Radio Access Networks (RANs) are critical infrastructures that mobile operators continuously upgrade to accommodate increasing data traffic demands, stricter performance requirements, and evolutions in radio technologies. RAN updates can affect carrier-level Key Performance Indicators (KPIs) that are the foundational input to data-driven models for network management. However, to date, no study has systematically examined the dynamics of RAN deployments, and little is known about the actual prevalence of RAN updates or their impact on Machine Learning (ML) models for network automation. This paper presents a first characterization of RAN updates in a nationwide operational infrastructure composed of over 500,000 carriers. A network-side vantage point lets us (i) investigate the type and frequency of RAN modifications, (ii) assess the impact of such changes on a primary KPI for network management, i.e., the traffic volume served by individual carriers, and (iii) verify the final effects on a classical downstream ML application, i.e., traffic prediction. Our results reveal that RAN updates take place with notable frequency, e.g., occurring every few days even in medium-sized cities. Also, they affect in a significant way the demands at a considerable fraction of pre-existing carriers, where they can curb the accuracy of ML traffic forecasting models.

The rollout of 5G represents a significant advancement in the telecommunications industry, offering the potential for markedly enhanced speeds, reduced latency, and improved connectivity. Considering these anticipated advantages, it is interesting to understand the progressive adoption of the new technology by operators and their subscribers. In this paper, we analyze the evolution and current operation of the nation-wide 5G network of Orange, a leading mobile operator in France. By inspecting longitudinal data about (i) the over-five-year-long development of the country-wide 5G radio access infrastructure and (ii) the last two years of 5G traffic demands, we unveil how the operator has planned the deployment of the 5G radio access and characterize the actual usage patterns of the available 5G infrastructure. We also investigate the recent introduction of a 5G Standalone (SA) commercial service and its adoption by the mobile subscribers. We show that by mid 2025, the 5G network under study has achieved substantial coverage of populated areas and the operator has very recently started adding capacity layers to its 5G radio access. However, our investigation reveals that such massive infrastructure deployment efforts are not matched by a commensurate adoption of the technology by the end users, as the 5G capacity -especially for SA- stays largely underutilized.

Trajectory datasets collected by network operators and service providers offer detailed information about individual mobility and have wide application in business and research. However, managing such data raises privacy risks, as the unique movement patterns of individuals pose significant re-identification risks and make common countermeasures like pseudonymization ineffective. The privacy-preserving data publishing (PPDP) of trajectory datasets that maintains post-anonymization accuracy and truthfulness is an open problem -especially for large datasets with millions of records like those gathered by major actors in the telco ecosystem. We close this gap with k-scale, a framework that implements k-anonymity in massive mobile user trajectory datasets, removing uniqueness while safeguarding accuracy at the record level. Not only k-scale is the first model capable of scaling k-anonymization to a dataset of one million trajectories, but it does so while also outperforming state-of-the-art methods for trajectory data publishing in terms of preserved data quality, which we prove in real-world massive datasets and applications.

The vision of intelligent networks capable of automatically configuring crucial parameters for tasks such as resource provisioning, anomaly detection or load balancing largely hinges upon efficient AI-based algorithms. Time series forecasting is a fundamental building block for network-oriented AI and current trends lean towards the systematic adoption of models based on deep learning approaches. In this paper, we pave the way for a different strategy for the design of predictors for mobile network environments, and we propose the Thresholded Exponential Smoothing (TES) framework, a hybrid Statistical Modeling and Deep Learning tool that allows for improving the performance of network Key Performance Indicator (KPI) forecasting. We adapt our framework to two state-of-the-art deep learning tools for time series forecasting, based on Recurrent Neural Networks and Transformer architectures. We experiment with TES by showcasing its superior support for three practical network management use cases, i.e. (i) anticipatory allocation of network resources, (ii) mobile traffic anomaly prediction, and (iii) mobile traffic load balancing. Our results, derived from traffic measurements collected in operational mobile networks, demonstrate that the TES framework can yield substantial performance gains over current state-of-the-art predictors in the applications considered.

In high-speed data center networks, persistent flows are repeatedly observed over extended periods, potentially signaling threats such as stealthy DDoS or botnet attacks. Monitoring every flow in production-grade hardware switches that feature limited memory, however, is challenging under typical high flow rates and data volumes. To tackle this, approximate data structures, like sketches, are often employed. Yet many existing methods rely on per-time-window flag resets, which require frequent control-plane interventions that make them unsuitable for high-speed traffic. This paper introduces Pallas, a fully data-plane-implementable sketch for detecting persistent flows in high-speed networks with high accuracy, obviating the need for time-window-based resets. We further propose Opt-Pallas, an enhanced variant of Pallas that improves detection accuracy by incorporating flow arrival patterns. We present a rigorous error bound analysis for both Pallas and Opt-Pallas, along with extensive performance evaluations using a P4-based prototype on an Intel Tofino switch. Pallas scales persistent flow detection to line-rate capacity, while state-of-the-art solutions fail to operate beyond a few Mbps. Our results show that Pallas and Opt-Pallas can accurately detect persistent flows in traffic volumes over 60× higher than those handled by the best existing approach. Additionally, even under low-speed traffic, Pallas and Opt-Pallas achieve 4.21% and 7.85% higher lookup accuracy while consuming only 8.5% and 9.7% of switch resources, respectively. Extensive trace-driven results on a CPU platform further validate the high detection accuracy of Opt-Pallas compared to existing methods.

The rollout of 5G promises major improvements in speed, latency, and connectivity over previous-generation radio access technologies. Our study analyzes Orange's nationwide 5G network in France, combining longitudinal data on infrastructure deployment with data traffic patterns. Early results show that while 5G coverage has steadily expanded and is presently reaching the vast majority of the user population, adoption by mobile subscribers remains limited, leaving much of the new capacity underutilized.

Summer Olympic Games are one of the major sports and social events worldwide, attracting global media attention, thousands of athletes, and large crowds to the hosting country.As such, the Olympics also represent a moment of severe strain for local infrastructures, including the telecommunication one.Yet, very little is known about how this large event affects demands for telco services.In this paper, we explore how the 2024 Summer Olympics hosted by Paris, France conditioned local mobile data traffic volumes and dynamics.We do so from a privileged vantage point by analyzing measurements collected in Orange's production network, largest mobile operator in the country and the official communications partner to the event organization.Our results shed light on a variety of aspects, including how Olympic Games affect consumption of mobile services, the burden that the event imposes on the local mobile network infrastructure, and how operators prepare for it.

In today's web-scale, data-driven environments, real-time detection of persistent items that consistently recur over time is essential for maintaining system integrity, reliability, and security. Persistent items often signal critical anomalies, such as stealthy DDoS and botnet attacks in web infrastructures. Although various methods exist for identifying such items as well as for determining their frequency, they require recording every item for processing, which is impractical at very high data rates achieved by modern data streams. In this paper, we introduce Pontus, a novel approach that uses an approximate data structure (sketch) specifically designed for the efficient and accurate detection of persistent items. Our method not only achieves fast and precise lookup but is also flexible, allowing for minor modifications to accommodate other types of persistence-based item detection tasks, such as detecting persistent items with low frequency. We rigorously validate our approach through formal methods, offering detailed proofs of time/space complexity and error bounds to demonstrate its theoretical soundness. Our extensive trace-driven evaluations across various persistence-based tasks further demonstrate Pontus's effectiveness in significantly improving detection accuracy and enhancing processing speed compared to existing approaches. We implement Pontus in an experimental platform with industry-grade Intel Tofino switches and demonstrate the practical feasibility of our approach in a real-world memory-constrained environment.

The ability to perform mobile traffic forecasting effectively with Deep Neural Networks (DNN) is instrumental to optimize resource management in 5G and beyond generation mobile networks. However, despite their capabilities, these Deep Neural Networks (DNN)s often act as complex opaque-boxes with decisions that are difficult to interpret. Even worse, they have proven vulnerable to adversarial attacks which undermine their applicability in production networks. Unfortunately, although existing state-of-the-art EXplainable Artificial Intelligence (XAI) techniques are often demonstrated in computer vision and Natural Language Processing (NLP), they may not fully address the unique challenges posed by spatio-temporal time-series forecasting models. To address these challenges, we introduceDeExpin this paper, a tool that flexibly builds upon legacy EXplainable Artificial Intelligence (XAI) techniques to synthesize compact explanations by making it possible to understand which Base Stations (BSs) are more influential for forecasting from a spatio-temporal perspective. Armed with such knowledge, we run state-of-the-art Adversarial Machine Learning (AML) techniques on those BSs to measure the accuracy degradation of the predictors under adversarial attacks. Our comprehensive evaluation uses real-world mobile traffic datasets and demonstrates that legacy XAI techniques spot different types of vulnerabilities. While Gradient-weighted Class Activation Mapping (GC) is suitable to spot BSs sensitive to moderate/low traffic injection, LayeR-wise backPropagation (LRP) is suitable to identify BSs sensitive to high traffic injection. Under moderate adversarial attacks, the prediction error of the BSs identified as vulnerable can increase by more than 250%.

Forecasting is increasingly considered a fundamental enabler for the management of next-generation mobile networks. While deep neural networks excel at short- and long-term forecasting, their complexity hinders interpretability, a crucial factor for production deployment. The existing EXplainable Artificial Intelligence (XAI) techniques, primarily designed for computer vision and natural language processing, struggle with time series data due to their lack of understanding of temporal characteristics of the input data. In this paper, we take the research on EXplainable Artificial Intelligence (XAI) for time series forecasting one step further by proposingAIChronoLens, a new tool that links legacy XAI explanations with the temporal properties of the input.AIChronoLensallows diving deep into the behavior of time series predictors and spotting, among other aspects, the hidden causes of forecast errors. We show thatAIChronoLens’s output can be utilized for meta-learning to predict when the original time series forecasting model makes errors and fix them in advance, thereby improving the accuracy of the predictors. Extensive evaluations with real-world mobile traffic traces pinpoint model behaviors that would not be possible to identify otherwise and show how model performance can be improved by 32 % upon re-training and by up to 39 % with meta-learning.

In the past few years, network infrastructures have transitioned from prominently hardware-based models to networks of functions, where software components provide the required functionalities with unprecedented scalability and flexibility. However, this new vision entails a completely new set of problems related to resource provisioning and the network function operation, making it difficult to manage the network function lifecycle management with traditional, human-in-the-loop approaches. Novel zero-touch management solutions promise autonomous network operation with limited human interactions. However, modeling network function behavior into compelling variables and algorithm is an aspect that such solutions must take into account. In this paper, we propose AZTEC+, a data-driven solution for anticipatory resource provisioning in network slicing scenarios. By leveraging a hybrid and modular deep learning architecture, AZTEC+ not only forecasts the future demands for target services but also identifies the best trade-offs to balance the costs due to the instantiation and reconfiguration of such resources. Our experimental evaluation, based on real-world network data, shows how AZTEC+ can outperform state-of-the-art management solutions for a large set of metrics.

