Goal-oriented Communication (GoC) is a new paradigm that activates data transmission only when it is instrumental for the receiver to achieve a certain goal. This leads to the advantage of reducing the frequency of transmissions significantly while maintaining adherence to the receiver’s objectives. However, GoC scheduling also opens a timing-based side channel that an eavesdropper can exploit to estimate the state of the system. This type of attack sidesteps even information-theoretic security, as it exploits the timing of updates rather than their content. In this work, we study such an eavesdropping attack against pull-based goal-oriented scheduling for remote monitoring and control of Markov processes. We provide a theoretical framework for defining the effectiveness of the attack and propose possible countermeasures, including three heuristics that provide a balance between the performance gains offered by GoC and the amount of leaked information. Our results show that, while a naive GoC scheduler allows the eavesdropper to correctly guess the system state about 60% of the time, our heuristic defenses can halve the leakage with a marginal reduction of the benefits of goal-oriented approaches.

Goal-oriented communication entails the timely transmission of updates related to a specific goal defined by the application. In a distributed setup with multiple sensors, each individual sensor knows its own observation and can determine its freshness, as measured by Age of Incorrect Information (AoII). This local knowledge is suited for distributed medium access, where the transmission strategies have to deal with collisions. We present Dynamic Epistemic Logic for Tracking Anomalies (DELTA), a medium access protocol that limits collisions and minimizes AoII in anomaly reporting over dense networks. Each sensor knows its own AoII, while it can compute the belief about the AoII for all other sensors,based on their Age of Information (AoI), which is inferred from the acknowledgments. This results in a goal-oriented approach based on dynamic epistemic logic emerging from public information. We analyze the resulting DELTA protocol both from a theoretical standpoint and with Monte Carlo simulations, showing that it is significantly more efficient and robust than classical random access, while outperforming state-of-the-art scheduled schemes by at least 30%, even with imperfect feedback.

Artificial Intelligence (AI) is a key component of$\mathbf{6 G}$networks, as it enables communication and computing services to adapt to end users' requirements and demand patterns. The management of Mobile Edge Computing (MEC) is a meaningful example of AI application: computational resources available at the network edge need to be carefully allocated to users, whose jobs may have different priorities and latency requirements. The research community has developed several AI algorithms to accomplish this goal, but it has neglected a key aspect: learning is itself a computationally demanding task, and considering free training results in idealized conditions and performance in simulations. In this work, we consider a more realistic framework that explicitly accounts for the cost of learning, presenting a new algorithm to dynamically select when to train a Deep Reinforcement Learning (DRL) agent that allocates resources in a MEC facility. Our method is highly general, as it can be directly applied to any scenario involving a training overhead, and it can approach the same performance as an ideal learning agent even under realistic training conditions.

Age of Incorrect Information (AoII) is particularly relevant in systems where real time responses to anomalies are required, such as natural disaster alerts, cybersecurity warnings, or medical emergency notifications. Keeping system control with wrong information for too long can lead to inappropriate responses. In this paper, we study the Peak AoII (PAoII) for multisource status reporting by independent devices over a collision channel, following a zero-threshold ALOHA access where nodes observing an anomaly immediately start transmitting about it. If a collision occurs, nodes reduce the transmission probability to allow for a resolution. Finally, wrong or lost feedback messages may lead a node that successfully updated the destination to believe a collision happened. The PAoII for this scenario is computed in closed-form. We are eventually able to derive interesting results concerning the minimization of PAoII, which can be traded against the overall goodput and energy efficiency, but may push the system to the edge of congestion collapse.

This paper investigates how to achieve both low-power operations of sensor nodes and accurate state estimation using Kalman filter for internet of things (IoT) monitoring employing wireless sensor networks under radio resource constraint. We consider two policies used by the base station to collect observations from the sensor nodes: (i) an oblivious policy, based on statistics of the observations, and (ii) a decentralized policy, based on autonomous decision of each sensor based on its instantaneous observation. This work introduces a wake-up receiver and wake-up signaling to both policies to improve the energy efficiency of the sensor nodes. The decentralized policy designed with random access prioritizes transmissions of instantaneous observations that are highly likely to contribute to the improvement of state estimation. Our numerical results show that the decentralized policy improves the accuracy of the estimation in comparison to the oblivious policy under the constraint on the radio resource and consumed energy when the correlation between the processes observed by the sensor nodes is low. We also clarify the degree of correlation in which the superiority of two policies changes.

With the unprecedented diffusion of virtual reality, the number of application scenarios is continuously growing. As commercial and gaming applications become pervasive, the need for the secure and convenient identification of users, often overlooked by the research in immersive media, is becoming more and more pressing. Networked scenarios such as Cloud gaming or cooperative virtual training and teleoperation require both a user-friendly and streamlined experience and user privacy and security. In this work, we investigate the possibility of identifying users from their movement patterns and data traffic traces while playing four commercial games, using a publicly available dataset. If, on the one hand, this paves the way for easy identification and automatic customization of the virtual reality content, it also represents a serious threat to users’ privacy due to network analysis-based fingerprinting. Based on this, we analyze the threats and opportunities for virtual reality users’ security and privacy.

