Network slicing enables the coexistence of multiple virtual networks over shared infrastructure, each designed to meet distinct service-level requirements. Ensuring SLA compliance in such environments demands efficient and fine-grained monitoring, particularly within the transport network (TN). Existing In-band Network Telemetry (INT) solutions, however, face challenges related to scalability and overhead. We propose SliceMon, a slice-aware SLA monitoring framework for P4enabled TNs. SliceMon deploys lightweight machine learning models in the data plane to locally infer potential SLA violations, while the controller aggregates these inferences using ensemble methods and activates detailed telemetry collection only when necessary. We evaluate SliceMon under delay-based SLAs using datasets gathered via emulation with Mininet and BMv2. Results demonstrate that SliceMon maintains high detection accuracy while significantly reducing monitoring overhead and improving responsiveness compared to traditional periodic telemetry approaches.

Social Virtual Reality (VR) applications enable real-time interaction among users in shared immersive environments, but pose stringent requirements on latency, bandwidth, and computational resources. In this poster, we consider the placement of rendering Virtual Network Functions (VNFs) on edge servers in a 5G backhaul network, with the goal of minimizing energy consumption while satisfying strict Motion-to-Photon (MTP) latency constraints. We formulate the problem as a non-linear programming model that accounts for multicast transmission of avatars, path delays, and computing constraints. Preliminary results show that our approach outperforms a baseline strategy, achieving lower energy usage while satisfying latency limits.

5G and beyond support the deployment of vertical applications, which is particularly appealing in combination with network slicing and edge computing to create a logically isolated environment for executing customer services. Even if serverless computing has gained significant interest as a cloud-native technology its adoption at the edge is lagging, especially because of the need to support stateful tasks, which are commonplace in, e.g., cognitive services, but not fully amenable to being deployed on limited and decentralized computing infrastructures. In this work, we study the emerging paradigm of stateful Function as a Service (FaaS) with lightweight task abstractions in WebAssembly. Specifically, we assess the implications of deploying inter-dependent tasks with an internal state on edge computing resources using a stateless vs. stateful approach and then derive a mathematical model to estimate the energy consumption of a workload with given characteristics, considering the power used for both processing and communication. The model is used in extensive simulations to determine the impact of key factors and assess the energy trade-offs of stateless vs. stateful.

A science is defined by a set of encyclopedic knowledge related to facts or phenomena following rules or evidenced by experimentally-driven observations. Computer Science and in particular computer networks is a relatively new scientific domain maturing over years and adopting the best practices inherited from more fundamental disciplines. The design of past, present and future networking components and architectures have been assisted, among other methods, by experimentally-driven research and in particular by the deployment of test platforms, usually named as testbeds . However, often experimentally-driven networking research used scattered methodologies, based on ad-hoc, small-sized testbeds , producing hardly repeatable results. We believe that computer networks needs to adopt a more structured methodology, supported by appropriate instruments, to produce credible experimental results supporting radical and incremental innovations. This paper reports lessons learned from the design and operation of test platforms for the scientific community dealing with digital infrastructures. We introduce the SLICES initiative as the outcome of several years of evolution of the concept of a networking test platform transformed into a scientific instrument. We address the challenges, requirements and opportunities that our community is facing to manage the full research-life cycle necessary to support a scientific methodology.

One of the main operational challenges faced by the operators of one-way car-sharing systems is to ensure vehicle availability across the regions of the service areas with uneven patterns of rental requests. Fleet balancing strategies are required to maximise the demand served while minimising the relocation costs. However, the design of optimal relocation policies is a complex problem, and global optimisation solutions are often limited to very small network sizes for computational reasons. In this work, we propose a multi-stage decision support system for vehicle relocation that decomposes the general relocation problem into three independent decision stages to allow scalable solutions. Furthermore, we adopt a rolling horizon control strategy to cope with demand uncertainty. Our approach is highly modular and flexible, and we leverage it to design user-based, operator-based and robotic relocation schemes. Besides, we formulate the relocation problem considering both conventional cars and a new class of compact stackable vehicles that can be driven in a road train. We compare the proposed relocation schemes with two recognised benchmarks using a large data set of taxi trips in New York. Our results show that our approach is scalable and outperforms the benchmark schemes in terms of quality of service, vehicle utilisation and relocation efficiency. Furthermore, we find that stackable vehicles can achieve a relocation performance close to that of autonomous cars, even with a small workforce of relocators.