The mobile telecommunication industry is undergoing continuous evolution to cope with ever increasing service requirements and expectations of end users. This has recently led to the rise of Mobile Network Aggregators (MNAs), a new type of global virtual operators that deliver mobile communication services by utilizing multiple Mobile Network Operators (MNOs), dynamically connecting to the one that best meets their customers’ needs based on location and time. MNAs can then offer optimized global coverage by connecting to local MNOs that have limited (e.g., national) geographic service. In this paper, we provide a first in-depth analysis of the operations of three major MNAs: Google Fi, Twilio, and Truphone. We conduct performance measurements across these MNAs for critical applications spanning DNS, web browsing, and video streaming, and compare their performance against that of a traditional MNO from two very diverse geographical locations, US and Spain. We find that MNAs may introduce some delay compared to local MNOs in the region where the user is roaming, yet they offer significant performance improvements over the traditional MNOs roaming model, such as home-routed roaming. To fully assess the potential benefits of the MNA model, we also carry out emulation studies assessing the potential performance gains that MNAs could achieve by deploying both control and user plane functions of open-source 5G implementations across different Amazon Web Services locations.

Integrating machine learning (ML) models directly in the network user plane enables inference on data traffic at line rate, and can dramatically reduce the latency and improve the scalability of key functionalities like traffic classification or intrusion detection. Yet, the hardware that can be used for this purpose, in particular programmable switches, present stringent constraints in terms of limited memory and little support for mathematical operations or data types that render ML model deployment a substantial technical challenge. In this paper, we make a step forward in user-plane ML by introducingFlowrest, a solution that redefines the state of the art in flow-level inference for programmable switches.Flowrestallows implementing general-purpose Random Forest (RF) models in industry-grade switches by ($\boldsymbol {i}$) suitably handling stateful flow-level (FL) features in the switch ASIC, ($\boldsymbol {ii}$) achieving low-collision flow management, and ($\boldsymbol {iii}$) customizing RF models right from the design phase for in-switch operation. We developFlowrestas an open-source software using the P4 language and evaluate its performance in an experimental testbed with Intel Tofino switches. Experiments with inference tasks of varying complexity prove that our solution improves accuracy by over 10 percent points on average with respect to the second-best competitor out of five recent approaches for RF-based in-switch inference, while maintaining sub-microsecond latency.

Recent works have shown that Machine Learning (ML) models can be deployed in P4-programmable user planes for line rate inference on live traffic and that these user planes can also be used to accelerate the 5G User Plane Function (UPF). This work builds on these capabilities to explore how ML inference in the user plane can facilitate real-time intrusion detection in 5G networks. As a proof-of-concept, we describe how an ML model could be deployed into the UPF as a special Packet Detection Rule (PDR). We then train and deploy a tree-based classifier into a P4programmable switch acting as the UPF and conduct experiments on a testbed with off-the-shelf hardware using experimental data from a 5 G test network on a university campus. Our results confirm that running ML-based intrusion detection on P4-based UPFs ensures line-rate attack detection and classification with an accuracy of up to$98 \%$in terms of F1 score, while keeping switch resource consumption increase under control.

Smartphones and mobile applications are staple tools in the operation of current-age public demonstrations, where they support organizers and participants in, \eg scaling the management of the events or communicating live about their objectives and traction. % The widespread use of mobile services during protests also presents interesting opportunities to observe the dynamics of these manifestations from a digital perspective. Previous studies in that direction have focused on the analysis of content posted in selected social media so as to forecast, survey or ascertain the success of public protests. In this paper, we take a different viewpoint and present a holistic characterization of the consumption of the whole spectrum of mobile applications during social protests. Hinging upon pervasive measurements in the production network of the incumbent network operator and focusing on the 2023 French pension reform strikes, we unveil how large masses of protesters generate a clearly recognizable footprint on mobile service demands in the examined events. In fact, the footprint is so strong that it lets us develop models informed by the usage of selected mobile applications that are capable of (i) tracking the spatiotemporal evolution of the target demonstrations and (ii) estimate the time-varying number of attendees from aggregate network operator data only. We demonstrate the utility of such privacy-preserving models to perform a-posteriori analyses of the public protests that reveal, e.g., the precise progression of the marches, alternate minor routes taken by participants or their dispersal at the end of the events.

Embedding machine learning (ML) models in programmable switches realizes the vision of high-throughput and low-latency inference at line rate. Recent works have made breakthroughs in embedding Random Forest (RF) models in switches for either packet-level inference or flow-level inference. The former relies on simple features from packet headers that are simple to implement but limit accuracy in challenging use cases; the latter exploits richer flow features to improve accuracy, but leaves early packets in each flow unclassified. We propose Jewel, an in-switch ML model based on a fully joint packet-and flow-level design, which takes the best of both worlds by classifying early flow packets individually and shifting to flow-level inference when possible. Our proposal involves (i) a single RF model trained to classify both packets and flows, and (ii) hardware-aware model selection and training techniques for resource footprint minimization. We implement Jewel in P4 and deploy it in a testbed with Intel Tofino switches, where we run extensive experiments with a variety of real-world use cases. Results reveal how our solution outperforms four state-of-the-art benchmarks, with accuracy gains in the 2.0%–5.3% range.

Next-generation mobile networks will increasingly rely on the ability to forecast traffic patterns for resource management. Usually, this translates into forecasting diverse objectives like traffic load, bandwidth, or channel spectrum utilization, measured over time. Among the other techniques, Long-Short Term Memory (LSTM) proved very successful for this task. Unfortunately, the inherent complexity of these models makes them hard to interpret and, thus, hampers their deployment in production networks. To make the problem worsen, EXplainable Artificial Intelligence (XAI) techniques, which are primarily conceived for computer vision and natural language processing, fail to provide useful insights: they are blind to the temporal characteristics of the input and only work well with highly rich semantic data like images or text. In this paper, we take the research on XAI for time series forecasting one step further proposing AIChronoLens, a new tool that links legacy XAI explanations with the temporal properties of the input. In such a way, AIChronoLens makes it possible to dive deep into the model behavior and spot, among other aspects, the hidden cause of errors. Extensive evaluations with real-world mobile traffic traces pinpoint model behaviors that would not be possible to spot otherwise and model performance can increase by 32%.

The roll out of 5G, coupled with the traffic monitoring capabilities of modern industry-grade networks, offers an unprecedented opportunity to closely observe the impact that the introduction of a new major wireless technology has on the end users. In this paper, we seize such a unique chance, and carry out a first-of-its-kind in-depth analysis of 5G adoption along spatial, temporal and service dimensions. Leveraging massive measurement data about application-level demands collected in a nationwide 4G/5G network, we characterize the impact of the new technology on when, where and how mobile subscribers consume 5G traffic both in aggregate and for individual types of services. This lets us unveil the overall incidence of 5G in the total mobile network traffic, its spatial and temporal fluctuations, its effect on the way 5G services are consumed, the way individual services and geographical locations contribute to fluctuations in the 5G demand, as well as surprising connections between socioeconomic status of local populations and the way the 5G technology is presently consumed.

RAN virtualization is revolutionizing the telco industry, enabling 5G Distributed Units to run using general-purpose platforms equipped with Hardware Accelerators (HAs). Recently, GPUs have been proposed as HAs, hinging on their unique capability to execute 5G PHY operations efficiently while also processing Machine Learning (ML) workloads. While this ambivalence makes GPUs attractive for cost-effective deployments, we experimentally demonstrate that multiplexing 5G and ML workloads in GPUs is in fact challenging, and that using conventional GPU-sharing methods can severely disrupt 5G operations. We then introduce YinYangRAN, an innovative O-RAN-compliant solution that supervises GPU-based HAs so as to ensure reliability in the 5G processing pipeline while maximizing the throughput of concurrent ML services. YinYangRAN performs GPU resource allocation decisions via a computationally-efficient approximate dynamic programming technique, which is informed by a neural network trained on real-world measurements. Using workloads collected in real RANs, we demonstrate that YinYangRAN can achieve over 50% higher 5G processing reliability than conventional GPU sharing models with minimal impact on co-located ML workloads. To our knowledge, this is the first work identifying and addressing the complex problem of HA management in emerging GPU-accelerated vRANs, and represents a promising step towards multiplexing PHY and ML workloads in mobile networks.

Open and virtualized Radio Access Networks (vRANs) are breeding a new market with unprecedented opportunities. However, carrier-grade vRANs today are expensive and energy-hungry, as they rely on hardware accelerators (HAs) that are dedicated to individual distributed units (DUs). In this paper, we argue that sharing pools of heterogeneous processors among DUs leads to more cost- and energy-efficient vRANs. We then design CloudRIC, a system that, powered by lightweight data-driven models, meets specific reliability targets while (i) coordinating access between DUs and heterogeneous computing infrastructure; and (ii) assisting DUs with compute-aware radio scheduling procedures. Experiments on a GPU-accelerated O-Cloud show that CloudRIC can achieve, respectively, 3x and 15x mean gains in energy- and cost-efficiency under real RAN workloads while ensuring 99.999% reliability even in dense scenarios.

Open and virtualized Radio Access Networks (vRANs) are breeding a new market with unprecedented opportunities. However, carrier-grade vRANs today are expensive and energy-hungry, as they rely on hardware accelerators (HAs) that are dedicated to individual distributed units (DUs). We demonstrate CloudRIC [17], a system that, powered by lightweight data-driven models, meets specific reliability targets while (i) coordinating access between DUs and heterogeneous computing infrastructure; and (ii) assisting DUs with compute-aware radio scheduling procedures. Using a user-friendly dashboard to control an experimental testbed remotely, we demonstrate that CloudRIC achieves comparable reliability performance to a DU-dedicated platform while offering up to 40x higher cost-efficiency and up to 6x higher energy efficiency when pooling resources for up to 70 DUs.

Growing network complexity has rendered human-in-the-loop network management approaches obsolete. The advent of Software-Defined Networking (SDN) has enabled network automation, with Machine Learning (ML) models running in the control plane. However, such control plane models do not run at line rate and would not satisfy the stringent latency requirements of time-sensitive next-generation applications. In this PhD project, we exploit recent advances in programmable switches and associated languages like P4 to enable data-driven management of networks by running ML models for inference in programmable switches at line rate, with high throughput and low latency. Resulting contributions include solutions for in-switch classification at packet level, flow level, or both, with use cases in network security, service identification, and device fingerprinting in commercial off-the-shelf switches.

Encrypted Traffic Classification (ETC) has become an important area of research with Machine Learning (ML) methods being the state-of-the-art. However, most existing solutions either rely on offline ETC based on collected network data or on online ETC with models running in the control plane of Software-Defined Networks (SDN), all of which do not run at line rate and would not meet latency requirements of time-sensitive applications in modern networks. This work leverages recent advances in data plane programmability to achieve real-time ETC in programmable switches at line rate, with high throughput and low latency. The proposed solution comprises (i) an ETC-aware Random Forest (RF) modelling process where only features based on packet size and packet arrival times are used, and (ii) an encoding of the trained RF model into production-grade P4-programmable switches. The performance of the proposed in-switch ETC framework is evaluated using 3 encrypted traffic datasets with experiments in a real-world testbed with Intel Tofino switches, in the presence of background traffic at 40 Gbps. Results show how the solution achieves high classification accuracy of up to 95%, with sub-microsecond delay, while consuming on average less than 10% of total available switch hardware resources.

