Michele Zorzi

Berivan Isik, Francesco Pase, Deniz Gunduz et al. · stanford

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.

Berivan Isik, Francesco Pase, Deniz Gunduz et al. · stanford

The high communication cost of sending model updates from the clients to the server is a significant bottleneck for scalable federated learning (FL). Among existing approaches, state-of-the-art bitrate-accuracy tradeoffs have been achieved using stochastic compression methods -- in which the client $n$ sends a sample from a client-only probability distribution $q_{φ^{(n)}}$, and the server estimates the mean of the clients' distributions using these samples. However, such methods do not take full advantage of the FL setup where the server, throughout the training process, has side information in the form of a global distribution $p_θ$ that is close to the clients' distribution $q_{φ^{(n)}}$ in Kullback-Leibler (KL) divergence. In this work, we exploit this closeness between the clients' distributions $q_{φ^{(n)}}$'s and the side information $p_θ$ at the server, and propose a framework that requires approximately $D_{KL}(q_{φ^{(n)}}|| p_θ)$ bits of communication. We show that our method can be integrated into many existing stochastic compression frameworks to attain the same (and often higher) test accuracy with up to $82$ times smaller bitrate than the prior work -- corresponding to 2,650 times overall compression.

Paolo Testolina, Francesco Barbato, Umberto Michieli et al.

Accurate scene understanding from multiple sensors mounted on cars is a key requirement for autonomous driving systems. Nowadays, this task is mainly performed through data-hungry deep learning techniques that need very large amounts of data to be trained. Due to the high cost of performing segmentation labeling, many synthetic datasets have been proposed. However, most of them miss the multi-sensor nature of the data, and do not capture the significant changes introduced by the variation of daytime and weather conditions. To fill these gaps, we introduce SELMA, a novel synthetic dataset for semantic segmentation that contains more than 30K unique waypoints acquired from 24 different sensors including RGB, depth, semantic cameras and LiDARs, in 27 different atmospheric and daytime conditions, for a total of more than 20M samples. SELMA is based on CARLA, an open-source simulator for generating synthetic data in autonomous driving scenarios, that we modified to increase the variability and the diversity in the scenes and class sets, and to align it with other benchmark datasets. As shown by the experimental evaluation, SELMA allows the efficient training of standard and multi-modal deep learning architectures, and achieves remarkable results on real-world data. SELMA is free and publicly available, thus supporting open science and research.

Nicolò Michelusi, Petar Popovski, Osvaldo Simeone et al.

This paper introduces a novel technique for access by a cognitive Secondary User (SU) using best-effort transmission to a spectrum with an incumbent Primary User (PU), which uses Type-I Hybrid ARQ. The technique leverages the primary ARQ protocol to perform Interference Cancellation (IC) at the SU receiver (SUrx). Two IC mechanisms that work in concert are introduced: Forward IC, where SUrx, after decoding the PU message, cancels its interference in the (possible) following PU retransmissions of the same message, to improve the SU throughput; Backward IC, where SUrx performs IC on previous SU transmissions, whose decoding failed due to severe PU interference. Secondary access policies are designed that determine the secondary access probability in each state of the network so as to maximize the average long-term SU throughput by opportunistically leveraging IC, while causing bounded average long-term PU throughput degradation and SU power expenditure. It is proved that the optimal policy prescribes that the SU prioritizes its access in the states where SUrx knows the PU message, thus enabling IC. An algorithm is provided to optimally allocate additional secondary access opportunities in the states where the PU message is unknown. Numerical results are shown to assess the throughput gain provided by the proposed techniques.

Francesco Pase, Marco Giordani, Giampaolo Cuozzo et al.

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.

Matteo Drago, Tommaso Zugno, Federico Mason et al.

Artificial intelligence (AI) techniques have emerged as a powerful approach to make wireless networks more efficient and adaptable. In this paper we present an ns-3 simulation framework, able to implement AI algorithms for the optimization of wireless networks. Our pipeline consists of: (i) a new geometry-based mobility-dependent channel model for V2X; (ii) all the layers of a 5G-NR-compliant protocol stack, based on the ns3-mmwave module; (iii) a new application to simulate V2X data transmission, and (iv) a new intelligent entity for the control of the network via AI. Thanks to its flexible and modular design, researchers can use this tool to implement, train, and evaluate their own algorithms in a realistic and controlled environment. We test the behavior of our framework in a Predictive Quality of Service (PQoS) scenario, where AI functionalities are implemented using Reinforcement Learning (RL), and demonstrate that it promotes better network optimization compared to baseline solutions that do not implement AI.