Internet of Things (IoT) devices provide constant, contextual data that can be leveraged to automatically reconfigure and optimize smart environments. Artificial Intelligence (AI) and deep learning techniques are tools of increasing importance for this, as Deep Reinforcement Learning (DRL) can provide a general solution to this problem. However, the heterogeneity of scenarios in which DRL models may be deployed is vast, making the design of universal plug-and-play models extremely difficult. Moreover, the real deployment of DRL models on the Edge, and in the IoT in particular, is limited by two factors: firstly, the computational complexity of the training procedure, and secondly, the need for a relatively long exploration phase, during which the agent proceeds by trial and error. A natural solution to both these issues is to use simulated environments by creating a Digital Twin (DT) of the environment, which can replicate physical entities in the digital domain, providing a standardized interface to the application layer. DTs allow for simulation and testing of models and services in a simulated environment, which may be hosted on more powerful Cloud servers without the need to exchange all the data generated by the real devices. In this paper, we present a novel architecture based on the emerging Web of Things (WoT) standard, which provides a DT of a smart environment and applies DRL techniques on real time data. We discuss the theoretical properties of DRL training using DTs, showcasing our system in an existing real deployment, comparing its performance with a legacy system. Our findings show that the implementation of a DT, specifically for DRL models, allows for faster convergence and finer tuning, as well as reducing the computational and communication demands on the Edge network. The use of multiple DTs with different complexities and data requirements can also help accelerate the training, progressing by steps. • Integration of Web of Things and Digital Twins for seamless DRL training. • Performance evaluation on a real dataset, showing the benefits of digital twins. • Legacy interoperability is demonstrated with off the shelf smart home software.

Sensors are increasingly being deployed to monitor critical infrastructure. However, as the number of sensors being deployed increases, so does the amount of sensor data that must be transmitted, stored, and analyzed. Thus, a significant number of methods have been proposed to improve sensor data acquisition and analytics. However, the proposed strategies and methods generally focus exclusively on either sensor data acquisition or analytics, thus ignoring the possible optimization that can be performed by taking a holistic view. To explore this opportunity, this paper provides an overview of sensor data acquisition and analytics and an analysis of two very different use cases, specifically monitoring wind turbines and measuring utility consumption using smart meters. Based on this analysis, the Framework for joint Sensory Data Acquisition and Analytics (SENDAI) is proposed, an integrated framework that models sensor data acquisition and analytics together, thus enabling holistic reasoning about sensor data acquisition and analytics. To demonstrate how the information in SENDAI can be used to reason about sensor data acquisition and analytics together, we show how sensor data acquisition can be optimized to respond efficiently to query workloads.

Pragmatic or goal-oriented communication can optimize communication decisions beyond the reliable transmission of data, instead aiming at directly affecting application performance with the minimum channel utilization. In this paper, we develop a general theoretical framework for the remote control of finite-state Markov processes, using pragmatic communication over a costly zero-delay communication channel. To that end, we model a cyber-physical system composed of an encoder, which observes and transmits the states of a process in real-time, and a decoder, which receives that information and controls the behavior of the process. The encoder and the decoder should cooperatively optimize the trade-off between the control performance (i.e., reward) and the communication cost (i.e., channel use). This scenario underscores a pragmatic (i.e., goal-oriented) communication problem, where the purpose is to convey only the data that is most valuable for the underlying task, taking into account the state of the decoder (hence, the pragmatic aspect). We investigate two different decision-making architectures: in pull-based remote control, the decoder is the only decision-maker, while in push-based remote control, the encoder and the decoder constitute two independent decision-makers, leading to a multi-agent scenario. We propose three algorithms to optimize our system (i.e., design the encoder and the decoder policies), discuss the optimality guarantees ofs the algorithms, and shed light on their computational complexity and fundamental limits.

Energy efficiency and information freshness are key requirements for sensor nodes serving Industrial Internet of Things (IIoT) applications, where a sink node collects informative and fresh data before a deadline, e.g., to control an external actuator. Content-based wake-up (CoWu) activates a subset of nodes that hold data relevant for the sink’s goal, thereby offering an energy-efficient way to attain objectives related to information freshness. This paper focuses on a scenario where the sink collects fresh information on top-kvalues, defined as data from the nodes observing thekhighest readings at the deadline. We introduce a new metric called top-kQuery Age of Information (k-QAoI), which allows us to characterize the performance of CoWu by considering the characteristics of the physical process. Further, we show how to select the CoWu parameters, such as its timing and threshold, to attain both information freshness and energy efficiency. The numerical results reveal the effectiveness of the CoWu approach, which is able to collect top-kdata with higher energy efficiency while reducingk-QAoI when compared to round-robin scheduling, especially when the number of nodes is large and the required size ofkis small.

The recently introduced QUIC protocol has greatly increased the flexibility of end-to-end transmissions on the Internet, surpassing the design limits of the most popular transport protocols: TCP and UDP. However, some of TCP’s main design principles were carried onto QUIC, which may not be suitable for real-time use-cases; primarily, full reliability, as it requires every packet to be retransmitted until acknowledged by the receiver. In this work, we present Dynamic Reliability (DRel), a partial reliability framework that allows for granular alteration of the reliability per packet at the transport layer. The framework is housed by QUIC and its multipath extension, yet offering no-ack and no-retransmit for the true meaning of unreliable packet transmission (congestion control does not impact and is not influenced by unreliable packets). The “dynamic” in DRel refers to interchangeable reliable and unreliable transmission, in one session and across multiple paths, depending on the volatility of the communication system; guided by reliability policies. Fluidly altering packet reliability may offer a means in meeting stringent 5G and Beyond transmission requirements, especially for xURLLC use-cases. We examine the performance of DRel in single- and multiple-path architectures through system-level simulation using Mininet. The results illustrate comparable performance to vital QoE metrics for the dynamic reliability policies compared to the original (MP)QUIC. Alternatively, the enhancements at the transport layer stem from a reduction in communication congestion by up to 80% for single- and multiple-path connections compared to the original (MP)QUIC. As a result, the amount of backlogged and out-of-order packets is reduced, downsizing intermediate and end-to-end buffer occupancies.