Metadata geolocation, i.e., mapping information collected at a cellular Base Station (BS) to the geographical area it covers, is a central operation in the production of statistics from mobile network measurements. This task requires modeling the probability that a device attached to a BS is at a specific location, and is presently addressed with simplistic approximations based on Voronoi tessellations. As we show, Voronoi cells exhibit poor accuracy compared to real-world geolocation data, which can, in turn, reduce the reliability of research results. We propose a new approach for data-driven metadata geolocation based on a teacher-student paradigm that combines probabilistic inference and deep learning. Our Deepmend model: ($i$) only needs BS positions as input, exactly like Voronoi tessellations; (ii) produces geolocation maps that are 56% and 33% more accurate than legacy Voronoi and their state-of-the-art VoronoiBoost calibration, respectively; and, (iii) generates geolocation data for thousands of BSs in minutes. We assess its accuracy against real-world multi-city geolocation data of 5, 947 BSs provided by a network operator, and demonstrate the impact of its enhanced metadata geolocation on two applications use cases.

Due to the fact that large swathes on Earth still lack broadband communication coverage, especially in remote/rural areas and developing countries, there have been several attempts, starting from 3GPP Release 17, to lay out the architectural amendments needed for the integration of terrestrial networks with their non-terrestrial counterparts. Such attempts have led to recent projects regarding such integration that consider either 5G/5G-Advanced networks or more revolutionary approaches for the forthcoming 6G networks. In this manuscript, we give an overview of the architectural framework, technical innovations, and considered use cases of the Horizon Europe ETHER project.

Heterogeneous networks have emerged as a popular solution for accommodating the growing number of connected devices and increasing traffic demands in cellular networks. While offering broader coverage, higher capacity, and lower latency, the escalating energy consumption poses sustainability challenges. In this paper a novel optimization approach for orthogonal heterogeneous networks is proposed to minimize transmission power while respecting individual users’ throughput constraints. The problem is formulated as a mixed integer geometric program, and optimizes at once multiple system variables such as user association, working bandwidth, and base stations transmission powers. Crucially, the proposed approach becomes a convex optimization problem when user-base station associations are provided. Evaluations in multiple realistic scenarios from the production mobile network of a major European operator and based on precise channel gains and throughput requirements from measured data validate the effectiveness of the proposed approach. Overall, our original solution paves the road for greener connectivity by reducing the energy footprint of heterogeneous mobile networks, hence fostering more sustainable communication systems.

Zero-touch network management is one of the most ambitious yet strongly required paradigms for beyond 5G and 6G mobile communication systems. Achieving full automation requires a closed loop that combines (i) network status data collection and processing, (ii) predictive capabilities based on such data to anticipate upcoming needs, and (iii) effective decision making that best addresses such future needs through proper network control and orchestration. Recent seminal works have proposed approaches to jointly implement the last two phases above via a single deep learning model trained on past network status to directly optimize future decisions. This is achieved by designing custom loss functions that directly embed the management task objective. Experiments with real-world measurement data have demonstrated that this strategy leads to substantial performance gains across diverse network management tasks. In this paper, we go one step beyond the loss tailoring schemes above, and introduce a loss meta-learning paradigm that (i) reduces the need for human intervention at model design stage, (ii) eases explainability and transferability of trained deep learning models for network management, and (iii) outperforms custom losses across a range of controlled experiments and practical use cases.

Indoor cellular networks (ICNs) are anticipated to become a principal component of 5G and beyond systems. ICNs aim at extending network coverage and enhancing users' quality of service and experience, consequently producing a substantial volume of traffic in the coming years. Despite the increasing importance that ICNs will have in cellular deployments, there is nowadays little understanding of the type of traffic demands that they serve. Our work contributes to closing that gap, by providing a first characterization of the usage of mobile services across more than 4, 500 cellular antennas deployed at over 1,000 indoor locations in a whole country. Our analysis reveals that ICNs inherently manifest a limited set of mobile application utilization profiles, which are not present in conventional outdoor macro base stations (BSs). We interpret the indoor traffic profiles via explainable machine learning techniques, and show how they are correlated to the indoor environment. Our findings show how indoor cellular demands are strongly dependent on the nature of the deployment location, which allows anticipating the type of demands that indoor 5G networks will have to serve and paves the way for their efficient planning and dimensioning.

We analyze 4G and 5G transport-layer sessions generated by a wide range of mobile services at over 282,000 base stations (BSs) of an operational mobile network, and carry out a statistical characterization of their demand rates, associated traffic volume and temporal duration. Based on the gained insights, we model the arrival process of sessions at heterogeneously loaded BSs, the distribution of the session-level load and its relationship with the session duration, using simple yet effective mathematical approaches. Our models are fine-tuned to a variety of services, and complement existing tools that mimic packet-level statistics or aggregated spatiotemporal traffic demands at mobile network BSs. They thus offer an original angle to mobile traffic data generation, and support a more credible performance evaluation of solutions for network planning and management. We assess the utility of the models in practical application use cases, demonstrating how they enable a more trustworthy evaluation of solutions for the orchestration of sliced and virtualized networks.

A variety of network management and orchestration (MANO) tasks take advantage of predictions to support anticipatory decisions. In many practical scenarios, such predictions entail two largely overlooked challenges: (i) the exact relationship between the predicted values (e.g., reserved resources) and the performance objective (e.g., quality of experience of end users) is often tangled and cannot be known a priori, and (ii) the objective is linked in many cases to multiple predictions that contribute to it in an intertwined way (e.g., resources to reserved are limited and must be shared among competing flows). We present AutoManager, a novel meta-learning model that can support complex MANO tasks by addressing these two challenges. Our solution learns how multiple intertwined predictions affect a common performance goal, and steers them so as to attain the correct operation point under a-priori unknown loss functions. We demonstrate AutoManager in practical, complex use cases based on real-world traffic measurements; our experiments show that the model produces forecasts that are accurate and tailored to the MANO task in a fully automated way.

The ability to forecast mobile traffic patterns is key to resource management for mobile network operators and planning for local authorities. Several Deep Neural Networks (DNN) have been designed to capture the complex spatio-temporal characteristics of mobile traffic patterns at scale. These models are complex black boxes whose decisions are inherently hard to explain. Even worse, they have proven vulnerable to adversarial attacks which undermine their applicability in production networks. In this paper, we conduct a first in-depth study of the vulnerabilities of DNNs for large-scale mobile traffic forecasting. We propose DeExp, a new tool that leverages EXplainable Artificial Intelligence (XAI) to understand which Base Stations (BSs) are more influential for forecasting from a spatio-temporal perspective. This is challenging as existing XAI techniques are usually applied to computer vision or natural language processing and need to be adapted to the mobile network context. Upon identifying the more influential BSs, we run state-of-the art Adversarial Machine Learning (AML) techniques on those BSs and measure the accuracy degradation of the predictors. Extensive evaluations with real-world mobile traffic traces pinpoint that attacking BSs relevant to the predictor significantly degrades its accuracy across all the scenarios.

Cloud-native mobile networks pave the road for Network Slicing as a Service (NSaaS), where slice overbooking is a promising management strategy to maximize the revenues from admitted slices by exploiting the fact they are unlikely to fully utilize their reserved resources concurrently. While seminal works have shown the potential of overbooking for NSaaS in simplistic cases, its realization is challenging in practical scenarios with realistic slice demands, where its actual performance remains to be tested. In this paper, we propose kaNSaaS, a complete solution for NSaaS management with slice overbooking that combines deep learning and classical optimization to jointly solve the key tasks of admission control and resource allocation. Experiments with large-scale measurement data of actual tenant demands show that kaNSaaS increases the network operator profits by 300% with respect to NSaaS management strategies that do not employ overbooking, while outperforming by more than 20% state-of-the-art overbooking-based approaches.

Recent endeavours have enabled the integration of trained machine learning models like Random Forests in resource-constrained programmable switches for line rate inference. In this work, we first show how packet-level information can be used to classify individual packets in production-level hardware with very low latency. We then demonstrate how the newly proposed Flowrest framework improves classification performance relative to the packet-level approach by exploiting flow-level statistics to instead classify traffic flows entirely within the switch without considerably increasing latency. We conduct experiments using measurement data in a real-world testbed with an Intel Tofino switch and shed light on how Flowrest achieves an F1-score of 99% in a service classification use case, outperforming its packet-level counterpart by 8%.

Next-generation mobile networks are expected to flaunt highly (if not fully) automated management. To achieve such a vision, Artificial Intelligence (AI) and Machine Learning (ML) techniques will be key enablers to craft the required intelligence for networking, i.e., Network Intelligence (NI), empowering myriad of orchestrators and controllers across network domains. In this paper, we elaborate on the DAEMON architectural model, which proposes introducing a NI Orchestration layer for the effective end-to-end coordination of NI instances deployed across the whole mobile network infrastructure. Specifically, we first outline requirements and specifications for NI design that stem from data management, control timescales, and network technology characteristics. Then, we build on such analysis to derive initial principles for the design of the NI Orchestration layer, focusing on (i) proposals for the interaction loop between NI instances and the NI Orchestrator, and (ii) a unified representation of NI algorithms based on an extended MAPE-K model. Our work contributes to the definition of the interfaces and operation of a NI Orchestration layer that foster a native integration of NI in mobile network architectures.

Cell densification through the installation of smallcells and femtocells in indoor environments is an emerging solution to enhance the operation of wireless networks. The deployment of new components within the heart of the radio access network calls for expedient tools that assist and ensure their optimal placement within the existing network infrastructure. In this paper, we introduce metrics that can characterize indoor wireless network performance (IWNP) in terms of coverage and capacity, and we evaluate them via physics-based propagation models. In particular, we exploit a deterministic propagation model, i.e., a ray-tracer, as well as a novel machine learning-based propagation model. We demonstrate that data-driven propagation models can be leveraged for the rigorous evaluation of the IWNP metrics, yielding a remarkable computational efficiency compared to the conventional deterministic models. The use of physics-based site-specific propagation models allows for the particularities of each indoor geometry to be taken into account, and also makes feasible the consideration of uncertainties related to the indoor environment. In this case, the IWNP metrics are expressed as stochastic quantities and a stochastic solution is derived through an efficient polynomial chaos expansion representation, enabling on-the-fly computation of the IWNP metrics statistics.

Intent-Based Networking mandates that high-level human-understandable intents are automatically interpreted and implemented by network management entities. As a key part in this process, it is required that network orchestrators activate the correct automated decision model to meet the intent objective. In anticipatory networking tasks, this requirement maps to identifying and deploying a tailored prediction model that can produce a forecast aligned with the specific –and typically complex– network management goal expressed by the original intent. Current forecasting models for network demands or network management optimize generic, non-flexible, and manually designed objectives, hence do not fulfil the needs of anticipatory Intent-Based Networking. To close this gap, we propose LossLeaP, a novel forecasting model that can autonomously learn the relationship between the prediction and the target management objective, steering the former to minimize the latter. To this end, LossLeaP adopts an original deep learning architecture that advances current efforts in automated machine learning, towards a spontaneous design of loss functions for regression tasks. Extensive experiments in controlled environments and in practical application case studies prove that LossLeaP outperforms a wide range of benchmarks, including state-of-the-art solutions for network capacity forecasting.