Francesco Pase, Deniz Gündüz, Michele Zorzi

We consider a rate-constrained contextual multi-armed 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 rate-constrained 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.

Nicola Marchetti, Irene Macaluso, Nicholas Kaminski et al.

We propose a new paradigm for telecommunications, and develop a framework drawing on concepts from information (i.e., different metrics of complexity) and computational (i.e., agent based modeling) theory, adapted from complex system science. We proceed in a systematic fashion by dividing network complexity understanding and analysis into different layers. Modelling layer forms the foundation of the proposed framework, supporting analysis and tuning layers. The modelling layer aims at capturing the significant attributes of networks and the interactions that shape them, through the application of tools such as agent-based modelling and graph theoretical abstractions, to derive new metrics that holistically describe a network. The analysis phase completes the core functionality of the framework by linking our new metrics to the overall network performance. The tuning layer augments this core with algorithms that aim at automatically guiding networks toward desired conditions. In order to maximize the impact of our ideas, the proposed approach is rooted in relevant, near-future architectures and use cases in 5G networks, i.e., Internet of Things (IoT) and self-organizing cellular networks.

Pietro Talli, Francesco Pase, Federico Chiariotti et al.

The coordination of robotic swarms and the remote wireless control of industrial systems are among the major use cases for 5G and beyond systems: in these cases, the massive amounts of sensory information that needs to be shared over the wireless medium can overload even high-capacity connections. Consequently, solving the effective communication problem by optimizing the transmission strategy to discard irrelevant information can provide a significant advantage, but is often a very complex task. In this work, we consider a prototypal system in which an observer must communicate its sensory data to an actor controlling a task (e.g., a mobile robot in a factory). We then model it as a remote Partially Observable Markov Decision Process (POMDP), considering the effect of adopting semantic and effective communication-oriented solutions on the overall system performance. We split the communication problem by considering an ensemble Vector Quantized Variational Autoencoder (VQ-VAE) encoding, and train a Deep Reinforcement Learning (DRL) agent to dynamically adapt the quantization level, considering both the current state of the environment and the memory of past messages. We tested the proposed approach on the well-known CartPole reference control problem, obtaining a significant performance increase over traditional approaches

Filippo Bragato, Tommaso Lotta, Gianmaria Ventura et al.

To ensure safety in teleoperated driving scenarios, communication between vehicles and remote drivers must satisfy strict latency and reliability requirements. In this context, Predictive Quality of Service (PQoS) was investigated as a tool to predict unanticipated degradation of the Quality of Service (QoS), and allow the network to react accordingly. In this work, we design a reinforcement learning (RL) agent to implement PQoS in vehicular networks. To do so, based on data gathered at the Radio Access Network (RAN) and/or the end vehicles, as well as QoS predictions, our framework is able to identify the optimal level of compression to send automotive data under low latency and reliability constraints. We consider different learning schemes, including centralized, fully-distributed, and federated learning. We demonstrate via ns-3 simulations that, while centralized learning generally outperforms any other solution, decentralized learning, and especially federated learning, offers a good trade-off between convergence time and reliability, with positive implications in terms of privacy and complexity.

Mattia Figaro, Francesco Rossato, Marco Giordani et al.

This paper overviews the 3GPP 5G NR-NTN standard, detailing the evolution from Rel. 18 to 19 and innovations for Rel. 20. Using realistic ns-3 simulations validated against 3GPP calibration data, we evaluate various satellite network configurations. The results highlight the potential of NTNs to extend wireless connectivity to remote areas, serve requests during emergency, and alleviate terrestrial network congestion.