The rapid development of Virtual Reality (VR) technology has led the industry and research community to look at its major challenges with increased interest. The main challenge in ensuring a high Quality of Experience (QoE) for users is represented by cybersickness, a phenomenon similar to motion sickness experienced by many VR users, while at the same time, the high data rates needed by VR require the definition of traffic models for network optimization. These two problems are intertwined, but have never been studied jointly before due to the lack of suitable datasets. In this paper, we present Questset, the first dataset designed for this purpose. Questset contains over 40 hours of VR traces from 70 users playing commercially available video games, and includes both traffic data for network optimization, and movement and user experience data for cybersickness analysis. Therefore, Questset represents an enabler to jointly address the main VR challenges in the near future.

The use of Wake-Up Radio (WUR) in Internet of Things (IoT) networks can significantly improve their energy efficiency: battery-powered sensors can remain in a low-power (sleep) mode while listening for wake-up messages using their WUR and reactivate only when polled. However, polling-based WUR may still lead to wasted energy if values sensed by the polled sensors provide no new information to the receiver, or in general have a low Value of Information (VoI). In this paper, we present Wake-Up with Awareness of VoI and Energy (WAVE), a scheme that combines the benefits of ID- and content-based WUR techniques by adapting to the update VoI. We analyze the trade-off between the tracking error and the battery lifetime of the sensors, showing that WAVE can provide fine-grained control of this trade-off and significantly increase the battery lifetime of the node with a minimal Mean Squared Error (MSE) increase.

Low Earth orbit (LEO) satellite mega-constellations with hundreds or thousands of satellites and inter-satellite links (ISLs) have the potential to provide global end-to-end connectivity. Furthermore, if the physical distance between source and destination is sufficiently long, end-to-end routing over the LEO constellation can provide lower latency when compared to the terrestrial infrastructure due to the faster propagation of electromagnetic waves in space than in optic fiber. However, the frequent route changes due to the movement of the satellites result in the out-of-order delivery of packets, causing sudden changes to the Round-Trip Time (RTT) that can be misinterpreted as congestion by congestion control algorithms. In this paper, the performance of three widely used congestion control algorithms, Cubic, Reno, and BBR, is evaluated in an emulated LEO satellite constellation with Free-Space Optical (FSO) ISLs. Furthermore, we perform a sensitivity analysis for Cubic by changing the satellite constellation parameters, length of the routes, and the positions of the source and destination to identify problematic routing scenarios. The results show that route changes can have profound transient effects on the goodput of the connection, posing problems for typical broadband applications.

Mobile Edge Computing (MEC) is expected to play a significant role in the development of 6G networks, as new applications such as cooperative driving and eXtended Reality (XR) require both communication and computational resources from the network edge. However, the limited capabilities of edge servers may be strained to perform complex computational tasks within strict latency bounds for multiple clients. In these contexts, both maintaining a low expected Age of Information (AoI) and guaranteeing a low Peak AoI (PAoI) even in the worst case may have significant user experience and safety implications. In this work, we investigate a theoretical model of a MEC server, deriving the expected AoI and the PAoI and latency distributions under the First In First Out (FIFO) and Generalized Processor Sharing (GPS) resource allocation policies. We consider both synchronized and unsynchronized systems, and draw insights on the robust design of resource allocation policies from the analytical results, as well as considering the dimensioning of MEC capabilities in two realistic 6G use cases.

Age of Information (AoI) is a metric often used to represent the freshness of the information exchanged between a sensing source and a receiver. We consider a system where these two nodes are connected through an error-prone time-slotted channel, and a relay node is also present to assist the transmission. We consider both the sensor and the relay as intermittently and independently active nodes, whose activity rate may be adjusted, resulting in different levels of freshness and corresponding energy costs. To this end, the activity pattern can either follow a Bernoulli random process or a periodic duty cycle with adjustable duration. After computing the expected AoI and the complete Peak Age of Information (PAoI) distribution for both cases, we consider a fully distributed game theoretic duty cycle optimization, in which the two nodes independently tune their own activity rate, finding a balance between freshness and cost. The equilibrium of the resulting game is found to be both efficient from the perspective of the resulting performance and computationally lightweight for a distributed robust control implementation.

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 theeffective communicationproblem 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.

Over the past few years, the concept of Virtual Reality (VR) has attracted increasing interest thanks to its extensive industrial and commercial applications. Currently, the 3D models of the virtual scenes are generally stored in the VR visor itself, which operates as a standalone device. However, applications that entail multi-party interactions will likely require the scene to be processed by an external server and then streamed to the visors. However, the stringent Quality of Service (QoS) constraints imposed by the VR's interactive nature require Network Slicing (NS) solutions, for which profiling the traffic generated by the VR application is crucial. To this end, we collected more than 4 hours of traces in a real setup and analyzed their temporal correlation, focusing on the CBR encoding mode, which should generate more predictable traffic streams. From the collected data, we then distilled two prediction models for future frame size, which can be instrumental in the design of dynamic resource allocation algorithms. Our results show that even the state-of-the-art H.264 CBR mode may have significant frame size fluctuations, impacting NS optimization. We then exploited the models to dynamically determine requirements in an NS scenario, providing the required QoS while minimizing resource usage.