The COVID-19 pandemic has affected our lives and how we use network infrastructures in an unprecedented way. While early studies have started shedding light on the link between COVID-19 containment measures and mobile network traffic, we presently lack a clear understanding of the implications of the virus outbreak, and of our reaction to it, on the usage of mobile apps. We contribute to closing this gap, by investigating how the spatiotemporal usage of mobile services has evolved through different response measures enacted in France during a continued seven-month period in 2020 and 2021. Our work complements previous studies in several ways: (i) it delves into individual service dynamics, whereas previous studies have not gone beyond broad service categories; (ii) it encompasses different types of containment strategies, allowing to observe their diverse effects on mobile traffic; (iii) it covers both spatial and temporal behaviors, providing a comprehensive view on the phenomenon. These elements of novelty let us lay new insights on how the demands for hundreds of different mobile services are reacting to the new environment set forth by the pandemics.

Mobile network traffic data offers unprecedented opportunities for innovative studies within and beyond networking. However, progress is hindered by the very limited access that the research community at large has to the real-world mobile network data that is needed to develop and dependably test mobile traffic data-driven solutions. As a contribution to overcome this barrier, we propose CartaGenie, a generator of realistic mobile traffic snapshots at city scale. Taking a deep generative modeling approach and through a tailored conditional generator design, CartaGenie can synthesize high-fidelity and artifact-free spatial traffic snapshots using only contextual information about the target geographical region that is easily found in public repositories. Hence, CartaGenie allows researchers to create their own realistic datasets of spatial traffic from open data about their region of interest. Experiments with real-world mobile traffic measurements collected in multiple metropolitan areas show that CartaGenie can produce dependable network traffic loads for areas where no prior traffic information is available, significantly outperforming a comprehensive set of benchmarks. Moreover, tests with practical case studies demonstrate that the synthetic data generated by CartaGenie is as good as real data in supporting diverse research-oriented mobile traffic data-driven applications.

Mapping information collected at the level of individual base stations onto the geographical space is a required operation for many works relying on mobile network metadata. The common practice is to represent base station coverage as Voronoi cells, and assume that users are uniformly distributed therein. In this paper, we leverage a large-scale dataset of realistic spatial association probabilities to over 5,000 operational base stations, and quantify the substantial problems of such a simplistic mapping approach. To address the limitations of legacy Voronoi representations, we develop VoronoiBoost, a data-driven model that scales Voronoi cells to match the probabilistic distribution of users associated to each base station. VoronoiBoost relies on the same input as traditional Voronoi decompositions, but provides a richer and more accurate rendering of where users are located: hence, it can be readily used by researchers to substantially improve the spatial representation of mobile network metadata. Our experiments demonstrate that VoronoiBoost improves the quality of mapping by 44% on average over standard Voronoi cells. We also showcase the utility of our model in a practical Edge network planning use case, where the information produced by VoronoiBoost drives a deployment up to 28% more accurate than that obtained with Voronoi cells.

Forecasting is a task of ever increasing importance for the operation of mobile networks, where it supports anticipatory decisions by network intelligence and enables emerging zero-touch service and network management models. While current trends in forecasting for anticipatory networking lean towards the systematic adoption of models that are purely based on deep learning approaches, we pave the way for a different strategy to the design of predictors for mobile network environments. Specifically, following recent advances in time series prediction, we consider a hybrid approach that blends statistical modelling and machine learning by means of a joint training process of the two methods. By tailoring this mixed forecasting engine to the specific requirements of network traffic demands, we develop a Thresholded Exponential Smoothing and Recurrent Neural Network (TES-RNN) model. We experiment with TES-RNN in two practical network management use cases, i.e., (i) anticipatory allocation of network resources, and (ii) mobile traffic anomaly prediction. Results obtained with extensive traffic workloads collected in an operational mobile network show that TES-RNN can yield substantial performance gains over current state-of-the-art predictors in both applications considered.

We study the interaction between the consumption of digital services via mobile devices and urbanization levels, using measurement data collected in an operational network serving the whole territory of France. We unveil that such an interaction follows a power law, or, in other words, there exists an emergent behavior that prompts subscribers living in increasingly extended and populated urban areas to exhibit a surging individual consumption of mobile traffic. The result holds for the global traffic, but is also consistently observed across a range of mobile services, although with varying intensity. An unprecedented longitudinal analysis of the phenomenon unveils how the imbalance in the per-capita mobile data traffic usage across cities of different size has grown steadily and substantially in the 2014–2019 time frame in France. Our study raises questions on the presence of second-level digital divides in developed countries, and paves the road to further investigations.

Recently, the topic of indoor outdoor detection (IOD) has seen its popularity increase, as IOD models can be leveraged to augment the performance of numerous Internet of Things and other applications. IOD aims at distinguishing in an efficient manner whether a user resides in an indoor or an outdoor environment, by inspecting the cellular phone sensor recordings. Legacy IOD models attempt to determine a user’s environment by comparing the sensor measurements to some threshold values. However, as we also observe in our experiments, such models exhibit limited scalability, and their accuracy can be poor. Machine learning (ML)-based IOD models aim at removing this limitation, by utilizing a large volume of measurements to train ML algorithms to classify a user’s environment. Yet, in most of the existing research, the temporal dimension of the problem is disregarded. In this article, we propose treating IOD as a multivariate time-series classification (TSC) problem, and we explore the performance of various deep learning (DL) models. We demonstrate that a multivariate TSC approach can be used to monitor a user’s environment, and predict changes in its state, with greater accuracy compared to conventional approaches that ignore the feature variation over time. Additionally, we introduce a new DL model for multivariate TSC, exploiting the concept of self-attention and atrous spatial pyramid pooling. The proposed DL multivariate TSC framework exploits only low power consumption sensors to infer a user’s environment, and it outperforms state-of-the-art models, yielding a higher accuracy combined with a smaller computational cost.

Service-level mobile traffic data enables research studies and innovative applications with a potential to shape future service-oriented communication systems and beyond. However, real-world datasets reporting measurements at the individual service level are hard to access as such data is deemed commercially sensitive by operators. APPSHOT is a model for generating synthetic high-fidelity city-scale snapshots of service level mobile traffic. It can operate in any geographical region and relies solely on easily available spatial context information such as population density, thus allowing the generation of new and open traffic datasets for the research community. The design of APPSHOT is informed by an original characterization of service-level mobile traffic data. APPSHOT is a novel conditional GAN design instantiated by a convolutional neural network generator and two discriminators. The model features several other innovative mechanisms including multi-channel and overlapping patch based generation to address the unique challenges involved in generating mobile service traffic snapshots. Experiments with ground-truth data collected by a major European operator in multiple metropolitan areas show that APPSHOT can produce realistic network loads at the service level for areas where it has no prior traffic knowledge, and that such data can reliably support service-oriented networking studies.

This paper introduces CloudLSTM, a new branch of recurrent neural models tailored to forecasting over data streams generated by geospatial point-cloud sources. We design a Dynamic Point-cloud Convolution (DConv) operator as the core component of CloudLSTMs, which performs convolution directly over point-clouds and extracts local spatial features from sets of neighboring points that surround different elements of the input. This operator maintains the permutation invariance of sequence-to-sequence learning frameworks, while representing neighboring correlations at each time step -- an important aspect in spatiotemporal predictive learning. The DConv operator resolves the grid-structural data requirements of existing spatiotemporal forecasting models and can be easily plugged into traditional LSTM architectures with sequence-to-sequence learning and attention mechanisms. We apply our proposed architecture to two representative, practical use cases that involve point-cloud streams, i.e. mobile service traffic forecasting and air quality indicator forecasting. Our results, obtained with real-world datasets collected in diverse scenarios for each use case, show that CloudLSTM delivers accurate long-term predictions, outperforming a variety of competitor neural network models.

City-scale spatiotemporal mobile network traffic data can support numerous applications in and beyond networking. However, operators are very reluctant to share their data, which is curbing innovation and research reproducibility. To remedy this status quo, we propose SpectraGAN, a novel deep generative model that, upon training with real-world network traffic measurements, can produce high-fidelity synthetic mobile traffic data for new, arbitrary sized geographical regions over long periods. To this end, the model only requires publicly available context information about the target region, such as population census data. SpectraGAN is an original conditional GAN design with the defining feature of generating spectra of mobile traffic at all locations of the target region based on their contextual features. Evaluations with mobile traffic measurement datasets collected by different operators in 13 cities across two European countries demonstrate that SpectraGAN can synthesize more dependable traffic than a range of representative baselines from the literature. We also show that synthetic data generated with SpectraGAN yield similar results to that with real data when used in applications like radio access network infrastructure power savings and resource allocation, or dynamic population mapping.

RAN energy consumption is a major OPEX source for mobile telecom operators, and 5G is expected to increase these costs by several folds. Moreover, paradigm-shifting aspects of the 5G RAN architecture like RAN disaggregation, virtualization and cloudification introduce new traffic-dependent resource management decisions that make the problem of energy-efficient 5G RAN orchestration harder. To address such a challenge, we present a first comprehensive virtualized RAN (vRAN) system model aligned with 5G RAN specifications, which embeds realistic and dynamic models for computational load and energy consumption costs. We then formulate the vRAN energy consumption optimization as an integer quadratic programming problem, whose NP-hard nature leads us to develop GreenRAN, a novel, computationally efficient and distributed solution that leverages Lagrangian decomposition and simulated annealing. Evaluations with real-world mobile traffic data for a large metropolitan area are another novel aspect of this work, and show that our approach yields energy efficiency gains up to 25% and 42%, over state-of-the-art and baseline traditional RAN approaches, respectively.

The characterization of user behavior is key to perform traffic analysis, modeling and optimization of network components and protocols. This is especially true for 5G and beyond networks that will heavily rely on machine learning for network optimization. In Base Station (BS) traffic traces, users are uniquely identified by a Radio Network Temporary Identifier (RNTI) assigned to them. RNTIs are not bound to a user but are reused upon expiration of an inactivity timer, whose duration is operator dependent. This implies that, over time, multiple users can be mapped to the same RNTI ID in diverse ways. Hence, when using real-world radio access measurement traces for traffic analysis, distinguishing individual users within the RNTI space is a non-trivial task. In this paper, we provide the first in-depth study of the RNTI allocation process and shed light not only on the setting of the inactivity timer, but also on the relationship of the RNTI allocation scheme and the user characteristics. For this, we collect a large dataset of mobile traffic from multiple BSs of several mobile network operators. The analysis of the decoded control messages of the BS unveils that the RNTI allocation process changes over time depending on the BSs observed load and time of day. We also observe that the RNTI expiration threshold is on the order of minutes, and demonstrate how using thresholds around 10~s that are reported in the vast majority of the literature can bias subsequent analyses. Overall, our work provides an important step towards dependable mobile network trace analysis, and lays more solid foundations to research relying on traffic traces for data-driven analysis and simulation.