Alexander Bonora, Marco Giordani, Michele Zorzi

The evolution of 6th generation (6G) networks increasingly relies on satellite-based Non-Terrestrial Networks (NTNs) to extend broadband connectivity to remote and unserved regions, and to support public safety. In this paper we compare two representative and conceptually different satellite constellation architectures, namely Starlink and IRIS 2. Starlink is a commercial private Internet constellation by SpaceX, based on dense Low Earth Orbit (LEO) satellites. It is primarily designed to deliver high-capacity broadband services for civil applications, with performance targets comparable to those of terrestrial networks. In contrast, IRIS 2 is a planned public initiative to be deployed by the European Union, based on a multi-layer combination of LEO, Medium Earth Orbit (MEO), and Geo-stationary Earth Orbit (GEO) satellites. It is primarily designed to provide a secure, resilient, and sovereign infrastructure for government and critical communications. After describing the main technical characteristics of Starlink and IRIS 2, we run a comprehensive simulation campaign to evaluate the design tradeoffs between the two. Specifically, we evaluate the per-cell and per-user achievable capacity, the impact of satellite mobility and handover, and identify the capability of each architecture to support global and reliable connectivity. We also provide design suggestions for possible future IRIS 2 deployment extensions.

Alessandro Traspadini, Michele Zorzi, Marco Giordani

The integration of Internet of Things (IoT) and Non-Terrestrial Networks (NTNs) has emerged as a key paradigm to provide connectivity for sensors and actuators via satellite gateways in remote areas where terrestrial infrastructure is limited or unavailable. Among other Low-Power Wide-Area Network (LPWAN) technologies for IoT, Long Range (LoRa) holds great potential given its long range, energy efficiency, and flexibility. In this paper, we explore the feasibility and performance of LoRa to support large-scale IoT connectivity through Low Earth Orbit (LEO) satellite gateways. To do so, we developed a new ns3-LoRa-NTN simulation module, which integrates and extends the ns3-LoRa and ns3-NTN modules, to enable full-stack end-to-end simulation of satellite communication in LoRa networks. Our results, given in terms of average data rate and Packet Reception Ratio (PRR), confirm that LoRa can effectively support direct communication from the ground to LEO satellites, but network optimization is required to mitigate collision probability when end nodes use the same Spreading Factors (SFs) over long distances.

Giacomo Avanzi, Marco Giordani, Michele Zorzi

Teleoperated driving (TD) is envisioned as a key application of future sixth generation (6G) networks. In this paradigm, connected vehicles transmit sensor-perception data to a remote (software) driver, which returns driving control commands to enhance traffic efficiency and road safety. This scenario imposes to maintain reliable and low-latency communication between the vehicle and the remote driver. To this aim, a promising solution is Predictive Quality of Service (PQoS), which provides mechanisms to estimate possible Quality of Service (QoS) degradation, and trigger timely network corrective actions accordingly. In particular, Reinforcement Learning (RL) agents can be trained to identify the optimal PQoS configuration. In this paper, we develop and implement two integrated RL agents that jointly determine (i) the optimal compression configuration for TD sensor data to balance the trade-off between transmission efficiency and data quality, and (ii) the optimal scheduling configuration to minimize the end-to-end latency by allocating radio resources according to different priority levels. We prove via full-stack ns-3 simulations that our integrated agents can deliver superior performance than any standalone model that only optimizes either compression or scheduling, especially in constrained or congested networks. While these agents can be deployed using either centralized or decentralized learning, we further propose a new meta-learning agent that dynamically selects the most appropriate strategy between the two based on current network conditions and application requirements.

Francesco Pase, Marco Giordani, Sara Cavallero et al.

Industrial Internet of Things (IIoT) networks will provide Ultra-Reliable Low-Latency Communication (URLLC) to support critical processes underlying the production chains. However, standard protocols for allocating wireless resources may not optimize the latency-reliability trade-off, especially for uplink communication. For example, centralized grant-based scheduling can ensure almost zero collisions, but introduces delays in the way resources are requested by the User Equipments (UEs) and granted by the gNB. In turn, distributed scheduling (e.g., based on random access), in which UEs autonomously choose the resources for transmission, may lead to potentially many collisions especially when the traffic increases. In this work we propose DIStributed combinatorial NEural linear Thompson Sampling (DISNETS), a novel scheduling framework that combines the best of the two worlds. By leveraging a feedback signal from the gNB and reinforcement learning, the UEs are trained to autonomously optimize their uplink transmissions by selecting the available resources to minimize the number of collisions, without additional message exchange to/from the gNB. DISNETS is a distributed, multi-agent adaptation of the Neural Linear Thompson Sampling (NLTS) algorithm, which has been further extended to admit multiple parallel actions. We demonstrate the superior performance of DISNETS in addressing URLLC in IIoT scenarios compared to other baselines.