The recent rise of the Cloud Virtual Reality (VR) paradigm, in which VR frames are streamed from a remote server to the user's Head-Mounted Display (HMD), poses some interesting challenges from a networking perspective. VR flows are high-throughput and have strict latency requirements. In this paper, we analyze on-the-fly edge transcoding, which follows the opposite philosophy from the more common slicing approach: instead of adapting resource allocation to the content, we compress the frames to fit into the allocated bandwidth, guaranteeing limited latency. Our results show that this strategy is effective in maintaining low latency, but the picture quality is highly dependent on the computing power of the Base Station (BS).

In the past few years, Deep Reinforcement Learning (DRL) has become a valuable solution to automatically learn efficient resource management strategies in complex networks. In many scenarios, the learning task is performed in the Cloud, while experience samples are generated directly by edge nodes or users. Therefore, the learning task involves some data exchange which, in turn, subtracts a certain amount of transmission resources from the system. This creates a friction between the need to speed up convergence towards an effective strategy, which requires the allocation of resources to transmit learning samples, and the need to maximize the amount of resources used for data plane communication, maximizing users' Quality of Service (QoS), which requires the learning process to be efficient, i.e., minimize its overhead. In this paper, we investigate this trade-off and propose a dynamic balancing strategy between the learning and data planes, which allows the centralized learning agent to quickly converge to an efficient resource allocation strategy, while minimizing the impact on QoS. Simulation results show that the proposed method outperforms static allocation methods, converging to the optimal policy (i.e., maximum efficacy and minimum overhead of the learning plane) in the long run.

The integration of Low Earth Orbit (LEO) satellite constellations into 5G and Beyond is essential to achieve efficient global connectivity. As LEO satellites are a global infrastructure with predictable dynamics, a pre-planned fair and load-balanced allocation of the radio resources to provide efficient downlink connectivity over large areas is an achievable goal. In this paper, we propose a distributed and a global optimal algorithm for satellite-to-cell resource allocation with multiple beams. These algorithms aim to achieve a fair allocation of time-frequency resources and beams to the cells based on the number of users in connected mode (i.e., registered). Our analyses focus on evaluating the trade-offs between average per-user throughput, fairness, number of cell handovers, and computational complexity in a downlink scenario with fixed cells, where the number of users is extracted from a population map. Our results show that both algorithms achieve a similar average per-user throughput. However, the global optimal algorithm achieves a fairness index over 0.9 in all cases, which is more than twice that of the distributed algorithm. Furthermore, by correctly setting the handover cost parameter, the number of handovers can be effectively reduced by more than 70% with respect to the case where the handover cost is not considered.

The Age of Incorrect Information (AoII) is a metric that can combine the freshness of the information available to a gateway in an Internet of Things (IoT) network with the accuracy of that information. As such, minimizing the AoII can allow the operators of IoT systems to have a more precise and up-to-date picture of the environment in which the sensors are deployed. However, most IoT systems do not allow for centralized scheduling or explicit coordination, as sensors need to be extremely simple and consume as little power as possible. Finding a decentralized policy to minimize the AoII can be extremely challenging in this setting. This paper presents a heuristic to optimize AoII for a slotted ALOHA system, starting from a threshold-based policy and using dual methods to converge to a better solution. This method can significantly outperform state-independent policies, finding an efficient balance between frequent updates and a low number of packet collisions.

The use of Reconfigurable Intelligent Surface (RIS) technology to extend coverage and allow for better control of the wireless environment has been proposed in several use cases, including Ultra-Reliable Low-Latency Communications (URLLC) communications. However, the extremely challenging latency constraint makes explicit channel estimation difficult, so positioning information is often used to configure the RIS and illuminate the receiver device. In this work, we analyze the effect of imperfections in the positioning information on the reliability, deriving an upper bound to the outage probability. We then use this bound to perform power control, efficiently finding the minimum power that respects the URLLC constraints under positioning uncertainty. The optimization is conservative, so that all points respect the URLLC constraints, and the bound is relatively tight, with an optimality gap between 1.5 and 4.5 dB.

Taking inspiration from linguistics, the communications theoretical community has recently shown a significant recent interest inpragmatic, or goal-oriented, communication. In this paper, we tackle the problem of pragmatic communication with multiple clients with different, and potentially conflicting, objectives. We capture the goal-oriented aspect through the metric of Value of Information (VoI), which considers the estimation of the remote process as well as the timing constraints. However, the most common definition of VoI is simply the Mean Square Error (MSE) of the whole system state, regardless of the relevance for a specific client. Our work aims to overcome this limitation by including different summary statistics, i.e., value functions of the state, for separate clients, and a diversified query process on the client side, expressed through the fact that different applications may request different functions of the process state at different times. A query-aware Deep Reinforcement Learning (DRL) solution based on statically defined VoI can outperform naive approaches by 15-20%.

The stringent timing and reliability requirements in mission-critical applications require a detailed statistical characterization of end-to-end latency. Teleoperation is a representative use case, in which a human operator (HO) remotely controls a robot by exchanging command and feedback signals. We present a framework to analyze the latency of a closed-loop teleoperation system consisting of three entities: an HO, a robot located in remote environment, and a Base Station (BS) with Mobile edge Computing (MEC) capabilities. A model of each component is used to analyze the closed-loop latency and optimize the compression strategy. The closed-form expression of the distribution of the closed-loop latency is difficult to estimate, such that suitable upper and lower bounds are obtained. We formulate a non-convex optimization problem to minimize the closed-loop latency. Using the obtained upper and lower bound on the closed-loop latency, a computationally efficient procedure to optimize the closed-loop latency is presented. The simulation results reveal that compression of sensing data is not always beneficial, while system design based on average performance leads to under-provisioning and may cause performance degradation. The applicability of the proposed analysis is much wider than teleoperation, including a large class of systems whose latency budget consists of many components.