Beyond 5G mobile networks have to support a wide range of performance requirements and unprecedented levels of flexibility. To this end, massive MIMO is a critical technology to improve spectral efficiency and thus scale up network capacity, by increasing the number of antenna elements. This also increases the overhead of Channel State Information (CSI) estimation and obtaining accurate CSI is a fundamental problem in massive MIMO systems. In this paper, we focus on scheduling uplink Sounding Reference Signals (SRSs) that carry pilot symbols for CSI estimation. Under the large number of users and high load that are expected to characterize beyond 5G systems, the limited amount of resources available for SRSs makes the legacy 3GPP periodic allocation scheme largely inefficient. We design TRADER, an SRS resource allocation framework that minimizes the age of channel estimates by taking advantage of machine learning-based short-term traffic forecasts at the base station level. By anticipating traffic bursts, TRADER schedules SRS resources so as to obtain CSI for each user right before the corresponding traffic arrives. Experiments with extensive real-world mobile network traces show that our solution is efficient and robust in high load scenarios: with respect to a round robin schedule of aperiodic SRS, TRADER provides more often CSI within the coherence time (up to 5× for given scenarios), leading to channel gains of up to 2 dB.

The combination of network softwarization with network slicing enables the provisioning of very diverse services over the same network infrastructure. However, it also creates a complex environment where the orchestration of network resources cannot be guided by traditional, human-in-the-loop network management approaches. New solutions that perform these tasks automatically and in advance are needed, paving the way to zero-touch network slicing. In this paper, we propose AZTEC, a data-driven framework that effectively allocates capacity to individual slices by adopting an original multi-timescale forecasting model. Hinging on a combination of Deep Learning architectures and a traditional optimization algorithm, AZTEC anticipates resource assignments that minimize the comprehensive management costs induced by resource overprovisioning, instantiation and reconfiguration, as well as by denied traffic demands. Experiments with real-world mobile data traffic show that AZTEC dynamically adapts to traffic fluctuations, and largely outperforms state-of-the-art solutions for network resource orchestration.

The growing diversification of mobile services imposes requirements on network performance that are ever more stringent and heterogeneous. Network slicing aligns mobile network operation to this context, by enabling operators to isolate and customize network resources on a per-service basis. A key input for provisioning resources to slices is real-time information about the traffic demands generated by individual services. Acquiring such knowledge is however challenging, as legacy approaches based on in-depth inspection of traffic streams have high computational costs, which inflate with the widening adoption of encryption over data and control traffic. In this paper, we present a new approach to service-level demand estimation for slicing, which hinges on decomposition, i.e., the inference of per-service demands from traffic aggregates. By operating on total traffic volumes only, our approach overcomes the complexity and limitations of legacy traffic classification techniques, and provides a suitable input to recent 'Network Slice as a Service' (NSaaS) models. We implement decomposition through Microscope, a novel framework that uses deep learning to infer individual service demands from complex spatiotemporal features hidden in traffic aggregates. Microscope (i) transforms traffic data collected in irregular radio access deployments in a format suitable for convolutional learning, and (ii) can accommodate a variety of neural network architectures, including original 3D Deformable Convolutional Neural Networks (3D-DefCNNs) that we explicitly design for decomposition. Experiments with measurement data collected in an operational network demonstrate that Microscope accurately estimates per-service traffic demands with relative errors below 1.2%. Further, tests in practical NSaaS management use cases show that resource allocations informed by decomposition yield affordable costs for the mobile network operator.

The dynamic management of network resources is both a critical and challenging task in upcoming multi-tenant mobile networks, which requires allocating capacity to individual network slices so as to accommodate future time-varying service demands. Such an anticipatory resource configuration process must be driven by suitable predictors that take into account the monetary cost associated to overprovisioning or underprovisioning of networking capacity, computational power, memory, or storage. Legacy models that aim at forecasting traffic demands fail to capture these key economic aspects of network operation. To close this gap, we present DeepCog, a deep neural network architecture inspired by advances in image processing and trained via a dedicated loss function. Unlike traditional traffic volume predictors, DeepCog returns a cost-aware capacity forecast, which can be directly used by operators to take short- and long-term reallocation decisions that maximize their revenues. Extensive performance evaluations with real-world measurement data collected in a metropolitan-scale operational mobile network demonstrate the effectiveness of our proposed solution, which can reduce resource management costs by over 50% in practical case studies.

Network slicing is a new paradigm for future 5G networks where the network infrastructure is divided into slices devoted to different services and customized to their needs. With this paradigm, it is essential to allocate to each slice the needed resources, which requires the ability to forecast their respective demands. To this end, we present DeepCog, a novel data analytics tool for the cognitive management of resources in 5G systems. DeepCog forecasts the capacity needed to accommodate future traffic demands within individual network slices while accounting for the operator's desired balance between resource overprovisioning (i.e., allocating resources exceeding the demand) and service request violations (i.e., allocating less resources than required). To achieve its objective, DeepCog hinges on a deep learning architecture that is explicitly designed for capacity forecasting. Comparative evaluations with real-world measurement data prove that DeepCog's tight integration of machine learning into resource orchestration allows for substantial (50% or above) reduction of operating expenses with respect to resource allocation solutions based on state-of-the-art mobile traffic predictors. Moreover, we leverage DeepCog to carry out an extensive first analysis of the trade-off between capacity overdimensioning and unserviced demands in adaptive, sliced networks and in presence of real-world traffic.

In this paper, we study the capacity (i.e., the maximum achievable throughput) of radio access networks that exploit mobile small cell base stations carried by vehicles for adaptive densification in urban areas. While traditional approaches for radio access network densification with fixed small cell base stations are proving ineffective and extremely costly, mobile small cell base stations carried by vehicles can provide adaptive densification while achieving higher efficiency and lower cost. As a matter of fact, the existence of correlations between the number of mobile network subscribers and the number of vehicles in a given area allows for the spontaneous creation of temporary dense small cell deployments where and when needed. Ultimately, this approach to Radio Access Network densification increases efficiency, hence reducing costs for the operator. In this context, we first present an approach for the computation of the maximum throughput that can be obtained in an area served by traditional fixed base stations and mobile small cell base stations. We then provide initial estimates for the throughput improvements with respect to traditional deployments that rely on fixed base stations only. Evaluations in the realistic case study of the main railway station area in Milan, Italy, reveal that the use of mobile base stations achieves throughout gains up to 120% over legacy fixed access infrastructures, while granting higher fairness among subscribers.

Human mobility analysis is a multidisciplinary research subject that has attracted a growing interest over the last decade. A substantial amount of such recent studies is driven by the availability of original sources of real-world information about individual movement patterns. An important task in the analysis of mobility data is reliably distinguishing between the stop locations and movement phases that compose the trajectories of the monitored subjects. The problem is especially challenging when mobility is inferred from mobile phone location data: here, oscillations in the association of mobile devices to base stations lead to apparent user mobility even in absence of actual movement. In this paper, we leverage a unique dataset of spatiotemporal individual trajectories that allows capturing both the user and network operator perspectives in mobile phone location data, and investigate the oscillation phenomenon. We present probabilistic and machine learning approaches for detecting oscillations in mobile phone location data, and a filtering technique for removing those. Our analyses and comparison with state-of-the-art approaches demonstrate the superiority of our solution, both in terms of removed oscillations and of error with respect to ground-truth trajectories.

We investigate spatial patterns in mobile service consumption that emerge at national scale. Our investigation focuses on a representative case study, i.e., France, where we find that: (i) the demand for popular mobile services is fairly uniform across the whole country, and only a reduced set of peculiar services (mainly operating system updates and long-lived video streaming) yields geographic diversity; (ii) even for such distinguishing services, the spatial heterogeneity of demands is limited, and a small set of consumption behaviors is sufficient to characterize most of the mobile service usage across the country; (iii) the spatial distribution of these behaviors correlates well with the urbanization level, ultimately suggesting that the adoption of geographically-diverse mobile applications is linked to a dichotomy of cities and rural areas. We derive our results through the analysis of substantial measurement data collected by a major mobile network operator, leveraging an approach rooted in information theory that can be readily applied to other scenarios.

Communication-enabled devices routinely carried by individuals have become pervasive, opening unprecedented opportunities for collecting digital metadata about the mobility of large populations. In this paper, we propose a novel methodology for the estimation of people density at metropolitan scales, using subscriber presence metadata collected by a mobile operator. Our approach suits the estimation of static population densities, i.e., of the distribution of dwelling units per urban area contained in traditional censuses. More importantly, it enables the estimation of dynamic population densities, i.e., the time-varying distributions of people in a conurbation. By leveraging substantial real-world mobile network metadata and ground-truth information, we demonstrate that the accuracy of our solution is superior to that granted by state-of-the-art methods in practical heterogeneous urban scenarios.

The economic sustainability of future mobile networks will largely depend on the strong specialization of its offered services. Network operators will need to provide added value to their tenants, by moving from the traditional one-size-fits-all strategy to a set of virtual end-to-end instances of a common physical infrastructure, named network slices, which are especially tailored to the requirements of each application. Implementing network slicing has significant consequences in terms of resource management: service customization entails assigning to each slice fully dedicated resources, which may also be dynamically reassigned and overbooked in order to increase the cost-efficiency of the system. In this paper, we adopt a data-driven approach to quantify the efficiency of resource sharing in future sliced networks. Building on metropolitan-scale real-world traffic measurements, we carry out an extensive parametric analysis that highlights how diverse performance guarantees, technological settings, and slice configurations impact the resource utilization at different levels of the infrastructure in presence of network slicing. Our results provide insights on the achievable efficiency of network slicing architectures, their dimensioning, and their interplay with resource management algorithms at different locations and reconfiguration timescales.

Upcoming mobile network technologies developed in the context of 5G and DSRC are expected to finally legitimize direct data transfers among vehicles as a standard communication paradigm. We investigate fundamental properties of the topology of vehicular networks built on top of these emerging vehicle-to-vehicle communication technologies. Our study yields multiple elements of originality: ( i) it addresses temporal connectivity, which has been poorly investigated despite a high relevance for vehicular network operations; (ii) it introduces exact but computationally efficient models of the temporal connectivity of vehicular networks; (iii) it evaluates the proposed models in urban settings that exhibit an unprecedented combination of dependability, scale and generality of vehicular mobility. This approach lets us unveil an apparent scale-and city-invariant law of temporal reachability in vehicular networks. Finally, we open our original scenarios to the research community, so as to ensure reproducibility of our results and foster further investigations of vehicular network performance.

The development of virtualization techniques enables an architectural shift in mobile networks, where resource allocation, or even signal processing, become software functions hosted in a data center. The centralization of computing resources and the dynamic mapping between baseband processing units (BBUs) and remote antennas (RRHs) provide an increased flexibility to mobile operators, with important reductions of operational costs. Most research efforts on Cloud Radio Access Networks (CRAN) consider indeed an operator perspective and network-side performance indicators. The impact of such new paradigms on user experience has been instead overlooked. In this paper, we shift the viewpoint, and show that the dynamic assignment of computing resources enabled by CRAN generates a new class of mobile terminal handover that can impair user quality of service. We then propose an algorithm that mitigates the problem, by optimizing the mapping between BBUs and RRHs on a time-varying graph representation of the system. Furthermore, we show that a practical online BBU-RRH mapping algorithm achieves results similar to an oracle-based scheme with perfect knowledge of future traffic demand. We test our algorithms with two large-scale real-world datasets, where the total number of handovers, compared with the current architectures, is reduced by more than 20%. Moreover, if a small tolerance to dropped calls is allowed, 30% less handovers can be obtained.