Davide Cosimo, Davide Costa, Riccardo Costanzi et al.

This paper presents a modular software architecture that enables environmental-aware coordination of heterogeneous Autonomous Underwater Vehicles (AUVs) to improve underwater acoustic connectivity. The architecture combines a Robot Operating System 2 application layer with the DESERT Underwater communication framework through the rmw_desert middleware, and integrates a Robot Operating System 1 bridge to ensure interoperability with legacy vehicle front-seat controllers. This design enables fine-grained, cross-layer configurability of the communication stack and supports onboard processing of environmental measurements to inform adaptive communication behaviors. As a representative use case, this architecture is used to implement a lightweight depth-optimization strategy that exploits environmental awareness and AUV mobility to improve acoustic link performance. The complete software stack is validated through sea trials conducted off the Gulf of La Spezia in littoral water with an average depth of approximately 100m using a deployment involving three AUVs with distinct operational roles. Experimental results indicate that depth-adaptive repositioning yields measurable gains in packet reception at horizontal separation of approximately 1km, while differences are negligible at shorter ranges where the received signal energy remains above demodulation thresholds. Beyond link-level performance the sea trials confirm the feasibility, modularity, and practical deployability of the proposed architecture on existing AUV platforms.

Alessandro Traspadini, Matteo Pagin, Raphaël Ihamouine et al.

Millimeter wave (mmWave) 5th generation (5G) networks offer high data rates but face coverage challenges due to severe path loss and blockage. These problems motivate the use of Integrated Access and Backhaul (IAB) as a flexible wireless backhaul solution that extends connectivity to cell boundaries and unfibered areas, including maritime environments. This paper overviews the latest 3GPP specifications for IAB networks in Releases 16 through 18. Then, it presents an ns-3 module for IAB, featuring a complete end-to-end protocol stack, including the backhaul adaptation protocol (BAP) layer, flexible slot and control configurations, and multiplexing schemes based on both time and frequency division. We test the IAB module via extensive system-level simulations in a custom maritime scenario where vessels, equipped with IAB-nodes, can simultaneously act as access points and relays, forming dynamic multi-hop networks that maintain connectivity via wireless backhaul to shore-based stations. We evaluate different topologies and channel conditions, providing insights into the design and deployment of mmWave IAB networks in offshore environments.

Mattia Lecci, Matteo Drago, Andrea Zanella et al.

Thanks to recent advancements in the technology, eXtended Reality (XR) applications are gaining a lot of momentum, and they will surely become increasingly popular in the next decade. These new applications, however, require a step forward also in terms of models to simulate and analyze this type of traffic sources in modern communication networks, in order to guarantee to the users state of the art performance and Quality of Experience (QoE). Recognizing this need, in this work, we present a novel open-source traffic model, which researchers can use as a starting point both for improvements of the model itself and for the design of optimized algorithms for the transmission of these peculiar data flows. Along with the mathematical model and the code, we also share with the community the traces that we gathered for our study, collected from freely available applications such as Minecraft VR, Google Earth VR, and Virus Popper. Finally, we propose a roadmap for the construction of an end-to-end framework that fills this gap in the current state of the art.

Mattia Lecci, Andrea Zanella, Michele Zorzi

Next-generation wireless communication technologies will allow users to obtain unprecedented performance, paving the way to new and immersive applications. A prominent application requiring high data rates and low communication delay is Virtual Reality (VR), whose presence will become increasingly stronger in the years to come. To the best of our knowledge, we propose the first traffic model for VR applications based on traffic traces acquired from a commercial VR streaming software, allowing the community to further study and improve the technology to manage this type of traffic. This work implements ns-3 applications able to generate and process large bursts of packets, enabling the possibility of analyzing APP-level end-to-end metrics, making the source code as well as the acquired VR traffic traces publicly available and open-source.

Pietro Talli, Francesco Pase, Federico Chiariotti et al.