The use of parallel communication paths to provide reliable, low-latency service is a significant trend in cellular networks, as it can provide a way to satisfy the exacting Quality of Service (QoS) requirements of 5G-enabled applications. In particular, coding data across multiple paths can significantly improve reliability and reduce overall latency, compensating for stragglers and lost packets with the redundant information from other paths. However, the design trade-offs in optimizing these systems are non-trivial, particularly when considering Age of Information (AoI). In this work, we derive the latency and Peak Age of Information (PAoI) distributions for such a multipath coded system, drawing design insights on how to optimize either. While preemption is always the optimal choice to minimize AoI in a single-path, uncoded queuing system, the trade-off in this case is more complex, as dropping a late packet on one path might affect the reliability of the whole block. Our results show that the parameters to minimize the PAoI lead to poor latency performance, and optimizing both at once might require significant resource overprovisioning.

Over the last few years, the Deep Reinforcement Learning (DRL) paradigm has been widely adopted for 5G and beyond network optimization because of its extreme adaptability to many different scenarios. However, collecting and processing learning data entail a significant cost in terms of communication and computational resources, which is often disregarded in the networking literature. In this work, we analyze the cost of learning in a resource-constrained system, defining an optimization problem in which training a DRL agent makes it possible to improve the resource allocation strategy but also reduces the number of available resources. Our simulation results show that the cost of learning can be critical when evaluating DRL schemes on the network edge and that assuming a cost-free learning model can lead to significantly overestimating performance.

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.

With the advent of 5G technology, the notion oflatencygot a prominent role in wireless connectivity, serving as a proxy term for addressing the requirements for real-time communication. As wireless systems evolve toward 6G, the ambition to immerse the digital into physical reality will increase. Besides making the real-time requirements more stringent, this immersion will bring the notions of time, simultaneity, presence, and causality to a new level of complexity. A growing body of research points out that latency is insufficient to parameterize all real-time requirements. Notably, one such requirement that received significant attention is information freshness, defined through the Age of Information (AoI) and its derivatives. In general, the metrics derived from a conventional black-box approach to communication network design are not representative of new distributed paradigms, such as sensing, learning, or distributed consensus. The objective of this article is to investigate the general notion of timing in wireless communication systems and networks, and its relation to effective information generation, processing, transmission, and reconstruction at the senders and receivers. We establish a general statistical framework oftimingrequirements in wireless communication systems, which subsumes both latency and AoI. The framework is made by associating a timing component with the two basic statistical operations: decision and estimation. We first use the framework to present a representative sample of the existing works that deal with timing in wireless communication. Next, it is shown how the framework can be used with different communication models of increasing complexity, starting from the basic Shannon one-way communication model and arriving at communication models for consensus, distributed learning, and inference. Overall, this article fills an important gap in the literature by providing a systematic treatment of various timing measures in wireless communication and sets the basis for design and optimization for the next-generation real-time systems.

The evolution of 5G and Beyond networks has enabled new applications with stringent end-to-end latency requirements, but providing reliable low-latency service with high throughput over public wireless networks is still a significant challenge. One of the possible ways to solve this is to exploit path diversity, encoding the information flow over multiple streams across parallel links. The challenge presented by this approach is the design of joint coding and scheduling algorithms that adapt to the state of links to take full advantage of path diversity. In this paper, we address this problem for a synchronous traffic source that generates data blocks at regular time intervals (e.g., a video with constant frame rate) and needs to deliver each block within a predetermined deadline. We first develop a closed-form performance analysis in the simple case of two parallel servers without any buffering and single-packet blocks, and propose a model for the general problem based on a Markov Decision Process (MDP). We apply policy iteration to obtain the coding and scheduling policy that maximizes the fraction of source blocks delivered within the deadline: our simulations show the drawbacks of different commonly applied heuristic solutions, drawing general design insights on the optimal policy.

Age of Information (AoI) has become an important concept in communications, as it allows system designers to measure the freshness of the information available to remote monitoring or control processes. However, its definition tacitly assumes that new information is used at any time, which is not always the case: the instants at which information is collected and used may be dependent on a certain query process, and resource-constrained environments such as most Internet of Things (IoT) use cases require precise timing to fully exploit the limited available transmissions. In this work, we consider apull-based communication modelin which the freshness of information is only important when the receiver generates a query: if the monitoring process is not using the value, the age of the last update is irrelevant. We optimize the Age of Information at Query (QAoI), a metric that samples the AoI at relevant instants, better fitting the pull-based resource-constrained scenario, and show how this can lead to very different choices. Our results show that QAoI-aware optimization can significantly reduce the average and worst-case perceived age for both periodic and stochastic queries.

Multipath communication is a critical technology to provide Quality of Service (QoS) to interactive and Internet of Things (IoT) monitoring and control applications. In this work, we model the exemplary case with two paths and consider different strategies that exploit redundancy and coding to improve the timing performance of wireless communications. We consider two disparate scenarios, in which the data blocks are generated via a Markovian and a deterministic process, respectively. We consider simple scheduling and coding schemes, considering both lossless and lossy encoding, and modeling the resulting process as a fork-join queue with different arrival processes. We analyze the full distribution of two relevant metrics for the two-path case: the packet delay and the Peak Age of Information (PAoI), which measures the freshness of the information at the receiver. The results show interesting trade-offs between the update frequency, latency, PAoI, and level of compression, with interesting implications for system designers.