Multi-access edge computing (MEC) is emerging as an effective solution to fulfil the requirements of time-critical Internet of Things (IoT) applications. However, to improve the network efficiency and QoS support of MEC systems, it is essential to jointly optimise edge resource management, IoT data collection and IoT device association in the context of heterogeneous services and edge resources. In this study, we address these challenges by formulating the resource allocation, device association and data routing problem in a multi-cell MEC network as a mixed-integer non-linear programming problem. We also propose a best-fit greedy heuristic method to determine an approximate solution to the optimisation problem for online resource management. Simulation results confirm the effectiveness of the proposed algorithm compared to three alternative benchmarks.

In this study, we focus on the problem of managing a hybrid, shared IoT-based monitoring system, in which stationary sensor devices are complemented with user-carried personal devices embedded with sensing capabilities. The envisioned crowd-assisted monitoring system must support the sharing of the sensing infrastructure among multiple concurrent sensing tasks that can have highly varying QoS requirements. In such a scenario, a key issue is to maximise the utilisation efficiency of the physical sensing resources and the QoS satisfaction of sensing tasks while limiting the redundancy of collected data. As in previous research, we advocate the use of an IoT Broker, an intermediary entity that (i) interacts with the IoT applications to collect their QoS requirements (i.e., spatial coverage, data notification frequency); and (ii) coordinates with the redundant sensor deployments and mobile devices to selectively activate and configure the data streams that are needed to fulfil application requirements in a cost-efficient way. Then, we have developed an optimisation framework to jointly select the set of physical sensing resources to activate and the data update frequency for maximising the overall sensing performance while limiting redundant data. A key feature of our proposed framework is to be privacy-friendly as it only requires coarse-grained space-time knowledge of device location. Extensive simulations under realistic WSN deployments and real-life mobility patterns confirm the efficiency of the proposed solution in terms of data-coverage gain and reduction of data redundancy with respect to classical non-hybrid monitoring systems.

The Shared Sensor Network (SSN) model has recently emerged to reduce the high deployment and management costs of application-specific WSNs. In a SSN, the underlying physical infrastructure is shared among multiple applications simultaneously. Sensor tasks are likely to have different QoS requirements, e.g. in terms of sensing rate and coverage. In this scenario, we advocate the use of a local application broker (typically deployed on an edge device) to act as a mediator between the physical sensing resources and the sensing tasks, and to support efficient resource allocation and controlled sharing of cached data between applications. To this end, we have formulated an optimisation problem to determine: (i) the set of sensing resources to use; (ii) a mapping between activated sensor resources and admitted applications; and (iii) the probing rate of the broker for the activated sensors. The objective of our model is to maximise both the number of admitted applications (i.e. the revenue for the provider of the sensing infrastructure) and the system lifetime. We have implemented a prototype of the proposed broker using native CoAP functionalities, and we have conducted an extensive evaluation in an emulation environment. Results showed that our application broker provides better performance in terms of the number of admitted applications and energy efficiency than a state-of-the-art benchmark.

Data management using Device-to-Device (D2D) communications and opportunistic networks (ONs) is one of the main focuses of human-centric pervasive Internet services. In the recently proposed "Internet of People" paradigm, accessing relevant data dynamically generated in the environment nearby is one of the key services. Moreover, personal mobile devices become proxies of their human users while exchanging data in the cyber world and, thus, largely use ONs and D2D communications for exchanging data directly. Recently, researchers have successfully demonstrated the viability of embedding human cognitive schemes in data dissemination algorithms for ONs. In this paper, we consider one such scheme based on the recognition heuristic, a human decision-making scheme used to efficiently assess the relevance of data. While initial evidence about its effectiveness is available, the evaluation of its behaviour in large-scale settings is still unsatisfactory. To overcome these limitations, we have developed a novel hybrid modelling methodology, which combines an analytical model of data dissemination within small-scale communities of mobile users, with detailed simulations of interactions between different communities. This methodology allows us to evaluate the algorithm in large-scale city- and country-wide scenarios. Results confirm the effectiveness of cognitive data dissemination schemes, even when content popularity is very heterogenous.