The remote wireless control of industrial systems is one of the major use cases for 5G and beyond systems: in these cases, the massive amounts of sensory information that need to be shared over the wireless medium may overload even high-capacity connections. Consequently, solving the effective communication problem by optimizing the transmission strategy to discard irrelevant information can provide a significant advantage, but is often a very complex task. In this work, we consider a prototypal system in which an observer must communicate its sensory data to a robot controlling a task (e.g., a mobile robot in a factory). We then model it as a remote Partially Observable Markov Decision Process (POMDP), considering the effect of adopting semantic and effective communication-oriented solutions on the overall system performance. We split the communication problem by considering an ensemble Vector Quantized Variational Autoencoder (VQ-VAE) encoding, and train a Deep Reinforcement Learning (DRL) agent to dynamically adapt the quantization level, considering both the current state of the environment and the memory of past messages. We tested the proposed approach on the well-known CartPole reference control problem, obtaining a significant performance increase over traditional approaches.

Federico Mason, Tommaso Zugno, Matteo Drago et al.

Predictive Quality of Service (PQoS) makes it possible to anticipate QoS changes, e.g., in wireless networks, and trigger appropriate countermeasures to avoid performance degradation. Hence, PQoS is extremely useful for automotive applications such as teleoperated driving, which poses strict constraints in terms of latency and reliability. A promising tool for PQoS is given by Reinforcement Learning (RL), a methodology that enables the design of decision-making strategies for stochastic optimization. In this manuscript, we present PRATA, a new simulation framework to enable PRedictive QoS based on AI for Teleoperated driving Applications. PRATA consists of a modular pipeline that includes (i) an end-to-end protocol stack to simulate the 5G Radio Access Network (RAN), (ii) a tool for generating automotive data, and (iii) an Artificial Intelligence (AI) unit to optimize PQoS decisions. To prove its utility, we use PRATA to design an RL unit, named RAN-AI, to optimize the segmentation level of teleoperated driving data in the event of resource saturation or channel degradation. Hence, we show that the RAN-AI entity efficiently balances the trade-off between QoS and Quality of Experience (QoE) that characterize teleoperated driving applications, almost doubling the system performance compared to baseline approaches. In addition, by varying the learning settings of the RAN-AI entity, we investigate the impact of the state space and the relative cost of acquiring network data that are necessary for the implementation of RL.

Francesco Pase, Szymon Kobus, Deniz Gunduz et al.

We consider the problem of communicating a sequence of concepts, i.e., unknown and potentially stochastic maps, which can be observed only through examples, i.e., the mapping rules are unknown. The transmitter applies a learning algorithm to the available examples, and extracts knowledge from the data by optimizing a probability distribution over a set of models, i.e., known functions, which can better describe the observed data, and so potentially the underlying concepts. The transmitter then needs to communicate the learned models to a remote receiver through a rate-limited channel, to allow the receiver to decode the models that can describe the underlying sampled concepts as accurately as possible in their semantic space. After motivating our analysis, we propose the formal problem of communicating concepts, and provide its rate-distortion characterization, pointing out its connection with the concepts of empirical and strong coordination in a network. We also provide a bound for the distortion-rate function.

Francesco Pase, Deniz Gunduz, Michele Zorzi

We consider a remote contextual multi-armed bandit (CMAB) problem, in which the decision-maker observes the context and the reward, but must communicate the actions to be taken by the agents over a rate-limited communication channel. This can model, for example, a personalized ad placement application, where the content owner observes the individual visitors to its website, and hence has the context information, but must convey the ads that must be shown to each visitor to a separate entity that manages the marketing content. In this remote CMAB (R-CMAB) problem, the constraint on the communication rate between the decision-maker and the agents imposes a trade-off between the number of bits sent per agent and the acquired average reward. We are particularly interested in characterizing the rate required to achieve sub-linear regret. Consequently, this can be considered as a policy compression problem, where the distortion metric is induced by the learning objectives. We first study the fundamental information theoretic limits of this problem by letting the number of agents go to infinity, and study the regret achieved when Thompson sampling strategy is adopted. In particular, we identify two distinct rate regions resulting in linear and sub-linear regret behavior, respectively. Then, we provide upper bounds on the achievable regret when the decision-maker can reliably transmit the policy without distortion.

Federico Mason, Matteo Drago, Tommaso Zugno et al.