Age of Information (AoI) has gained significant attention from the research community because of its applications to Internet of Things (IoT) monitoring and control. In this work, we treat multihop connections over queuing networks with tributary flows and non-preemptive service: packets cannot be discarded because they are utilized for other system objectives, such as data analytics. Without preemption, the key tool for optimizing AoI is then the scheduling policy between the different data flows at each intermediate node. This is the subject of our analysis, along with the impact of packet erasure on the age. We derive upper and lower bounds for the average AoI considering several queuing policies in arbitrary network topologies, and present the results in different scenarios. Network topology, tributary traffic load, and link characteristics such as packet erasure generate complex trade-offs, which affect the optimal operation point and the age performance. The scheduling strategy at each node can also affect performance and fairness among users, particularly at critical bottleneck links, which have a significant impact on the overall performance of the whole network.

Over the last few years, the many uses of Unmanned Aerial Vehicles (UAVs) have captured the interest of both the scientific and the industrial communities. A typical scenario consists in the use of UAVs for surveillance or target-search missions over a wide geographical area. In this case, it is fundamental for the command center to accurately estimate and track the trajectories of the UAVs by exploiting their periodic state reports. In this work, we design anad hoctracking system that exploits the Long Range Wide Area Network (LoRaWAN) standard for communication and an extended version of the Constant Turn Rate and Acceleration (CTRA) motion model to predict drone movements in a 3D environment. We analyze the trade-off in setting the main parameters of the communication system and Adaptive Data Rate (ADR) scheme, showing how our tracking system can handle large swarms of drones at distances up to 4 km. Simulation results on a publicly available dataset show that our system can reliably estimate the position and trajectory of a swarm of UAVs, significantly outperforming baseline tracking approaches.

Security is a critical consideration in Underwater Acoustic Networks (UANs) due to the importance of the applications in which these types of networks are often employed, from military applications to marine natural disaster prevention. Furthermore, even simple Denial of Service (DoS) attacks such as jamming can be very effective in disrupting the communication, with significant negative consequences for these critical applications. While jamming has been widely studied in the context of terrestrial networks, the peculiarities of propagation in UANs, such as the low propagation speed, the multipath, and the high delay spread, need to be considered: the relative positions of the jammer, transmitter, and receiver can have a huge impact on the feasibility and impact of reactive jamming, opening the way for the exploitation of other jamming models. In this paper, we analyze the effectiveness of a reactive and a blind jammer through a game theoretical framework, comparing them for different geometries of the scenario. We assess the impact of the different jammers, employing different active and evasive strategies, where the first type of countermeasure implies the use of additional energy to protect the communication, while the second tries to avoid the jamming signals by randomizing the transmission pattern.

Age of Information (AoI) has become an important concept in communications, as it allows system designers to measure the freshness of the information available to remote monitoring or control processes. However, its definition tacitly assumes that new information is used at any time, which is not always the case. Instead instants at which information is collected and used are dependent on a certain query process. We propose a model that accounts for the discrete time nature of many monitoring processes, by considering a pull-based communication model in which the freshness of information is only important when the receiver generates a query. We then define the Age of Information at Query (QAoI), a more general metric that fits the pull-based scenario, and show how its optimization can lead to very different choices from traditional push-based AoI optimization when using a Packet Erasure Channel (PEC).

The fifth generation (5G) of wireless systems has a platform-driven approach, aiming to support heterogeneous connections with very diverse requirements. The shared wireless resources should be sliced in a way that each user perceives that its requirements have been met. Heterogeneity challenges the traditional notion of resource efficiency, as the resource usage has to cater for, e.g., rate maximization for one user and a timeliness requirement for another user. This paper treats a model for radio access network (RAN) uplink, where a throughput-demanding broadband user shares wireless resources with an intermittently active user that wants to optimize the timeliness, expressed in terms of latency-reliability or Age of Information (AoI). We evaluate the trade-offs between throughput and timeliness for Orthogonal Multiple Access (OMA) as well as Non-Orthogonal Multiple Access (NOMA) with successive interference cancellation (SIC). We observe that NOMA with SIC, in a conservative scenario with destructive collisions, is just slightly inferior to that of OMA, which indicates that it may offer significant benefits in practical deployments where the capture effect is frequently encountered. On the other hand, finding the optimal configuration of NOMA with SIC depends on the activity pattern of the intermittent user, to which OMA is insensitive.

Age of Information (AoI) is a critical metric for several Internet of Things (IoT) applications, where sensors keep track of the environment by sending updates that need to be as fresh as possible. The development of edge computing solutions has moved the monitoring process closer to the sensor, reducing the communication delays, but the processing time of the edge node needs to be taken into account. Furthermore, a reliable system design in terms of freshness requires the knowledge of the full distribution of the Peak AoI (PAoI), from which the probability of occurrence of rare, but extremely damaging events can be obtained. In this work, we model the communication and computation delay of such a system as two First Come First Serve (FCFS) queues in tandem, analytically deriving the full distribution of the PAoI for the M/M/1 - M/D/1 and the M/M/1 - M/M/1 tandems, which can represent a wide variety of realistic scenarios.

