Michele Zorzi’s research while affiliated with University of Padua and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (93)


Unmanned Marine Vehicle Cooperative Offshore Infrastructure Monitoring with Multimodal Underwater Feature Transmission
  • Conference Paper

June 2024

·

6 Reads

Alessia Ortile

·

·

Federico Chiariotti

·

Michele Zorzi



Fig. 3: Cumulative distribution of the per-user uplink and downlink throughput for different pilot assignment strategies for a small-scale scenario, M = 50, K = 12, L = 1, and τ p = 3.
Pilot Reuse in Cell-Free Massive MIMO Systems: A Diverse Clustering Approach
  • Preprint
  • File available

December 2022

·

27 Reads

Distributed or Cell-free (CF) massive Multiple-Input, Multiple-Output (mMIMO), has been recently proposed as an answer to the limitations of the current network-centric systems in providing high-rate ubiquitous transmission. The capability of providing uniform service level makes CF mMIMO a potential technology for beyond-5G and 6G networks. The acquisition of accurate Channel State Information (CSI) is critical for different CF mMIMO operations. Hence, an uplink pilot training phase is used to efficiently estimate transmission channels. The number of available orthogonal pilot signals is limited, and reusing these pilots will increase co-pilot interference. This causes an undesirable effect known as pilot contamination that could reduce the system performance. Hence, a proper pilot reuse strategy is needed to mitigate the effects of pilot contamination. In this paper, we formulate pilot assignment in CF mMIMO as a diverse clustering problem and propose an iterative maxima search scheme to solve it. In this approach, we first form the clusters of User Equipments (UEs) so that the intra-cluster diversity maximizes and then assign the same pilots for all UEs in the same cluster. The numerical results show the proposed techniques' superiority over other methods concerning the achieved uplink and downlink average and per-user data rate.

Download

Rate-Constrained Remote Contextual Bandits

December 2022

·

3 Reads

·

11 Citations

IEEE Journal on Selected Areas in Information Theory

We consider a rate-constrained contextual multiarmed bandit (RC-CMAB) problem, in which a group of agents are solving the same contextual multi-armed bandit (CMAB) problem. However, the contexts are observed by a remotely connected entity, i.e., the decision-maker, that updates the policy to maximize the returned rewards, and communicates the arms to be sampled by the agents to a controller over a rate-limited communications channel. This framework can be applied to personalized ad placement, whenever the content owner observes the website visitors, and hence has the context, but needs to transmit the ads to be shown to a controller that is in charge of placing the marketing content. Consequently, the rateconstrained CMAB (RC-CMAB) problem requires the study of lossy compression schemes for the policy to be employed whenever the constraint on the channel rate does not allow the uncompressed transmission of the decision-maker’s intentions. We characterize the fundamental information theoretic limits of this problem by letting the number of agents go to infinity, and study the regret that can be achieved, identifying the two distinct rate regions leading to linear and sub-linear regrets respectively. We then analyze the optimal compression scheme achievable in the limit with infinite agents, when using the forward and reverse KL divergence as distortion metric. Based on this, we also propose a practical coding scheme, and provide numerical results.


Distributed Resource Allocation for URLLC in IIoT Scenarios: A Multi-Armed Bandit Approach

November 2022

·

25 Reads

·

Marco Giordani

·

Giampaolo Cuozzo

·

[...]

·

Michele Zorzi

This paper addresses the problem of enabling inter-machine Ultra-Reliable Low-Latency Communication (URLLC) in future 6G Industrial Internet of Things (IIoT) networks. As far as the Radio Access Network (RAN) is concerned, centralized pre-configured resource allocation requires scheduling grants to be disseminated to the User Equipments (UEs) before uplink transmissions, which is not efficient for URLLC, especially in case of flexible/unpredictable traffic. To alleviate this burden, we study a distributed, user-centric scheme based on machine learning in which UEs autonomously select their uplink radio resources without the need to wait for scheduling grants or preconfiguration of connections. Using simulation, we demonstrate that a Multi-Armed Bandit (MAB) approach represents a desirable solution to allocate resources with URLLC in mind in an IIoT environment, in case of both periodic and aperiodic traffic, even considering highly populated networks and aggressive traffic.


Figure 3: Accuracy and bitrate comparison of FedPM with baselines SignSGD (Bernstein et al., 2018), TernGrad (Wen et al., 2017), QSGD (Alistarh et al., 2017), DRIVE (Vargaftik et al., 2021), EDEN (Vargaftik et al., 2022), and FedMask (Li et al., 2021), all performing in the same bitrate regime, on CIFAR-10, CIFAR-100, MNIST, and EMNIST datasets. Note that the bitrate in QSGD is adjustable via the number of levels. In these plots, we pick a level number that gives bitrate slightly larger than 1. However, the accuracy is still lower than FedPM accuracy. It is possible to reduce the bitrate in QSGD and still have some reasonable accuracy; however, the performance in that regime (lower bitrate, lower accuracy) is not a meaningful comparison point for FedPM.
Figure 4: Distributed mean estimation scheme in FedPM, modified for differential privacy.
Architectures for CONV-4, CONV-6, and CONV-10 models used in the experiments.
Sparse Random Networks for Communication-Efficient Federated Learning

September 2022

·

83 Reads

·

2 Citations

One main challenge in federated learning is the large communication cost of exchanging weight updates from clients to the server at each round. While prior work has made great progress in compressing the weight updates through gradient compression methods, we propose a radically different approach that does not update the weights at all. Instead, our method freezes the weights at their initial \emph{random} values and learns how to sparsify the random network for the best performance. To this end, the clients collaborate in training a \emph{stochastic} binary mask to find the optimal sparse random network within the original one. At the end of the training, the final model is a sparse network with random weights -- or a subnetwork inside the dense random network. We show improvements in accuracy, communication (less than 1 bit per parameter (bpp)), convergence speed, and final model size (less than 1 bpp) over relevant baselines on MNIST, EMNIST, CIFAR-10, and CIFAR-100 datasets, in the low bitrate regime under various system configurations.