In recent years, autonomous networks have been designed with Predictive Quality of Service (PQoS) in mind, as a means for applications operating in the industrial and/or automotive sectors to predict unanticipated Quality of Service (QoS) changes and react accordingly. In this context, Reinforcement Learning (RL) has come out as a promising approach to perform accurate predictions, and optimize the efficiency and adaptability of wireless networks. Along these lines, in this paper we propose the design of a new entity, implemented at the RAN-level that, with the support of an RL framework, implements PQoS functionalities. Specifically, we focus on the design of the reward function of the learning agent, able to convert QoS estimates into appropriate countermeasures if QoS requirements are not satisfied. We demonstrate via ns-3 simulations that our approach achieves the best trade-off in terms of QoS and Quality of Experience (QoE) performance of end users in a teleoperated-driving-like scenario, compared to other baseline solutions.

Mattia Lecci, Federico Chiariotti, Matteo Drago et al.

Over the past few years, eXtended Reality (XR) has attracted increasing interest thanks to its extensive industrial and commercial applications, and its popularity is expected to rise exponentially over the next decade. However, the stringent Quality of Service (QoS) constraints imposed by XR's interactive nature require Network Slicing (NS) solutions to support its use over wireless connections: in this context, quasi-Constant Bit Rate (CBR) encoding is a promising solution, as it can increase the predictability of the stream, making the network resource allocation easier. However, traffic characterization of XR streams is still a largely unexplored subject, particularly with this encoding. In this work, we characterize XR streams from more than 4 hours of traces captured in a real setup, analyzing their temporal correlation and proposing two prediction models for future frame size. Our results show that even the state-of-the-art H.264 CBR mode can have significant frame size fluctuations, which can impact the NS optimization. Our proposed prediction models can be applied to different traces, and even to different contents, achieving very similar performance. We also show the trade-off between network resource efficiency and XR QoS in a simple NS use case.

Mate Boban, Marco Giordani, Michele Zorzi

Recent advances in software, hardware, computing and control have fueled significant progress in the field of autonomous systems. Notably, autonomous machines should continuously estimate how the scenario in which they move and operate will evolve within a predefined time frame, and foresee whether or not the network will be able to fulfill the agreed Quality of Service (QoS). If not, appropriate countermeasures should be taken to satisfy the application requirements. Along these lines, in this paper we present possible methods to enable predictive QoS (PQoS) in autonomous systems, and discuss which use cases will particularly benefit from network prediction. Then, we shed light on the challenges in the field that are still open for future research. As a case study, we demonstrate whether machine learning can facilitate PQoS in a teleoperated-driving-like use case, as a function of different measurement signals.

Valentina Rossi, Paolo Testolina, Marco Giordani et al.

Fully autonomous driving systems require fast detection and recognition of sensitive objects in the environment. In this context, intelligent vehicles should share their sensor data with computing platforms and/or other vehicles, to detect objects beyond their own sensors' fields of view. However, the resulting huge volumes of data to be exchanged can be challenging to handle for standard communication technologies. In this paper, we evaluate how using a combination of different sensors affects the detection of the environment in which the vehicles move and operate. The final objective is to identify the optimal setup that would minimize the amount of data to be distributed over the channel, with negligible degradation in terms of object detection accuracy. To this aim, we extend an already available object detection algorithm so that it can consider, as an input, camera images, LiDAR point clouds, or a combination of the two, and compare the accuracy performance of the different approaches using two realistic datasets. Our results show that, although sensor fusion always achieves more accurate detections, LiDAR only inputs can obtain similar results for large objects while mitigating the burden on the channel.

Francesco Pase, Marco Giordani, Michele Zorzi

Federated learning (FL) has recently emerged as an attractive decentralized solution for wireless networks to collaboratively train a shared model while keeping data localized. As a general approach, existing FL methods tend to assume perfect knowledge of the Channel State Information (CSI) during the training phase, which may not be easy to acquire in case of fast fading channels. Moreover, literature analyses either consider a fixed number of clients participating in the training of the federated model, or simply assume that all clients operate at the maximum achievable rate to transmit model data. In this paper, we fill these gaps by proposing a training process that takes channel statistics as a bias to minimize the convergence time under imperfect CSI. Numerical experiments demonstrate that it is possible to reduce the training time by neglecting model updates from clients that cannot sustain a minimum predefined transmission rate. We also examine the trade-off between number of clients involved in the training process and model accuracy as a function of different fading regimes.

Federico Venturini, Federico Mason, Francesco Pase et al.

