This paper presents a machine learning-based approach for network intrusion detection. It relies on a fast binary classifier to quickly distinguish between benign and malicious traffic, being the actual type of attack detected in an independent second stage. The model was trained and evaluated using a benchmark dataset that combines the well known CIC-IDS2017, CIC-IDS2018, and UNSW-NB15 datasets, following extensive preprocessing and exploratory data analysis. We compared multiple algorithms (Neural Networks, XGBoost, Random Forest, and Logistic Regression) based on standard metrics like precision, F1 score, ROC-AUC, and inference time, among others. Experimental results show that XGBoost consistently achieves the best balance between classification performance and deployment efficiency.

Climate change is one of the main problems that humanity is facing, and reducing air pollution is one of the actions that must be taken to fight it. Before we can reduce it, we must be able to accurately measure the current air pollution and identify the contamination sources. In this work, we propose the use of unmanned aerial vehicles (UAVs) to measure the air pollution and identify the highest contaminated areas (i.e. the pollution source). We developed an automatic multicopter guidance system that is capable of obtaining pollution data from an area in an efficient, autonomous, and precise way. Our guidance algorithms move a UAV based on the data that it measures at each location, in such a way that the UAV spends most of the time flying/measuring in areas where the pollution is higher. Hence, we are able to precisely map the contamination sources while reducing flight time by up to 85% compared to a full sweep of the scenario.

The goal of this paper is to evaluate the dissemination of information using traces of real vehicles from two cities: Rome and San Francisco. Concretely, we perform an analysis of the temporal and spatial characteristics of these traces. Then, using the Epidemic Protocol, we evaluate the diffusion of the information according to the size of the message. As expected, the experiments show that the size of the message has a great impact on the dissemination of these messages. Eventually, we propose an improvement to the Epidemic protocol, called Epidem icX2, which is based on the division of large messages to increase their delivery opportunity. The results show that EpidemicX2 increases the diffusion (number of vehicles receiving the message) reducing the delivery time and the incurred overhead when compared to the Epidemic protocol.

Vehicular Social Networks (VSNs) consist of groups of individuals (i.e., people) who may share common interests, preferences and needs in the context of temporal spatial proximity on roads. In this environment, the impact of human social factors, such as mobility, willingness to cooperate and personal preferences, on vehicular connectivity is taken under consideration, thus extending the concept of Vehicular Ad-hoc Networks. In VSNs, vehicles are classified based on their social degree, a vehicle considered to be a “social” one if it accesses the vehicular social network and posts messages with a frequency higher than a given threshold. Therefore, to speed up the data dissemination process within a vehicular social network, a packet should be forwarded to those vehicles showing high social activity. In a previous paper, we introduced a new probabilistic-based broadcasting scheme called SCARF (SoCial-Aware Reliable Forwarding Technique for Vehicular Communications), and we analytically demonstrated its effectiveness in packet transmission reduction while guaranteeing network dissemination. In this paper, we assess SCARF in more realistic scenarios with real traffic traces, and we compare it with other similar techniques. We show that SCARF outperforms other approaches in terms of delivery ratio, while guaranteeing acceptable time delay values and average number of forwardings.

Unmanned Aerial Vehicles (UAVs), popularly known as drones, are foreseen as mobile infrastructures that support communications when a fixed infrastructure is missing in vehicular networks. UAVs can act as message relays between ground vehicles or broadcast alerts in emergency situations. Simulation that involves UAVs, combined with ground vehicles, should support 3D transmission features as the drone is not positioned in a flat surface. Results from real test bed experiments showed that irregular terrains that form hills and mountains can act as obstacles that limit the communications range. In this paper, we propose a simulation framework that runs within the OMNeT++ simulator which exhibits results comparable to the ones obtained in the real test bed, when applied to different scenarios. In the simulation, a measurement of the communications quality between UAVs and cars that considers 3D real-world terrain features which will have an impact on signal attenuation shows that the level of realism has improved when compared to simulation experiments that only consider planar communications.