Citations (45)


... Several low-cost and low-power acoustic underwater modems have been developed over the last years. Reviews of these modems are given in [3,4]. Recent examples of other modems are the Nanomodem [19], FAU modem [9], or Xiamen University modem [5]. ...

Reference:

Integration of the ahoi Modem into the MOLA Seafloor Lander System: Concept, Realization and Real-World Trials
Affordable underwater acoustic modems and their application in everyday life: a complete overview
  • Citing Conference Paper
  • June 2024

... In [27], the authors considered an agent sending observation information to another agent using a SC system to guide it to optimize its behavior. In [28], a remotely connected agent transmitted the SI of the observations, and analyzed the trade-off between the accuracy of the SI representation of actions and the system reward. However, the works in [27] and [28] did not consider the cooperation among multiple agents. ...

Rate-Constrained Remote Contextual Bandits
  • Citing Article
  • December 2022

IEEE Journal on Selected Areas in Information Theory

... Their proposal focuses on the awareness of each node of the network of the behavior of the network itself. In addition, the authors in [1], [2] designed a security mechanism for UANs based on the introduction of a "watchdog" layer within each node of the network that overhears neighboring transmissions and keeps track of which ones are successful or still have to occur. This insight is then used to update a local table that stores reputation values for each surrounding node, which translate to one of four possible reputation states. ...

A Secure Cross-Layer Communication Stack for Underwater Acoustic Networks

... The authors study the delayed noisy rewards with no prior information on the delay distribution, and develop a new algorithm based on Linear UCB (LinUCB) algorithm. The remote control of contextual bandits has also been investigated in [10]- [12]. The work in [10] assumes that the delay is a fixed, known, and constant, and proves it has an additive effect on the regret. ...

Remote Contextual Bandits
  • Citing Conference Paper
  • June 2022

... However, data processing based on machine learning (from compression and object detection and recognition to tracking and trajectory prediction) requires extensive computational resources, which may be challenging for GVs [10]. In an urban scenario, GVs can offload data to roadside units for Vehicular Edge Computing (VEC) [11] but, in poorly connected rural areas, NTNs emerge as a viable alternative [12]. To address this research, in [13] we proposed a framework to optimize data offloading via NTNs, focusing on HAPs. ...

UAV/HAP-Assisted Vehicular Edge Computing in 6G: Where and What to Offload?
  • Citing Conference Paper
  • June 2022

... Using the module presented in [12], it is possible to simulate 5G mmWave networks, analysing several key performance indicators (KPIs). Adhoc simulators can be developed, such as the one presented in [13] and used to study URLLC use cases in IIoT scenarios. In all the aforementioned works, 5G systems are examined in terms of KPIs, occasionally referencing requirements such as those defined by 5G-ACIA in [5]; however, none of them has addressed an experimental industrial use case with its own specific requirements. ...

Enabling URLLC in 5G NR IIoT Networks: A Full-Stack End-to-End Analysis
  • Citing Conference Paper
  • June 2022

... In [67], the paper tackles energy efficiency maximization in an UL RIS-aided mmWave NOMA network by jointly optimizing the UE TX power and the RIS phase shifts using iterative optimization techniques. In [68], the usage of both RISs and UAV is discussed in terms of providing energy-efficient communications on the DL while ensuring the QoS demands of the UEs. The proposed method uses Successive Convex Approximation (SCA) to iteratively determine a joint optimal solution for the UAV trajectory, RIS phase shifts, and the UAV TX power. ...

Energy-Efficient Design for RIS-assisted UAV communications in beyond-5G Networks
  • Citing Conference Paper
  • June 2022

... The deployment of large MIMO arrays in UAV-assisted communication for 6G systems presents several challenges due to the substantial load and size of these systems [15]. These challenges primarily revolve around addressing the limited power resources and stringent constraints on energy consumption, making ground base station (BS) deployment more practical in practice. ...

On the beamforming design of millimeter wave UAV networks: Power vs. capacity trade-offs
  • Citing Article
  • January 2022

Computer Networks

... In addition, this layer may also facilitate specific managementlevel and network-exposure-level capabilities to allow a tenant and a third-party to execute certain tasks within the proposed framework [137]. In a scenario where the provisioning of an NS crosses the boundaries of a network operator (such as in global coverage or public-private network infrastructure services), it is essential to equip the Business Management Layer of a network operator with the necessary capabilities to aggregate and compose resources, features, and capabilities from different Business Management Layer instances belonging to different 6G network operators [138]. ...

6G Drivers for B2B Market
  • Citing Chapter
  • October 2021