CoAP is an application protocol that provides standardised RESTful services for IoT devices. Since COAP messages are encapsulated into UDP datagrams, COAP specification provides: i) optional reliability mechanisms through retransmissions, and ii) simple congestion control mechanisms based on retransmission timeouts. Recent studies have demonstrated that these congestion control schemes may significantly underperform when operating with bursty traffic. To address these limitations, in this paper we propose COAP-R, an alternative solution for regulating the sending rate of CoAP sources, which adopts a rate-based approach for traffic control. Key features of COPA-R are: i) to leverage the tree-based routing structure of IoT networks to estimate the maximum throughput that can be obtained on the bottleneck link of every upward route, and ii) to perform in a distributed manner a max-min fair allocation of available network capacity on the basis of estimated bottleneck bandwidths. The proposed approach is evaluated by means of simulations considering a scenario in which traffic is generated in bursts, for instance as consequence of events detected by sensors. Our simulations demonstrate that the proposed approach ensures a fair allocation of network resources, and leads to a 40% decrease of the data collection delays when compared to COAP.

Presents the introductory welcome message from the conference proceedings. May include the conference officers' congratulations to all involved with the conference event and publication of the proceedings record.

CoAP is a web transfer protocol that provides basic RESTful services for IoT devices. Since CoAP operates on top of UDP, CoAP must support additional congestion control and reliability mechanisms. However, CoAP is originally designed to allow implementations that do not maintain any end-to-end connection information. Thus, the default congestion control mechanisms are not capable of adapting to network conditions. A few enhancements of the basic congestion control mechanism of CoAP have been recently proposed, and CoCoA+ is commonly considered the most mature of these proposals. However, the shortcomings of CoCoA+ have not been sufficiently investigated in the literature. In this paper, we present an in-depth analysis of the congestion control mechanisms of CoCoA+ and CoAP for a variety of traffic patterns under realistic IoT setups. The results of this evaluation indicate that CoCoA+ can perform significantly worse than default CoAP, especially with bursty traffic and in networks with few clients as a result of an improper selection of the retransmission timeouts. Furthermore, large variability of RTO values in CoCoA+ makes difficult to meet end-to-end delay requirements of typical IoT monitoring applications.

The number and capabilities of IoT devices will exponentially grow over the next years. Together with the pervasive diffusion of smart personal mobile devices this opens up unprecedented opportunities for contextualised services provided to mobile users, based on their current interests and behaviours. In addition, most of these services will be content-centric rather than host-centric. Cloud computing and Information-Centric Networking (ICN) are therefore two key technologies in this perspective. In both cases, solutions are typically designed for global Internet platforms, while mobile nodes are seen as edge devices from which data are fetched and sent back through pervasive wireless networks (typically, LTE). However, it is questionable whether such an approach will work as expected, e.g., due to data privacy concerns and expected bandwidth shortage of even last-generation cellular networks. In this paper we present a general framework where global cloud and ICN platforms are complemented in a totally synergic way by local clouds formed at the edge of the network by mobile devices, where service provisioning and data management functionalities are offloaded whenever possible (and appropriate). This results in a multi-layer, content- and service-centric approach to IoT data management and service provisioning. We then present performance evaluation results from applying this framework to a specific case where data-centric services are jointly provided by edge devices and by a global cloud platform. Results show that this approach is very promising, as it is able to drastically cut the related cellular-network traffic, and, at the same time, improve the effectiveness of service provisioning to users.

It is our pleasure to introduce the technical program of WoWMoM 2016.

LTE is rapidly gaining momentum for building future 4G cellular systems, and real operational networks are under deployment worldwide. To achieve high throughput performance, in addition to an advanced physical layer design LTE exploits a combination of sophisticated mechanisms at the radio resource management layer. Clearly, this makes difficult to develop analytical tools to accurately assess and optimise the user perceived throughput under realistic channel assumptions. Thus, most existing studies focus only on link-layer throughput or consider individual mechanisms in isolation. The main contribution of this paper is a unified modelling framework of the MAC-level downlink throughput of a sigle LTE cell, which caters for wideband CQI feedback schemes, AMC and HARQ protocols as defined in the LTE standard. We have validated the accuracy of the proposed model through detailed LTE simulations carried out with the ns-3 simulator extended with the LENA module for LTE.