Employing mobile devices to perform data analytics is a typical fog computing application that utilizes the intelligence at the edge of networks. Such an application relies on the knowledge of the mobility of mobile devices and their users, e.g., to deploy computation tasks efficiently at the edge. This paper surveys the literature on the mobility-related utilization of operator-collected CDR (charging data records) - the most sig- nificant proxy of large-scale human mobility studies. We provide an innovative introductory guide to the CDR data preliminary. It reveals original issues regarding CDR-based mobility feature computation and applications at the edge. Our survey plays an important role in utilizing mobile devices in terms of both human mobility investigation and fog computing.

By providing especially tailored instances of a virtual network,network slicing allows for a strong specialization of the offered services on the same shared infrastructure. Network slicing has profound implications on resource management, as it entails an inherent trade-off between: (i) the need for fully dedicated resources to support service customization, and (ii) the dynamic resource sharing among services to increase resource efficiency and cost-effectiveness of the system. In this paper, we provide a first investigation of this trade-off via an empirical study of resource management efficiency in network slicing. Building on substantial measurement data collected in an operational mobile network (i) we quantify the efficiency gap introduced by non-reconfigurable allocation strategies of different kinds of resources, from radio access to the core of the network, and (ii) we quantify the advantages of their dynamic orchestration at different timescales. Our results provide insights on the achievable efficiency of network slicing architectures, their dimensioning, and their interplay with resource management algorithms.

We investigate how individual mobile services are consumed at a national scale, by studying data collected in a 3G/4G mobile network deployed over a major European country. Through correlation and clustering analyses, our study unveils a strong heterogeneity in the demand for different mobile services, both in time and space. In particular, we show that: (i) somehow surprisingly, almost all considered services exhibit quite different temporal usage patterns; (ii) in contrast to such temporal behavior, spatial patterns are fairly uniform across all services; (iii) when looking at usage patterns at different locations, the average traffic volume per user is dependent on the urbanization level, yet its temporal dynamics are not. Our findings do not only have sociological implications, but are also relevant to the orchestration of network resources.

In this paper, we aim at answering the question "at what frequency should one sample individual human movements so that they can be reconstructed from the collected samples with minimum loss of information?". Our quest for a response unveils (i) seemingly universal spectral properties of human mobility, and (ii) a linear scaling law of the localization error with respect to the sampling interval. We conduct analyses using fine-grained GPS trajectories of 119 users worldwide. Our findings have potential applications in ubiquitous computing and mobile service design, in terms of energy efficiency, location-based service operations, active probing of subscribers' positions in mobile networks and trajectory data compression.

Mobile traffic data collected by network operators is a rich source of information about human habits, and its analysis provides insights relevant to many fields, including urbanism, transportation, sociology and networking. In this paper, we present an original approach to infer both spatial and temporal structures hidden in the mobile demand, via a first-time tailoring of Exploratory Factor Analysis (EFA) techniques to the context of mobile traffic datasets. Casting our approach to the time or space dimensions of such datasets allows solving different problems in mobile traffic analysis, i.e., network activity profiling and land use detection, respectively. Tests with real-world mobile traffic datasets show that, in both its variants above, the proposed approach (i) yields results whose quality matches or exceeds that of state-of-the-art solutions, and (ii) provides additional joint spatiotemporal knowledge that is critical to result interpretation.

Mobile network operators can track subscribers via passive or active monitoring of device locations. The recorded trajectories offer an unprecedented outlook on the activities of large user populations, which enables developing new networking solutions and services, and scaling up studies across research disciplines. Yet, the disclosure of individual trajectories raises significant privacy concerns: thus, these data are often protected by restrictive non-disclosure agreements that limit their availability and impede potential usages. In this paper, we contribute to the development of technical solutions to the problem of privacy-preserving publishing of spatiotemporal trajectories of mobile subscribers. We propose an algorithm that generalizes the data so that they satisfy-anonymity, an original privacy criterion that thwarts attacks on trajectories. Evaluations with real-world datasets demonstrate that our algorithm attains its objective while retaining a substantial level of accuracy in the data. Our work is a step forward in the direction of open, privacy-preserving datasets of spatiotemporal trajectories.

In this paper, we leverage two large-scale real-world datasets to provide the first results on the limits of predictability of cellular data traffic demands generated by individual users over time and space. Using information theory tools, we measure the maximum predictability that any algorithm has potential to achieve. We first focus on the predictability of mobile traffic consumption patterns in isolation. Our results show that it is theoretically possible to anticipate the individual demand with a typical accuracy of 85% and reveal that this percentage is consistent across all user types. Then, we analyze the joint predictability of the traffic demands and mobility patterns. We find that the two dimensions are correlated, which improves the predictability upper bound to 90% on average.

The integration of automotive communications in 5G systems must build on a clear understanding of the performance of services for connected vehicles in today's LTE deployments. In this paper, we carry out a simulation-based performance evaluation of automotive communications in LTE, with particular attention to realism: to that end, we investigate the impact of different road traffic models, employ a state-of-the-art commercial LTE tool, and study a practical service use case. Our results demonstrate that unrealistic road traffic datasets can bias network simulations in urban vehicular environments, and provide insights on the limitations of the current radio access architecture, when confronted to connected vehicles.

Urban landscapes present a variety of socio-topological environments that are associated to diverse human activities. As the latter affect the way individuals connect with each other, a bound exists between the urban tissue and the mobile communication demand. In this paper, we investigate the heterogeneous patterns emerging in the mobile communication activity recorded within metropolitan regions. To that end, we introduce an original technique to identify classes of mobile traffic signatures that are distinctive of different urban fabrics. Our proposed technique outperforms previous approaches when confronted to ground-truth information, and allows characterizing the mobile demand in greater detail than that attained in the literature to date. We apply our technique to extensive real-world data collected by major mobile operators in 10 cities. Results unveil the diversity of baseline communication activities across countries, but also provide evidence of the existence of a number of mobile traffic signatures that are common to all studied areas and specific to particular land uses.

In the next few years, mobile networks will undergo significant evolutions in order to accommodate the ever-growing load generated by increasingly pervasive smartphones and connected objects. Among those evolutions, cognitive networking upholds a more dynamic management of network resources that adapts to the significant spatiotemporal fluctuations of the mobile demand. Cognitive networking techniques root in the capability of mining large amounts of mobile traffic data collected in the network, so as to understand the current resource utilization in an automated manner. In this paper, we take a first step towards cellular cognitive networks by proposing a framework that analyzes mobile operator data, builds profiles of the typical demand, and identifies unusual situations in network-wide usages. We evaluate our framework on two real-world mobile traffic datasets, and show how it extracts from these a limited number of meaningful mobile demand profiles. In addition, the proposed framework singles out a large number of outlying behaviors in both case studies, which are mapped to social events or technical issues in the network.

Understanding of the network topology is a basic building block towards the design of efficient networking solutions. In the context of vehicular networks, such a step is especially crucial due to the highly dynamic nature of vehicles that can lead to strong instantaneous variations in the structure of the network. This notwithstanding, and despite the soon-to-come real-world deployment of vehicle-to-vehicle communication technologies, we still lack a clear understanding of vehicular network topological properties. In this paper, we present a complex network analysis of the instantaneous topology of a realistic vehicular network in Cologne, Germany. Our study unveils a poorly connected topology, with very limited availability, reliability, and navigability. We also examine the vehicular network topology in a second scenario, i.e., Zurich, Switzerland. The comparative analysis shows how simplistic mobility models can lead to unrealistic overly connected topologies.

Smartphones and other mobile devices are today pervasive across the globe. As an interesting side effect of the surge in mobile communications, mobile network operators can now easily collect a wealth of high-resolution data on the habits of large user populations. The information extracted from mobile network traffic data is very relevant in the context of population mapping: it provides a tool for the automatic and live estimation of population densities, overcoming the limitations of traditional data sources such as censuses and surveys. In this paper, we propose a new approach to infer population densities at urban scales, based on aggregated mobile network traffic metadata. Our approach allows estimating both static and dynamic populations, achieves a significant improvement in terms of accuracy with respect to state-of-the-art solutions in the literature, and is validated on different city scenarios.

The dramatic increase in the number and sensing capabilities of mobile devices is fostering opportunistic sensing as a paramount data collection paradigm in smart cities. According to this paradigm, sensing of large-scale phenomena is autonomously performed by mobile devices that provide irregular samples in time and space. The collected data is then transferred to a central controller, and processed so as to obtain a representation of the phenomenon. In this paper, we investigate the factors that impact the accuracy of mobile opportunistic sensing. Specifically, we characterize the accuracy of a phenomenon representation obtained from samples collected by mobile devices and processed through the popular LMMSE filter. We do so by drawing on random matrix theory, which allows us to deal with irregularly spaced samples. Our analytical expressions capture the fundamental relationships existing between the accuracy and the parameters of mobile opportunistic sensing. We apply our analytical results to a realistic scenario where atmospheric pollution samples are collected by vehicular and pedestrian users. We validate the proposed analytical framework, and then exploit the model to investigate the impact on mobile sensing accuracy of a number of parameters. These include the pedestrian and vehicle density, the participation ratio to the sensing application, the type of phenomenon to be sensed, and the level of noise and position errors affecting the collected samples.

Mobile traffic data has been recently used to characterize the urban environment in terms of urban fabric profiles. While showing promising results, the existing urban fabric detection solutions are built without a clear understanding of the detection process chain. In this paper, we distinguish and analyze the different steps common to all urban profiling techniques. By evaluating the impact of each step of the process, we are able to propose a new solution that outperforms the state of the art techniques. Our approach uses the weekly periodicity of human activities, as well as a median-based filtering technique, resulting in a better clustering in terms of both coverage and entropy, as shown by results obtained on two large scale mobile traffic datasets covering the urban areas of Milan and Turin, in Italy.

Preservation of user privacy is paramount in the publication of datasets that contain fine-grained information about individuals. The problem is especially critical in the case of mobile traffic datasets collected by cellular operators, as they feature high subscriber trajectory uniqueness and they are resistant to anonymization through spatiotemporal generalization. In this work, we first unveil the reasons behind such undesirable features of mobile traffic datasets, by leveraging an original measure of the anonymizability of users' mobile fingerprints. Building on such findings, we propose GLOVE, an algorithm that grants k-anonymity of trajectories through specialized generalization. We evaluate our methodology on two nationwide mobile traffic datasets, and show that it achieves k-anonymity while preserving a substantial level of accuracy in the data.

Connected vehicles promise to enable a wide range of new automotive services that will improve road safety, ease traffic management, and make the overall travel experience more enjoyable. However, they also open significant new surfaces for attacks on the electronics that control most of modern vehicle operations. In particular, the emergence of vehicle-to-vehicle (V2V) communication risks to lay fertile ground for self-propagating mobile malware that targets automobile environments. In this work, we perform a first study on the dynamics of vehicular malware epidemics in a large-scale road network, and unveil how a reasonably fast worm can easily infect thousands of vehicles in minutes. We determine how such dynamics are affected by a number of parameters, including the diffusion of the vulnerability, the penetration ratio and range of the V2V communication technology, or the worm self-propagation mechanism. We also propose a simple yet very effective numerical model of the worm spreading process, and prove it to be able to mimic the results of computationally expensive network simulations. Finally, we leverage the model to characterize the dangerousness of the geographical location where the worm is first injected, as well as for efficient containment of the epidemics through the cellular network.