Next-generation wireless networks are expected to enable new applications with strict latency constraints. However, existing transport layer protocols are unable to meet the stringent Quality of Service (QoS) requirements on throughput and maximum latency: excessive queuing due to capacity-oriented congestion control inflates end-to-end latency well beyond interactivity deadlines. In this work, we propose a novel framework that evolves best-effort communications into reliability- and latency-aware communications for QoS-sensitive applications. The new protocol, named High-reliability latency-bounded Overlay Protocol (HOP), provides a novel combination of packet-level Forward Error Correction (FEC) and multipath scheduling to compensate for capacity drops and meet pre-defined QoS requirements. More specifically, the sender splits the data and the associated redundancy between the paths by using a stochastic forecast of their future capacity and decides the amount of redundancy necessary to meet the application’s requirements without clogging the connections. We compare HOP’s performance with state-of-the-art multipath protocols in ns-3 simulations using both synthetic and live network traces, and confirm that our scheme can reliably deliver high-throughput data, reducing the number of late blocks by 2 to 5 times with respect to optimized Multipath TCP (MPTCP).

Age of Information is a new metric used in real-time status update tracking applications. It measures at the destination the time elapsed since the generation of the last received packet. In this paper, we consider the co-existence of critical and noncritical status updates in a two-hop system, for which the network assigns different scheduling priorities. Specifically, the high priority is reserved to the packets that traverse the two nodes, as they experience worse latency performance. We obtain the distribution of the age and its natural upper bound termed peak age. We provide tight upper and lower bounds for priority updates and the exact expressions for the non-critical flow of packets with a general service distribution. The results give fundamental insights for the design of age-sensitive multi-hop systems.

The recently proposed QUIC protocol has been widely adopted at the transport layer of the Internet over the past few years. Its design goals are to overcome some of TCP's performance issues, while maintaining the same properties and basic application interface. Two of the main drivers of its success were the integration with the innovative Bottleneck Bandwidth and Round-trip propagation time (BBR) congestion control mechanism, and the possibility of multiplexing different application streams over the same connection. Given the strong interest in QUIC shown by the ns-3 community, we present an extension to the native QUIC module that allows researchers to fully explore the potential of these two features. In this work, we present the integration of BBR into the QUIC module and the implementation of the necessary pacing and rate sampling mechanisms, along with a novel scheduling interface, with three different scheduling flavors. The new features are tested to verify that they perform as expected, using a web traffic model from the literature.

Security aspects in underwater wireless networks have not been widely investigated so far, despite the critical importance of the scenarios in which these networks can be employed. For example, an attack to a military underwater network for enemy targeting or identification can lead to serious consequences. Similarly, environmental monitoring applications, such as tsunami prevention, are also critical from a public safety point of view. In this article, we assess a scenario in which a malicious node tries to perform a jamming attack, degrading the communication quality of battery-powered underwater nodes. The legitimate transmitter may use packet-level coding to increase the chances of correctly delivering packets. Because of the energy limitation of the nodes, the jammer's objective is twofold: 1) disrupting the communication and 2) reducing the lifetime of the victim by making it send more redundancy. We model the jammer and the transmitter as players in a multistage game, deriving the optimal strategies. We evaluate the performance both in a model-based scenario and using real experimental data, and perform a sensitivity analysis to evaluate the performance of the strategies if the real channel model is different from the one they use.

Bike-sharing systems have become an established reality in cities all across the world and are a key component of the Smart City paradigm. However, the unbalanced traffic patterns during rush hours can completely empty some stations, while filling others, and the service becomes unavailable for further users. The traditional approach to solve this problem is to use rebalancing trucks, which take bikes from full stations and deposit them at empty ones, reducing the likelihood of system outages. Another paradigm that is gaining steam is gamification, i.e., incentivizing users to fix the system by influencing their behavior with rewards and prizes. In this work, we combine the two efforts and show that a joint optimization considering both rebalancing and incentives results in a higher service quality for a lower cost than using simple rebalancing. We use simulations based on the New York CitiBike usage data to validate our model and analyze several schemes to optimize the bike-sharing system.

Jamming may become a serious threat in Internet of Things networks of battery-powered nodes, as attackers can disrupt packet delivery and significantly reduce the lifetime of the nodes. In this work, we model an active defense scenario in which an energy-limited node uses power control to defend itself from a malicious attacker, whose energy constraints may not be known to the defender. The interaction between the two nodes is modeled as an asymmetric Bayesian game where the victim has incomplete information about the attacker. We show how to derive the optimal Bayesian strategies for both the defender and the attacker, which may then serve as guidelines to develop and gauge efficient heuristics that are less computationally expensive than the optimal strategies. For example, we propose a neural-network-based learning method that allows the node to effectively defend itself from the jamming with a significantly reduced computational load. The outcomes of the ideal strategies highlight the tradeoff between node lifetime and communication reliability and the importance of an intelligent defense from jamming attacks.

In this work, we face the issue of achieving an efficient dynamic mapping in vehicular networking scenarios, i.e., obtaining an accurate estimate of the positions and trajectories of connected vehicles in a certain area. State-of-the-art solutions are based on the periodic broadcasting of the position information of the network nodes, with an inter-transmission period set by a congestion control scheme. However, the movements and maneuvers of vehicles can often be erratic, making transmitted data inaccurate or downright misleading. To address this problem, we propose to adopt a dynamic transmission scheme based on the actual positioning error, sending new data when the estimate overcomes a preset error threshold. Furthermore, the proposed method adapts the error threshold to the operational context according to an innovative congestion control algorithm that limits the collision probability among broadcast packet transmissions. This threshold-based strategy can reduce the network load by avoiding the transmission of redundant messages, and is shown to improve the overall positioning accuracy by more than 20% in realistic urban scenarios.