Offloading is gaining momentum as a technique to overcome the cellular capacity crunch due to the surge of mobile data traffic demand. Multiple offloading techniques are currently under investigation, from modifications inside the cellular network architecture, to integration of multiple wireless broadband infrastructures, to exploiting direct communications between mobile devices. In this paper we focus on the latter type of offloading, and specifically on offloading through opportunistic networks. As opposed to most of the literature looking at this type of offloading, in this paper we consider the case where requests for content are non-synchronised, i.e. users request content at random points in time. We support this scenario through a very simple offloading scheme, whereby no epidemic dissemination occurs in the opportunistic network. Thus our scheme is minimally invasive for users' mobile devices, as it uses only minimally their resources. Then, we provide an analysis on the efficiency of our offloading mechanism (in terms of percentage of offloaded traffic) in representative vehicular settings, where content needs to be delivered to (subsets of the) users in specific geographical areas. Depending on various parameters, we show that a simple and resource-savvy offloading scheme can nevertheless offload a very large fraction of the traffic (up to more than 90%, and always more than 20%). We also highlight configurations where such a technique is less effective, and therefore a more aggressive use of mobile nodes resources would be needed.

Adaptive Modulation and Coding (AMC) in LTE networks is commonly employed to improve system throughput by ensuring more reliable transmissions. Most of existing AMC methods select the modulation and coding scheme (MCS) using pre-computed mappings between MCS indexes and channel quality indicator (CQI) feedbacks that are periodically sent by the receivers. However, the effectiveness of this approach heavily depends on the assumed channel model. In addition CQI feedback delays may cause throughput losses. In this paper we design a new AMC scheme that exploits a reinforcement learning algorithm to adjust at run-time the MCS selection rules based on the knowledge of the effect of previous AMC decisions. The salient features of our proposed solution are: i) the low-dimensional space that the learner has to explore, and ii) the use of direct link throughput measurements to guide the decision process. Simulation results obtained using ns3 demonstrate the robustness of our AMC scheme that is capable of discovering the best MCS even if the CQI feedback provides a poor prediction of the channel performance.

Lightweight link quality estimation is crucial in wireless sensor networks. Indeed, devices with limited capabilities shall trade off between consuming their resources to maintain a precise view of the neighbours' link quality and to build routes almost blindly. For instance, the Routing Protocol for Low-Power and Lossy Networks (RPL), which has been recently standardised by the IETF to enable IPv6-based sensor networks, only estimates the quality of the links used to deliver data packets. However, this solution has been demonstrated to cause periods of routing instability and reduced packet delivery rates since it estimates only the quality of utilised links. To address this issue in this work we propose a lightweight link estimation procedure that exploits Trickle-based topology maintenance techniques to simultaneously estimate link qualities and propagate routing information. Our proposed scheme has been integrated in the Contiki's RPL prototype implementation. Simulation results demonstrate that our proposal is capable of measuring the quality of the links to neighbours with small overhead, which results into better routing decisions and improved packet delivery rates.

The IPv6 routing protocol for low-power and lossy networks (RPL) has been recently standardized by the Internet Engineering Task Force (IETF) routing protocol for low-power and lossy networks (ROLL) working group to support IPv6 routing for resource-constrained devices in industrial, home, and urban environments. However, several studies have shown that RPL may experience (very) low delivery rates, particularly in large-scale deployments. In this paper, we provide an in-depth analysis of the protocol attributes and design choices that generate such unreliability issues. Then, we describe and evaluate a new implementation of the RPL standard for the Contiki operating system (OS) to improve data delivery reliability. The salient feature of our RPL implementation is to adopt a flexible cross-layering design that provides simple routing optimizations, enhanced link estimation capabilities, and efficient management of neighbor tables. In order to verify the effectiveness of our RPL implementation, we use an advanced metering infrastructure (AMI) as a case study. Results obtained using Cooja emulator in two sets of experiments, differentiated by the presence or lack of duty cycling, indicate that our RPL implementation outperforms the one provided in Contiki in terms of average packet delivery rates by up to 200% in networks with 100 nodes.

