint. To address the high dimensional optimization, an approximate constrained point-based value iteration (C-PBVI) method is developed, which simultaneously optimizes the primal and dual functions to meet the power constraint. Numerical results demonstrate a good match between the analysis and a simulation based on 2D mobility and 3D analog beamforming via uniform planar arrays at both BSs and UE, and reveal that C-PBVI performs near-optimally, and outperforms a baseline scheme with periodic beam training by 38% in spectral efficiency. Motivated by the structure of the C-PBVI policy, two heuristics are proposed, that trade complexity with sub-optimality, and achieve only 4% and 15% loss in spectral efficiency.

cs.IT↗

arXiv2020

Toric varieties and Gröbner bases: the complete Q-factorial case

We present two algorithms determining all the complete and simplicial fans admitting a fixed non-degenerate set of vectors $V$ as generators of their 1-skeleton. The interplay of the two algorithms allows us to discerning if the associated toric varieties admit a projective embedding, in principle for any values of dimension and Picard number. The first algorithm is slower than the second one, but it computes all complete and simplicial fans supported by $V$ and lead us to formulate a topological-combinatoric conjecture about the definition of a fan. On the other hand, we adapt the Sturmfels' arguments on the Gröbner fan of toric ideals to our complete case; we give a characterization of the Gröbner region and show an explicit correspondence between Gröbner cones and chambers of the secondary fan. A homogenization procedure of the toric ideal associated to $V$ allows us to employing GFAN and related software in producing our second algorithm. The latter turns out to be much faster than the former, although it can compute only the projective fans supported by $V$. We provide examples and a list of open problems. In particular we give examples of rationally parametrized families of $\Q$-factorial complete toric varieties behaving in opposite way with respect to the dimensional jump of the nef cone over a special fibre.

math.AG↗

arXiv2020

Adaptive Millimeter-Wave Communications Exploiting Mobility and Blockage Dynamics

Mobility may degrade the performance of next-generation vehicular networks operating at the millimeter-wave spectrum: frequent loss of alignment and blockages require repeated beam training and handover, thus incurring huge overhead. In this paper, an adaptive and joint design of beam training, data transmission and handover is proposed, that exploits the mobility process of mobile users and the dynamics of blockages to optimally trade-off throughput and power consumption. At each time slot, the serving base station decides to perform either beam training, data communication, or handover when blockage is detected. The problem is cast as a partially observable Markov decision process, and solved via an approximate dynamic programming algorithm based on PERSEUS [2]. Numerical results show that the PERSEUS-based policy performs near-optimally, and achieves a 55% gain in spectral efficiency compared to a baseline scheme with periodic beam training. Inspired by its structure, an adaptive heuristic policy is proposed with low computational complexity and small performance degradation.

cs.IT↗

arXiv2020

Fibration and classification of smooth projective toric varieties of low Picard number

In this paper we show that a smooth toric variety $X$ of Picard number $r\leq 3$ always admits a nef primitive collection supported on a hyperplane admitting non-trivial intersection with the cone $\Nef(X)$ of numerically effective divisors and cutting a facet of the pseudo-effective cone $\Eff(X)$, that is $\Nef(X)\cap\partial\overline{\Eff}(X)\neq\{0\}$. In particular this means that $X$ admits non-trivial and non-big numerically effective divisors. Geometrically this guarantees the existence of a fiber type contraction morphism over a smooth toric variety of dimension and Picard number lower than those of $X$, so giving rise to a classification of smooth and complete toric varieties with $r\leq 3$. Moreover we revise and improve results of Oda-Miyake by exhibiting an extension of the above result to projective, toric, varieties of dimension $n=3$ and Picard number $r=4$, allowing us to classifying all these threefolds. We then improve results of Fujino-Sato, by presenting sharp (counter)examples of smooth, projective, toric varieties of any dimension $n\geq4$ and Picard number $r=4$ whose non-trivial nef divisors are big, that is $\Nef(X)\cap\partial\overline{\Eff}(X)=\{0\}$. Producing those examples represents an important goal of computational techniques in definitely setting an open geometric problem. In particular, for $n=4$, the given example turns out to be a weak Fano toric fourfold of Picard number 4.

math.AG↗