Over the past few years, the use of swarms of Unmanned Aerial Vehicles (UAVs) in monitoring and remote area surveillance applications has become widespread thanks to the price reduction and the increased capabilities of drones. The drones in the swarm need to cooperatively explore an unknown area, in order to identify and monitor interesting targets, while minimizing their movements. In this work, we propose a distributed Reinforcement Learning (RL) approach that scales to larger swarms without modifications. The proposed framework relies on the possibility for the UAVs to exchange some information through a communication channel, in order to achieve context-awareness and implicitly coordinate the swarm's actions. Our experiments show that the proposed method can yield effective strategies, which are robust to communication channel impairments, and that can easily deal with non-uniform distributions of targets and obstacles. Moreover, when agents are trained in a specific scenario, they can adapt to a new one with minimal additional training. We also show that our approach achieves better performance compared to a computationally intensive look-ahead heuristic.

Mattia Lecci, Paolo Testolina, Mattia Rebato et al.

With the advent of millimeter wave (mmWave) communications, the combination of a detailed 5G network simulator with an accurate antenna radiation model is required to analyze the realistic performance of complex cellular scenarios. However, due to the complexity of both electromagnetic and network models, the design and optimization of antenna arrays is generally infeasible due to the required computational resources and simulation time. In this paper, we propose a Machine Learning framework that enables a simulation-based optimization of the antenna design. We show how learning methods are able to emulate a complex simulator with a modest dataset obtained from it, enabling a global numerical optimization over a vast multi-dimensional parameter space in a reasonable amount of time. Overall, our results show that the proposed methodology can be successfully applied to the optimization of thinned antenna arrays.

Paolo Testolina, Mattia Lecci, Mattia Rebato et al.

Complex phenomena are generally modeled with sophisticated simulators that, depending on their accuracy, can be very demanding in terms of computational resources and simulation time. Their time-consuming nature, together with a typically vast parameter space to be explored, make simulation-based optimization often infeasible. In this work, we present a method that enables the optimization of complex systems through Machine Learning (ML) techniques. We show how well-known learning algorithms are able to reliably emulate a complex simulator with a modest dataset obtained from it. The trained emulator is then able to yield values close to the simulated ones in virtually no time. Therefore, it is possible to perform a global numerical optimization over the vast multi-dimensional parameter space, in a fraction of the time that would be required by a simple brute-force search. As a testbed for the proposed methodology, we used a network simulator for next-generation mmWave cellular systems. After simulating several antenna configurations and collecting the resulting network-level statistics, we feed it into our framework. Results show that, even with few data points, extrapolating a continuous model makes it possible to estimate the global optimum configuration almost instantaneously. The very same tool can then be used to achieve any further optimization goal on the same input parameters in negligible time.

Michele Polese, Rittwik Jana, Velin Kounev et al.

The fifth generation of cellular networks (5G) will rely on edge cloud deployments to satisfy the ultra-low latency demand of future applications. In this paper, we argue that such deployments can also be used to enable advanced data-driven and Machine Learning (ML) applications in mobile networks. We propose an edge-controller-based architecture for cellular networks and evaluate its performance with real data from hundreds of base stations of a major U.S. operator. In this regard, we will provide insights on how to dynamically cluster and associate base stations and controllers, according to the global mobility patterns of the users. Then, we will describe how the controllers can be used to run ML algorithms to predict the number of users in each base station, and a use case in which these predictions are exploited by a higher-layer application to route vehicular traffic according to network Key Performance Indicators (KPIs). We show that the prediction accuracy improves when based on machine learning algorithms that rely on the controllers' view and, consequently, on the spatial correlation introduced by the user mobility, with respect to when the prediction is based only on the local data of each single base station.

Davide Zordan, Michele Rossi, Michele Zorzi

Many future wireless sensor networks and the Internet of Things are expected to follow a software defined paradigm, where protocol parameters and behaviors will be dynamically tuned as a function of the signal statistics. New protocols will be then injected as a software as certain events occur. For instance, new data compressors could be (re)programmed on-the-fly as the monitored signal type or its statistical properties change. We consider a lossy compression scenario, where the application tolerates some distortion of the gathered signal in return for improved energy efficiency. To reap the full benefits of this paradigm, we discuss an automatic sensor profiling approach where the signal class, and in particular the corresponding rate-distortion curve, is automatically assessed using machine learning tools (namely, support vector machines and neural networks). We show that this curve can be reliably estimated on-the-fly through the computation of a small number (from ten to twenty) of statistical features on time windows of a few hundreds samples.