Unmanned Aerial Vehicles (UAVs) can act as re-lays in areas with limited infrastructure to support car-to-car communications. Prior studies on UAV-to-car communications showed that the irregularity of the terrains has a significant impact on link quality. Thus, in this paper, we propose a positioning technique that relies on Particle Swarm optimization (PSO) to optimize the positioning of a UAV in the vehicular environment by considering the irregularities of the terrains that might hinder Line-of-Sight (LOS) conditions. The proposed technique takes into account the path loss caused by the terrains. Simulation results show that the optimization algorithm allows us to determine the best position for the deployed UAVs throughout time by considering the movement of the cars, and also accounting for adjustments in terms of flight altitude. In particular, the latter is adjusted by considering the position of the cars on the ground and the profile of surrounding terrains to determine potential communications blockages, while respecting international regulations regarding flight altitude restrictions.

Opportunistic Networking (OppNet) is a consolidated networking paradigm based on intermittent connectivity between wireless mobile devices. In OppNet, mobile nodes store, carry and forward messages by taking advantage of wireless ad-hoc communication opportunities. A common approach for the diffusion of information in OppNet is the epidemic protocol, which carries out a fast information diffusion at the expense of increasing the usage of local buffers on mobile nodes. Nevertheless, due to this high local buffer usage and the usually reduced hardware capabilities of Internet of Things (IoT) devices, OppNet utilisation for IoT applications can be severely reduced. Therefore, a way to reduce the local buffer usage is to set a message expiration timer that forces the removal of old messages in the local buffers. Nevertheless, since dropping messages may reduce the speed of message diffusion, we propose to conveniently adjust this expiration timer, using a model to obtain the optimal expiration times that achieve performances similar to those ideal approaches where no expiration is considered, with a significant reduction of local buffer usage.

Unmanned Aerial Vehicles (UAVs) are being widely used as a valuable tool in critical situations, especially when the access to the target area is difficult or hazardous. In these situations, a swarm of UAVs can provide additional benefits by covering larger areas, thereby speeding up the fulfillment of a mission, or by carrying heavier loads that a single UAV would not support. In this paper we present MUSCOP, a protocol that allows a set of UAVs to properly synchronize their flight in the fulfillment of planned missions. The results obtained show that MUSCOP achieves a high degree of swarm cohesion even in the presence of channel losses, introducing minimum synchronization delays and low position offset errors compared to an ideal scenario.

Message diffusion in Vehicular Networks takes place upon the establishment of ephemeral contacts among vehicles. This approach has one severe limitation: since the amount of information transmitted in a contact is limited by the transmission speed and the contact duration, large messages are not likely to be exchanged, and thus their diffusion is severely limited. In this paper, we study the impact that different social aspects have on the spreading of messages, showing that splitting a large message into smaller parts can improve its diffusion. Based on this idea, we propose an extension of the epidemic protocol called Xpread. We have developed an analytical model based on Population Processes to evaluate the impact and the efficiency of the partition scheme, showing that a fixed size partition is the best option, while also providing a simple expression to obtain the optimal size. Finally, we show that the diffusion of the messages is improved up to two times with a slight reduction in the delivery time and overhead. These results were also confirmed with simulation experiments.

The performance of Opportunistic Networks relies mainly on users mobility. It is in fact mobility that creates the opportunities for contacts and therefore for data forwarding. The evaluation of these networks is usually based on either synthetic mobility models or real mobility traces generally characterized by a fixed number of users. In this paper, we focus on a mostly unexplored area characterized by crowded scenarios with people renewal, i.e., with users that can either enter or leave the evaluated scenario. By using a pedestrian mobility simulator we define realistic people mobility traces that allow the evaluation of different degrees of users densities and renewal rates. Using this methodology, we studied the performance of various existing routing protocols in scenarios with high degree of people renewal and by varying the messages sizes. The experiments confirm that the renewal rate has an important impact on the performance of the protocols, which becomes particularly evident when the message size is large. Overall, we observe that controlled flooding algorithm such as Spray & Wait can obtain good packet delivery results with lower overhead than with respect to probability based protocols, avoiding also the implementation complexity of the latters.