Cellular communications are undergoing significant evolutions in order to accommodate the load generated by increasingly pervasive smart mobile devices. Dynamic access network adaptation to customers' demands is one of the most promising paths taken by network operators. To that end, one must be able to process large amount of mobile traffic data and outline the network utilization in an automated manner. In this paper, we propose a framework to analyze broad sets of Call Detail Records (CDRs) so as to define categories of mobile call profiles and classify network usages accordingly. We evaluate our framework on a CDR dataset including more than 300 million calls recorded in an urban area over 5 months. We show how our approach allows to classify similar network usage profiles and to tell apart normal and outlying call behaviors.

There is a growing need for vehicular mobility datasets that can be employed in the simulative evaluation of protocols and architectures designed for upcoming vehicular networks. Such datasets should be realistic, publicly available, and heterogeneous, i.e., they should capture varied traffic conditions. In this paper, we contribute to the ongoing effort to define such mobility scenarios by introducing a novel set of traces for vehicular network simulation. Our traces are derived from high-resolution real-world traffic counts, and describe the road traffic on two highways around Madrid, Spain, at several hours of different working days. We provide a thorough discussion of the real-world data underlying our study, and of the synthetic trace generation process. Finally, we assess the potential impact of our dataset on networking studies, by characterizing the connectivity of vehicular networks built on the different traces. Our results underscore the dramatic impact that relatively small communication range variations have on the network. Also, they unveil previously unknown temporal dynamics of the topology of highway vehicular networks, and identify their causes.

A number of vehicular networking applications require continuous knowledge of the location of vehicles and tracking of the routes they follow, including, e.g., real-time traffic monitoring, e-tolling, and liability attribution in case of accidents. Locating and tracking vehicles has however strong implications in terms of security and user privacy. On the one hand, there should be a mean for an authority to verify the correctness of positioning information announced by a vehicle, so as to identify potentially misbehaving cars. On the other, public disclosure of identity and position of drivers should be avoided, so as not to jeopardize user privacy. In this paper, we address such issues by introducing A-VIP, a secure, privacy-preserving framework for continuous tracking of vehicles. A-VIP leverages anonymous position beacons from vehicles, and the cooperation of nearby cars collecting and reporting the beacons they hear. Such information allows a location authority to verify the positions announced by vehicles, or to infer the actual ones if needed, without resorting to computationally expensive asymmetric cryptography. We assess the effectiveness of A-VIP via realistic simulation and experimental testbeds.

Bandwidth availability in the cellular backhaul is challenged by ever-increasing demand by mobile users. Vehicular users, in particular, are likely to retrieve large quantities of data, choking the cellular infrastructure along major thoroughfares and in urban areas. It is envisioned that alternative roadside network connectivity can play an important role in offloading the cellular infrastructure. We investigate the effectiveness of vehicular networks in this task, considering that roadside units can exploit mobility prediction to decide which data they should fetch from the Internet and to schedule transmissions to vehicles. Rather than adopting a specific prediction scheme, we propose a fog-of-war model that allows us to express and account for different degrees of prediction accuracy in a simple, yet effective, manner. We show that our fog-of-war model can closely reproduce the prediction accuracy of Markovian techniques. We then provide a probabilistic graph-based representation of the system that includes the prediction information and lets us optimize content prefetching and transmission scheduling. Analytical and simulation results show that our approach to content downloading through vehicular networks can achieve a 70% offload of the cellular network.

The surge in vehicular network research has led, over the last few years, to the proposal of countless network solutions specifically designed for vehicular environments. A vast majority of such solutions has been evaluated by means of simulation, since experimental and analytical approaches are often impractical and intractable, respectively. The reliability of the simulative evaluation is thus paramount to the performance analysis of vehicular networks, and the first distinctive feature that has to be properly accounted for is the mobility of vehicles, i.e., network nodes. Notwithstanding the improvements that vehicular mobility modeling has undergone over the last decade, no vehicular mobility dataset is publicly available today that captures both the macroscopic and microscopic dynamics of road traffic over a large urban region. In this paper, we present a realistic synthetic dataset, covering 24 hours of car traffic in a 400-km2region around the city of Köln, in Germany. We describe the generation process and outline how the dataset improves the traces currently employed for the simulative evaluation of vehicular networks. We also show the potential impact that such a comprehensive mobility dataset has on the network protocol performance analysis, demonstrating how incomplete representations of vehicular mobility may result in over-optimistic network connectivity and protocol performance.

Knowledge of the location of vehicles and tracking of the routes they follow are a requirement for a number of applications. However, public disclosure of the identity and position of drivers jeopardizes user privacy, and securing the tracking through asymmetric cryptography may have an exceedingly high computational cost. In this paper, we address all of the issues above by introducing A-VIP, a lightweight privacy-preserving framework for tracking of vehicles. A-VIP leverages anonymous position beacons from vehicles, and the cooperation of nearby cars collecting and reporting the beacons they hear. Such information allows an authority to verify the locations announced by vehicles, or to infer the actual ones if needed. We assess the effectiveness of A-VIP through testbed implementation results.

Efficient schemes for warning message dissemination in vehicular ad hoc networks (VANETs) use context information collected by vehicles about their neighbor nodes to guide the dissemination process. These schemes maximize their performance when all the vehicles advertise correct information about their positions, and hence position errors may drastically reduce the performance of the dissemination process. We present a proactive Cooperative Neighbor Position and Verification (CNPV) protocol that detects nodes advertising false locations so as to mitigate the impact of adversarial users. We combine our mechanism with two warning dissemination schemes for VANETs, and demonstrate how these algorithms can benefit from the use of our security scheme in the presence of malicious nodes trying to exploit the inherent vulnerabilities of each algorithm.

Despite the growing interest in a real-world deployment of vehicle- to-vehicle communication, many topological features of the resulting vehicular network remain largely unknown. We still lack a clear understanding of the level of connectivity achievable in large-scale urban scenarios, of the availability and reliability of connected multi-hop paths, and of the evolution of such features over daytime. In this paper, we investigate how the instantaneous topology of the vehicular network would look like in the case of Cologne, Germany, a typical middle-sized European city. Through a complex network analysis, we unveil the low connectivity, availability, reliability and navigability of the network, and exploit our findings to derive network design and usage guidelines.

In this paper, we argue that the ability to accurately spot random and social relationships in dynamic networks is essential to network applications that rely on human routines, such as, e.g., opportunistic routing. We thus propose a strategy to analyze users' interactions in mobile networks where users act according to their interests and activity dynamics. Our strategy, named Random rElationship ClASsifier sTrategy (RECAST), allows classifying users' wireless interactions, separating random interactions from different kinds of social ties. To that end, RECAST observes how the real system differs from an equivalent one where entities' decisions are completely random. We evaluate the effectiveness of the RECAST classification on real-world user contact datasets collected in diverse networking contexts. Our analysis unveils significant differences among the dynamics of users' wireless interactions in the datasets, which we leverage to unveil the impact of social ties on opportunistic routing.

Floating Car Data (FCD) is currently collected by moving vehicles and uploaded to Internet-based processing centers through the cellular access infrastructure. As FCD is foreseen to rapidly become a pervasive technology, the present network paradigm risks not to scale well in the future, when a vast majority of automobiles will be constantly sensing their operation as well as the external environment and transmitting such information towards the Internet. In order to relieve the cellular network from the additional load that widespread FCD can induce, we study a local gathering and fusion paradigm, based on vehicle-to-vehicle (V2V) communication. We show how this approach can lead to significant gain, especially when and where the cellular network is stressed the most. Moreover, we propose several distributed schemes to FCD offloading based on the principle above that, despite their simplicity, are extremely efficient and can reduce the FCD capacity demand at the access network by up to 95%.

The large-scale adoption of vehicle-to-vehicle (V2V) communication technologies risks to significantly widen the attack surface available to mobile malware targeting critical automobile operations. Given that outbreaks of vehicular computer worms self-propagating through V2V links could pose a significant threat to road traffic safety, it is important to understand the dynamics of such epidemics and to prepare adequate countermeasures. In this paper we perform a comprehensive characterization of the infection process of variously behaving vehicular worms on a road traffic scenario of unprecedented scale and heterogeneity. We then propose a simple yet effective data-driven model of the worm epidemics, and we show how it can be leveraged for smart patching infected vehicles through the cellular network in presence of a vehicular worm outbreak.

We present a communication protocol, called LINGER, for persistent dissemination of delay-tolerant information to vehicular users, within a geographical area of interest. The goal of LINGER is to dispatch and confine information in localized areas of a mobile network with minimal protocol overhead and without requiring knowledge of the vehicles' routes or destinations. LINGER does not require roadside infrastructure support: it selects mobile nodes in a distributed, cooperative way and lets them act as "information bearers", providing uninterrupted information availability within a desired region. We analyze the performance of our dissemination mechanism through extensive simulations, in complex vehicular scenarios with realistic node mobility. The results demonstrate that LINGER represents a viable, appealing alternative to infrastructure-based solutions, as it can successfully drive the information toward a region of interest from a far away source and keep it local with negligible overhead. We show the effectiveness of such an approach in the support of localized broadcasting, in terms of both percentage of informed vehicles and information delivery delay, and we compare its performance to that of a dedicated, state-of-the-art protocol.

A growing number of ad hoc networking protocols and location-aware services require that mobile nodes learn the position of their neighbors. However, such a process can be easily abused or disrupted by adversarial nodes. In absence of a priori trusted nodes, the discovery and verification of neighbor positions presents challenges that have been scarcely investigated in the literature. In this paper, we address this open issue by proposing a fully distributed cooperative solution that is robust against independent and colluding adversaries, and can be impaired only by an overwhelming presence of adversaries. Results show that our protocol can thwart more than 99 percent of the attacks under the best possible conditions for the adversaries, with minimal false positive rates.

We consider a system where users aboard communication-enabled vehicles are interested in downloading different contents from Internet-based servers. This scenario captures many of the infotainment services that vehicular communication is envisioned to enable, including news reporting, navigation maps, and software updating, or multimedia file downloading. In this paper, we outline the performance limits of such a vehicular content downloading system by modeling the downloading process as an optimization problem, and maximizing the overall system throughput. Our approach allows us to investigate the impact of different factors, such as the roadside infrastructure deployment, the vehicle-to-vehicle relaying, and the penetration rate of the communication technology, even in presence of large instances of the problem. Results highlight the existence of two operational regimes at different penetration rates and the importance of an efficient, yet 2-hop constrained, vehicle-to-vehicle relaying.

Content downloading by mobile users is expected to significantly increase the cellular network load. Vehicular users, in particular, are likely to engage in information retrieval on the move: in this context, Intelligent Transportation Systems (ITS) can play an important role in offloading the cellular infrastructure. We investigate the effectiveness of ITS in this task, considering that roadside units (RSUs) can exploit mobility prediction to decide which data they should fetch from the Internet and schedule transmissions to vehicles, either potential relays or downloaders. Rather than presenting a specific prediction scheme, we propose a model that allows us to express and account for any prediction technique in a simple, yet effective, manner. We then provide a probabilistic graph-based representation of the system that accounts for the prediction uncertainty. We use such a representation to study the network dynamics by efficiently solving a (non-integer) LP problem. Our results show that the above approach to content downloading through ITS can achieve an 80% offload of the cellular network. Also, we investigate the dependency of the system performance on the accuracy of the mobility prediction, and which prediction errors have the largest impact.