In recent years, the use of drones to map environments and survey them for items of interest such as forest fires, landslides or wild animals has gained traction in various research communities. However, the need for a human pilot or a pre-planned flight path severely limits the effectiveness of the drones, especially when a whole swarm is used. In this work, we propose a model of the drone survey problem and apply three well-known reinforcement learning strategies, showing that the performance loss due to the lack of explicit optimization and pre-programmed knowledge of the system statistics is negligible in the swarm scenario.

As the capacity and reliability of mobile networks increases, so does the demand for more responsive end-to-end services: applications such as augmented reality, live video conferencing, and smart or autonomous vehicles require reliable, throughput-intensive end-to-end communications with strict delay constraints. Only consistently reliable delivery of data flows well within human interactivity deadlines will enable a truly immersive user experience. To enable data delivery within pre-defined deadlines, controlled on demand by an application or its user, we propose and demonstrate a novel transport-layer protocol for explicit latency control called latency-controlled end-to-end aggregation protocol (LEAP). The LEAP splits a data flow with quality of service (QoS) constraints into multiple subflows that are delivered over multiple parallel links (e.g., Wi-Fi and LTE in a standard smartphone, WiGig, and 5G in the near future). The subflow data rates are set based on a novel proactive forecasting of the achievable channel capacity, subject to application-specific QoS constraints. Cross-path encoding and redundancy adaptation are then used to deliberately balance the trade-off between maximum throughput, required delay, and minimum reliability as function of application/user-specific input parameters. When compared to leading state-of-the-art transport protocols in live network experiments, LEAP exhibits a superior capacity to reliably provide a high and stable throughput with bounded latency, both in wired and wireless scenarios. The LEAP is also the first protocol to allow applications to explicitly set their priorities, giving them the freedom to set the operating point in the trade-off between throughput, latency, and reliability.

So far, research on Smart Cities and self-organizing networking techniques for fifth-generation (5G) cellular systems has been one-sided: a Smart City relies on 5G to support massive machine-to-machine (M2M) communications, but the actual network is unaware of the information flowing through it. However, a greater synergy between the two would make the relationship mutual, since the insights provided by the massive amount of data gathered by sensors can be exploited to improve the communication performance. In this paper, we concentrate on self-organization techniques to improve handover efficiency using vehicular traffic data gathered in London. Our algorithms exploit mobility patterns between cell coverage areas and road traffic congestion levels to optimize the handover bias in heterogeneous networks and dynamically manage mobility management entity (MME) loads to reduce handover completion times.

The role of mobile devices, like smartphones or tablets, is becoming more and more important in everyday life, at the point that their unavailability due to early or unexpected battery discharge is perceived as a serious issue. Therefore, there is an urge for smart and efficient battery management algorithms that can prolong the duration of the battery charge. To this end, a reliable prediction of the battery discharging process would represent a precious tool to enable energy-efficiency optimization mechanisms. In this paper, we address this challenge by considering different machine learning techniques to provide an accurate and user-dependent prediction of the discharging time of a mobile device and, eventually, we propose a Deep Neural Network model that provides the best performance. Unlike previous solutions proposed in the literature, our method exploits space-time data from the device operating system (Android) to learn the specific battery usage pattern of the user, thus offering a customized prediction of the discharge process. We show that such model outperforms the other machine-learning methods considered in this study, and achieves much better performance than the deterministic linear fitting methods widely used in commercial devices.

In this work, we propose an online adaptation logic for Dynamic Adaptive Streaming over HTTP (DASH) clients, where each client selects the representation that maximize the long term expected reward. The latter is defined as a combination of the decoded quality, the quality fluctuations and the rebuffering events experienced by the user during the playback. To solve this problem, we cast a Markov Decision Process (MDP) optimization for the selection of the optimal representations. System dynamics required in the MDP model are a priori unknown and are therefore learned through a Reinforcement Learning (RL) technique. The developed learning process exploits a parallel learning technique that improves the learning rate and limits sub-optimal choices, leading to a fast and yet accurate learning process that quickly converges to high and stable rewards. Therefore, the efficiency of our controller is not sacrificed for fast convergence. Simulation results show that our algorithm achieves a higher QoE than existing RL algorithms in the literature as well as heuristic solutions, as it is able to increase average QoE and reduce quality fluctuations.

We consider a multi-stage Bayesian game to model the interaction between an adaptive video streaming client and a congested network adopting a token-based policy for QoS provisioning. The Bayesian type of the network is its level of congestion, which is initially unknown to the client but heavily influences its payoff, so that the client may be interested in estimating it. Thus, we consider the Bayesian Nash equilibrium of the stage game and also we evaluate an iterative estimation process performed by the client throughout the stages, which allow to tune its equilibrium action. We discuss how the initial conditions can gauge the convergence speed of the estimate. We find out that, while the network type may be sometimes hard to evaluate, especially in low congestion scenarios, nevertheless the equilibrium action of the client is still very close to the ideal best response with full knowledge of the network type. We extend this result to the ability of the client to correctly estimate the prior distribution of the network type from multi-stage streaming games.

The spreading of video streaming services in the last few years is presenting new challenges in wireless networking; Video Rate Adaptation (VRA) is a technique that optimizes the bandwidth usage by adapting video quality as network conditions change. We propose two Quality of Experience (QoE) aware algorithms that perform VRA while guaranteeing user satisfaction.