Vehicular networks are gradually emerging due to the expected benefits in terms of enhanced safety and infotainment services. However, outside main metropolitan areas, little infrastructure currently deployed, which may hinder these services. To mitigate this problem, Unmanned Aerial Vehicles (UAVs) are envisioned as mobile infrastructure elements, supporting communications when fixed infrastructure is missing. This way, in emergency situations, UAVs can offer services to vehicles including broadcasting alerts or acting as message relays between ground vehicles. Our work attempts to be a first step in this direction by presenting experimental measurement results regarding communications quality between cars and UAVs. In particular, we varied the altitude of the drone and its antenna orientation, and the car's antenna location to assess their impact on performance. Based on the experimental results achieved, we find that UAVs communicating in the 5 GHz band using IEEE 802.11 technology are able to deliver data to moving cars within a range of more than three kilometers, achieving more than 0.5 of packet delivery ratio up to 2.5 kilometers under the optimal configuration settings.

The diffusion of information using Opportunistic Networks relies mainly on users mobility. In fact, mobility creates the opportunities for contacts and therefore for data forwarding. The evaluation of this process in crowded spaces with people renewal, i.e. with users that can either enter or leave the evaluated scenario, is still basically unexplored, despite being one of the most suitable areas of application for these networks. In this paper, we focus on evaluating the dissemination of information in a subway station. We obtained real data from a local subway station through observation. These data were used in a pedestrian mobility simulator to generate people mobility traces to be used in an Opportunistic Network simulator. The analysis of the temporal and spatial characteristics of these traces reflects the realism of the generated scenario. In the experiments, we compare the direct delivery and epidemic diffusion protocols. As expected, epidemic diffusion compared to direct delivery increases drastically the number of users that receive the message. Furthermore, we propose an improvement of this protocol, called EpidemicX2 that is particularly effective for large messages.

Opportunistic networks are a form of mobile ad-hoc networks that exploit users mobility to provide data sharing. Individual nodes store, carry and forward messages using direct communication between devices; no fixed infrastructure of communication is required. The final goal is to reach the totality of the participating users. The efficiency of message diffusion depends on the users' mobility, the transfer time, and the device buffer management. User's mobility determines the number of contacts and their duration while the transfer time depends on the size of the interchanged messages and the channel throughput. The device buffer management is a critical factor since, due to its limited size, the chosen policy for forwarding messages when a connection is available and for message dropping when the buffer is full can impact greatly on the diffusion of the messages. In this paper we focus on this issue, evaluating the impact of different buffer management approaches in two different scenarios: pedestrian and vehicular. The results show that the best buffer management is the combination of smallest message forwarding and largest message dropping. We also show that a TTL (Time to Live) of 24 hours allows the best diffusion because it corresponds to the daily movement pattern of the nodes.

Intersection management is one of the challenges of vehicular communications in urban environments. Maximizing message delivery at intersections requires gaining awareness about when the vehicle is at the center of intersection, as buildings tend to obstruct wireless signals especially in the 5 GHz band. This paper addresses the issue of communications on different types of intersections by characterizing each intersection type through both real experiments and analytic modeling. The study has found that antenna location, distance to the intersection and line of sight conditions are all important factors for vehicular communications at intersections. Furthermore, this communication success also depends on the intersections characteristics, as different degrees of obstruction cause different delivery probabilities of sending messages.

The duration of batteries is a key factor that must be considered in the design and deployment of applications and services on mobile devices. The users of this technology are not aware of the energetic impact of the applications executing in their devices, especially if these applications present an adaptive behavior. The goal of this paper is to study the power consumption associated to an opportunistic communication application. This will be achieved by analyzing the different operation stages of an application running on mobile and vehicular devices, using basic statistical analysis calculations on data collected by a logging application. Furthermore, this will allow analysing the impact of this kind of applications regarding battery performance, providing useful information to users and developers. The experiments were executed in an opportunistic network environment, using a floating content approach as a mechanism for information distribution in a geographically delimited zone.