Performance and reliability of content access in mobile networks is conditioned by the number and location of content replicas deployed at the network nodes. In this work, we design a practical, distributed solution to content replication that is suitable for dynamic environments and achieves load balancing. Simulation results show that our mechanism, which uses local measurements only, approximates well an optimal solution while being robust against network and demand dynamics. Also, our scheme outperforms alternative approaches in terms of both content access delay and access congestion.

We consider a complex (i.e., nonlinear) road scenario where users aboard vehicles equipped with communication interfaces are interested in downloading large files from road-side Access Points (APs). We investigate the possibility of exploiting opportunistic encounters among mobile nodes so to augment the transfer rate experienced by vehicular downloaders. To that end, we devise solutions for the selection of carriers and data chunks at the APs, and evaluate them in real-world road topologies, under different AP deployment strategies. Through extensive simulations, we show that carry&forward transfers can significantly increase the download rate of vehicular users in urban/suburban environments, and that such a result holds throughout diverse mobility scenarios, AP placements and network loads.

Content downloading in vehicular networks is a topic of increasing interest: services based upon it are expected to be hugely popular and investments are planned for wireless roadside infrastructure to support it. We focus on a content downloading system leveraging both infrastructure-to-vehicle and vehicle-to-vehicle communication. With the goal to maximize the system throughput, we formulate a max-flow problem that accounts for several practical aspects, including channel contention and the data transfer paradigm. Through our study, we identify the factors that have the largest impact on the performance and derive guidelines for the design of the vehicular network and of the roadside infrastructure supporting it.

We propose a fully-distributed approach to the on line estimation of vehicle traffic density. Our approach envisions vehicles communicating within a VANET and cooperating to collect density measurements through a uniform sampling of the road sections of interest. The proposed scheme does not require the presence of any network infrastructure, central controller or devices triggered by the passage of vehicles, and it is suitable for both highway and urban environments. Results derived through ns-2 simulations in realistic mobility scenarios show that our solution is very effective, providing accurate, on-line estimates of the traffic density with minimal protocol overhead.

Performance and reliability of content access in mobile networks is conditioned jointly by the number and location of content replicas deployed at the network nodes. The endeavour of this work is to address such an optimization problem with a distributed, lightweight solution that handles network dynamics. We devise a mechanism that lets nodes share the burden of storing and providing content, so as to achieve load balancing, and decide whether to replicate or drop the information so as to adapt to a dynamic content demand and time-varying topology. Simulation results show that our mechanism, which uses local measurements only, is: (i) extremely precise in approximating an optimal solution to content placement and replication; (ii) robust against network mobility; (iii) flexible in accommodating variation in time and space of the content demand.

We address cooperative caching in mobile ad hoc networks where information is exchanged in a peer-to-peer fashion among the network nodes. Our objective is to devise a fully-distributed caching strategy whereby nodes, independently of each other, decide whether to cache or not some content, and for how long. Each node takes this decision according to its perception of what nearby users may be storing in their caches and with the aim to differentiate its own cache content from the others'. We aptly named such algorithm "Hamlet". The result is the creation of a content diversity within the nodes neighborhood, so that a requesting user likely finds the desired information nearby. We simulate our caching algorithm in an urban scenario, featuring vehicular mobility, as well as in a mall scenario with pedestrians carrying mobile devices. Comparison with other caching schemes under different forwarding strategies confirms that Hamlet succeeds in creating the desired content diversity thus leading to a resource-efficient information access.

We investigate the problem of spreading information contents in a wireless ad hoc network. In our vision, information dissemination should satisfy the following requirements: (i) it should result in a desirable distribution of information replicas in the network and (ii) the information should be evenly and fairly carried by all nodes in their turn. In this paper, we show that these goals can be achieved by simple cache-and-forward mechanisms inspired by well-known node mobility models, provided that a sufficient number of information replicas are injected into the network. The proposed approach works under different network scenarios, is fully distributed and comes at a very low cost in terms of protocol overhead.

We target urban scenarios where vehicular users can download large files from road-side access points (APs), and define a framework to exploit opportunistic encounters between mobile nodes to increase their transfer rate. We first devise a technique for APs deployment, based on vehicular traffic flows analysis, which fosters cooperative download. Then, we propose and evaluate different algorithms for carriers selection and chunk scheduling in carry&forward data transfers. Results obtained under realistic road topology and vehicular mobility conditions show that coupling our APs deployment scheme with probabilistic carriers selection and redundant chunk scheduling yields a worst-case 2x gain in the average download rate with respect to direct download, as well as a 10x reduction in the rate of undelivered chunks with respect to a blind carry&forward.

We propose a framework, called LINGER, for the support of cooperative creation and distribution of information contents in vehicular networks. The goal of LINGER is to dispatch and confine information in localized areas of a mobile network with no infrastructure availability and minimal protocol overhead. LINGER selects mobile nodes in a distributed, cooperative way and lets them act as "information bearers", ensuring uninterrupted information availability to as many nodes as possible in a desired region. Simulation results in vehicular scenarios with realistic node mobility prove that LINGER successfully drives information toward a target area from a far away source and keeps it local with negligible overhead. Further tests with a beaconing application leveraging the LINGER framework show that a mobile information bearer may be as reliable as an infrastructure-based access point in providing service to users. Finally, in a large-scale scenario, LINGER is proven to be effective for delay-tolerant broadcast applications.

In this work we define and study a new model for the caching problem in a heterogeneous wireless network under a flash-crowd scenario. Using non-cooperative game theory, we cast the caching problem as an anti-coordination game. We start by defining the social optimum in the general case and then focus on a two-player game to obtain insights into the design of efficient caching strategies. Based the theoretical findings, our current work focuses on the development of strategies to be implemented in a practical network setting.

We consider that a given number of dissemination points (DPs) have to be deployed for disseminating information to vehicles travelling in an urban area. We formulate our problem as a maximum coverage problem (MCP) so as to maximize the number of vehicles that get in contact with the DPs and as a second step with a sufficient amount of time. Since the MCP is NP-hard, we solve it though heuristic algorithms. Evaluation of the proposed solutions in a realistic urban environment shows how knowledge of vehicular mobility plays a major role in achieving an optimal coverage of mobile users, and that simple heuristics provide near-optimal results even in large-scale scenarios.

The paper focuses on a cooperative environment in wireless ad hoc networks, where mobile nodes share information in a peer-to-peer fashion. Nodes follow a pure peer-to-peer approach (i.e., without the intervention of servers), thus requiring an efficient query/response propagation algorithm to prevent network congestion. The main contribution of the paper is the proposal of a novel solution, called Eureka, that identifies the regions of the network where the required information is more likely to be stored and steers the queries toward those regions. To discriminate among regions, the concept of information density is introduced, along with a procedure that allows nodes its estimation. Eureka does not require the use of satellite positioning systems, and proves to be very effective in both vehicular and pedestrian environments.

Mobility is the distinguishing feature of vehicular networks, affecting the evolution of network connectivity over space and time in a unique way. Connectivity dynamics, in turn, determine the performance of networking protocols, when they are employed in vehicle-based, large-scale communication systems. Thus, a key question in vehicular networking is: which effects does nodes mobility generate on the topology of a network built over vehicles? Surprisingly, such a question has been quite overlooked by the networking research community. In this paper, we present an in-depth analysis of the topological properties of a vehicular network, unveiling the physical reasons behind the peculiar connectivity dynamics generated by a number of mobility models. Results make one think about the validity of studies conducted under unrealistic car mobility and stimulate interesting considerations on how network protocols could take advantage of vehicular mobility to improve their performance.

Deployments of city-wide multi-hop 802.11 networks introduce challenges for maintaining client performance at vehicular speeds. We experimentally demonstrate that current network interfaces employ policies that result in long outage durations, even when clients are always in range of at least one access point. Consequently, we design and evaluate a family of client-driven handoff techniques that target vehicular mobility in multi-tier multi-hop wireless mesh networks. Our key technique is for clients to invoke an association change based on (i) joint use of channel quality measurements and AP quality scores that reflect long-term differences in AP performance and (ii) controlled measurement and hand-off time scales to balance the need for the instantaneously best association against performance penalties incurred from spurious handoffs due to channel fluctuations and marginally improved associations. We utilize a 4,000 user urban deployment to evaluate the performance of a broad class of hand-off policies.

We consider a cooperative environment in wireless mobile networks where information is exchanged among nodes in a peer-to-peer fashion. We apply a pure peer-to-peer approach (i.e., without the intervention of servers) and we seek to devise an efficient query/response propagation algorithm. Our approach, calledEureka, identifies the regions of the network where the required information is more likely to be stored and steers the queries toward those regions. To discriminate among regions, we introduce the concept ofinformation densityand a procedure that allows nodes its estimation. The effectiveness of our scheme is evaluated through simulation in a vehicular environment with realistic mobility models.

Within the extent of research activity in the field of vehicular networking, notwithstanding the quantity of simulative studies carried out on protocols performance, the impact of vehicular mobility characterization has been often overlooked. Realistic mobility models are seldom employed, and when used, no comment is usually provided on the different effects that such models have on the results with respect to simpler mobility descriptions. In this paper, we address this issue, and analyze the impact that various levels of details in vehicular mobility modeling have on the simulation of networking protocols.

Multiple description coding (MDC) combined with multi-path transmission is a viable solution to the growing demand of reliable multimedia video communication over the Internet. Despite the progress in MDC techniques, transport protocols based on TCP cannot efficiently handle multipath transmissions of this kind of traffic. In this paper we present MD-SCTP, a full-multipath partially-reliable SCTP-based protocol, providing different scheduling schemes and featuring a selective retransmission within time-to-delivery constraints of multimedia traffic. Performance analysis proves the validity of our scheduler and of our selective retransmission scheme.

A consequence of the increasing availability of multiple network interfaces on the same computer system is the request for new technologies capable of fully exploiting the available resources. As a matter of fact, today's most common transport and network layer protocols are unable to use more than one interface at a time, thus wasting part of the available network resources. However, the growing demand for bandwidth and quality of service introduced by multimedia applications in particular could find an answer in the correct utilization of multipath connections. In this paper, we present Westwood SCTP-PR, a partially-reliable SCTP-based protocol, providing full multipath support and featuring an original adaptive load balancing technique, characteristics which make it especially suitable to real-time traffic.

We propose an information-sharing application for wireless intervehicular networks (IVNs), called Infoshare. The Infoshare application leverages the broadcast nature of the wireless medium to achieve maximum spreading of information queries among vehicles, while a smart caching policy limits the overhead resulting from useless queries and duplicated replies. Simulation by ns2 is used to investigate the performance of Infoshare, highlighting the impact of various system parameters on spreading dynamics. A network scenario featuring one- and two-lane traffic traveling either at the same or at different speeds is considered. Simulation results come in handy for the development of analytical models, identifying critical parameters and justifying assumptions.