Nowadays, the impact of global warming is causing societies to become more aware and responsive to environmental problems. As a result, pollution sensing is gaining more relevance. In order to have a strict control over air quality, the use of mobile sensors is becoming a promising alternative to traditional air quality stations. Mobile sensors allow to easily perform measurements in many different places, thereby offering substantial improvements in terms of the spatial granularity of the data gathered. Pollution monitoring near large industrial areas or in rural areas where transportation facilities are poor or inexistent can complicate the mobile sensing approach. To address this problem, in this paper we propose endowing Unmanned Aerial Vehicles (UAVs) with pollution sensors, allowing them to become autonomous air monitoring stations. The proposed solution has the potential to quickly cover a target region at a low cost, and providing great flexibility.

Mobile ad-hoc networks (MANETs) assume that mobile nodes voluntary cooperate in order to work properly. This cooperation is a cost-intensive activity and some nodes can refuse to cooperate, leading to a selfish node behaviour. Thus, the overall network performance could be seriously affected. The use of watchdogs is a well-known mechanism to detect selfish nodes. However, the detection process performed by watchdogs can fail, generating false positives and false negatives that can induce to wrong operations. Moreover, relying on local watchdogs alone can lead to poor performance when detecting selfish nodes, in term of precision and speed. This is specially important on networks with sporadic contacts, such as delay tolerant networks (DTNs), where sometimes watchdogs lack of enough time or information to detect the selfish nodes. Thus, we propose collaborative contact-based watchdog (CoCoWa) as a collaborative approach based on the diffusion of local selfish nodes awareness when a contact occurs, so that information about selfish nodes is quickly propagated. As shown in the paper, this collaborative approach reduces the time and increases the precision when detecting selfish nodes.

A representative characterisation of inter-contact times between nodes is essential for the performance evaluation of mobile networks. The most common characterization of inter-contact times is based on the study of the aggregate distribution of contacts between individual pairs of nodes. The problem with this aggregate distribution is that it is not always representative of the individual pair distributions, especially in the short term and when the number of nodes in the network is high. Thus, deriving results from this characterisation, can lead to inaccurate performance evaluations results.

The Delay Tolerant Networks (DTN) approach is considered the best strategy to address the specific issues of the VANETs, namely high mobility, variable node density or frequent radio obstacles. Several protocols have been proposed for DTNs, being the epidemic routing (and variations of it) the most representative protocol. Nevertheless, the availability of navigation systems, thanks to which each vehicle is aware of its location within a map, introduces the possibility for a new routing approach, known as Geographic Routing. In this paper we analytically evaluate the performance of our previously presented Map-based Sensor-data Delivery Protocol (MSDP). We introduce an analytical model that takes into account the effect of constrained buffers. The results show that adopting the Map-based Sensor-data Delivery Protocol (MSDP) routing mechanism allows achieving a reasonable delivery time with an insignificant overhead compared with epidemic routing.

Mobile ad-hoc Networks (MANETs) and Delay Tolerant Networks (DTN) rely on network cooperation schemes to work properly. Nevertheless, if nodes have a selfish behaviour and are unwilling to cooperate, the overall network performance could be seriously affected. The use of watchdogs is a well-known mechanism to detect selfish nodes. Nevertheless, the detection process performed by watchdogs can fail, generating false positives and false negatives that can induce a wrong behaviour.

Real-time systems used in media processing and transmission produce bursty workloads with highly variable execution and transmission times. To avoid the drawbacks of using the worst-case approach with these workloads, this paper uses a variation of the usual real-time task model where the WCET is replaced by a discrete statistical distribution. Using this approach, tasks are characterized by their processing time over a sampling period. We could expect that increasing the sampling period would smooth, in principle, the workload variability and the proposed analysis would provide more deterministic long- term results. However, we have surprisingly observed that this variability does not decreases with the sampling period: work- loads are bursty on many time scales. This property is known as self-similarity and is measured using the Hurst parameter.This paper studies how to properly model and analyze self- similar task sets showing the influence of the Hurst parameter on the schedulability analysis. It shows, through an analytical model and simulations, that this parameter may have a very negative impact on system performance. As a conclusion, it can be stated that this factor should be taken into account for statistical analysis of real-time systems, since simplistic workload models can lead to inaccurate results. It also shows that the negative effect of this parameter can be bounded using scheduling policies based on the bandwidth isolation principle.

Switched Ethernet is being used for real time transmissions in industrial automation more and and more. Most modern industrial switches are equipped with mechanisms to deal with time predictability. However, real-time transmission not only requires these mechanisms, but also the proper policies for managing network resources. This paper proposes the use of contracts. A contract is a set of transmission specifications which are negotiated between the applications and the run-time support. They define the application workload and the required performance guarantees. We implement contracts for real-time streaming as an extension of FRESCOR (framework for real-time embedded systems based on contracts). This framework was initially thought for providing deterministic performance guarantees to strictly periodic workloads. This work extends it by using the concept of classes of service (CoS) to deal with a wider range of workloads and guarantees and, particularly, with the transmission of highly variable bit rate (VBR) streams, like video. CoS enables, for example, joint transmission of real-time periodic workloads and VBR streams. CoS are implemented using a combination of resource reservation and resource preallocation techniques. The packet scheduling facilities of managed switches and Linux are shown to be key for managing network resources. Evaluations about the effectiveness of the extended FRESCOR framework and the feasibility of using switched Ethernet in real-time industrial environments are also presented.

Switched Ethernet networks are spreading to industrial environments more are more. The current trend is using them at all levels of a factory, replacing this way field-buses and other industrial networks. Switched Ethernet lacks the drawback of the non-deterministic collision resolution of coax cabling. However there are still some sources of indeterminism, mostly due to contention problems in message queues at switches and network interfaces. These problems can be dealt with using traffic control mechanisms for packet prioritisation and scheduling. These features have been largely inaccessible in Ethernet for a long time but, nowadays, they are widely available in industrial switches and also in the Linux TC (Traffic Control) mechanism. This paper presents an evaluation of traffic control mechanisms in a scenario where real-time periodic messages are transmitted along with some other bandwidth-consuming workload like video streams. The evaluation focuses on the effect of two traffic control mechanisms: rate limiting and packet prioritisation. Results show that these mechanisms provide switched Ethernet with a reasonably good predictability, even with high bandwidth utilisations. Rate limiting mostly reduces the average delay while packet prioritisation reduces the jitter and the dispersion.

Network performance analysis relies mainly on two models: a workload model and a performance model. This paper proposes to use histograms for characterising the arrival workloads and a performance model based on a stochastic process. This new stochastic process works directly with histograms using a set of specific operators. The result is the buffer occupancy distribution. The loss rate and network delay distribution can be obtaining using this distribution. Three traffic models are proposed: the first model (the HD model) is a basic histogram model that is compact and short-range dependent, the second (HD(m) model) is based on obtaining several histograms using different time scales and the third (the HD(H) model) is based on the Hurst parameter and it is long-range dependent.

Expedited forwarding (EF) is the differentiated services class of service that provides high quality transmission with node bounded delay. Nevertheless, in order to obtain a bounded network delay it is necessary to compute a route that meets the required end-to-end delay. Therefore, we study the requirements (bandwidth, buffer and delay) for a new EF connection. As detailed in the paper, with these constraints the routing algorithm is a NP-complete problem. Therefore, we present an efficient routing scheme that has low polynomial computational cost. Finally, the evaluation of this routing shows that is as efficient as using the exact routing algorithm.

Real-time transmission implies guaranteeing a given quality of service (QoS), requiring large use of network resources. Backup channels introduce the notion of availability to real-time transmission at the cost of increasing the use of network resources. However, this over-provisioning of resources is potentially wasted, since fault rate is very low. This paper introduces a new failure detection scheme for real-time transmission called proactive backup channel. This scheme is based on activating the backup channel before a fail is produced. As proven in the paper, this scheme reduces the use of network resources and is suitable for integrated and differentiated services.