Low-Power and Lossy Networks (LLNs) are an enabling technology for many Internet of Things (IoT) applications. To allow LLNs to interoperate with the Internet the IETF is specifying IP-compatible standards that are specifically designed for embedded devices with small memory and limited computing capabilities. The purpose of this paper is to investigate the interplay between the recently proposed IETF standard for IPv6 routing in LLNs, called RPL, with state-of-the-art autoconfiguration algorithms that assign network addresses in a hierarchical manner. To this end, we have extended the default RPL implementation available in Contiki operating system to support T-DAAP, LISAA and DAAM addressing schemes. Our experimental results clearly show that the mechanisms used by RPL to configure and maintain a tree-based network topology frequently break address hierarchy, limiting the opportunities for route aggregation. Furthermore, classical techniques used to reduce network instability significantly increase the network set-up times and, in same cases, degrade path qualities.

In this paper we consider a vehicular sensor network in which vehicles are equipped with video cameras and continuously capture images from urban roads. Then, vehicles can use roadside wireless access points (APs) encountered during travel to deliver recorded image data to remote data collectors, where images streams from multiple sources are aggregated and processed. However, how to efficiently utilize the limited upload capacity of the wireless access network while reducing data redundancy due to spatial correlation of neighboring vehicles is a critical issue. To tackle this problem we propose a mechanism to dynamically adjust sampling rates of onboard cameras based on the vehicle status and the spatial distribution of roadside APs. The key idea is that vehicles traveling close to a roadside AP should use lower sampling rates than vehicles traveling in areas with a poor connectivity. We use network simulations with realistic vehicular mobility patterns to verify the performance gains of our proposed scheme compared to a baseline system that use fixed sampling rates. Simulation results show that our solution can ensure a more balanced and uniform coverage of the road network while reducing the amount of transferred data.

In vehicular multimedia sensor networks vehicles are equipped with cameras and they continuously capture images from urban streets. Then, vehicles can use roadside wireless access points encountered during travel to deliver recorded image data to remote data collectors, in which the information from several multimedia streams is aggregated and processed to enable new services, such as urban surveillance, or traffic and road monitoring. However, due to constraints on the wireless access network the amount of image data that can be transferred from vehicles is limited, and data redundancy should be avoided. In this paper we address this issue by using submodular optimization techniques to develop an efficient data collection algorithm capable of providing data redundancy elimination under network capacity constraints. We also design an alternative decentralized scheme that operates on longer time scales and relies only on basic aggregate information. We use network simulations with realistic vehicular mobility patterns to verify the performance gains of our proposed schemes compared to a baseline system that ignores data redundancy. Simulation results show that our data collection techniques can ensure a more accurate coverage of the road network while significantly reducing the amount of transferred data.

The purpose of this survey is to present a critical overview of smart grid concepts, with a special focus on the role that communication, networking and middleware technologies will have in the transformation of existing electric power systems into smart grids. First of all we elaborate on the key technological, economical and societal drivers for the development of smart grids. By adopting a data-centric perspective we present a conceptual model of communication systems for smart grids, and we identify functional components, technologies, network topologies and communication services that are needed to support smart grid communications . Then, we introduce the fundamental research challenges in this field including communication reliability and timeliness, QoS support, data management services, and autonomic behaviors. Finally, we discuss the main solutions proposed in the literature for each of them, and we identify possible future research directions.

When faced with large amounts of data, human brains are able to swiftly react to stimuli and assert relevance of discovered information, even under uncertainty and partial knowledge. These efficient decision-making abilities rely on so-called cognitive heuristics, which are rapid, adaptive, light-weight yet very effective schemes used by the brain to solve complex problems. In a content-centric future Internet where users generate and disseminate large amounts of content through opportunistic networking techniques, individual nodes should exhibit those properties to support a scalable content dissemination system. We therefore study whether such cognitive heuristics can also be used in such a networking environment. To this end, in this paper we develop an analytical model that describes a content dissemination mechanism for opportunistic networks based on one such heuristics, known as the recognition heuristic. Our model takes into account the different popularities of content types, and highlights the impact of the shared memory contributed by individual nodes to make the dissemination process more efficient. Furthermore, our model allows us to investigate the performance of the dissemination process for very large number of nodes, which might be very difficult to carry out through a simulation-based study.

WoWMoM has been known for its successful workshop programs as well as its high-quality main conference papers. WoWMoM has identified emerging hot workshop topics which also continued to last and grow in the field. The timely selections of new topics have yielded several multi-year workshops proudly hosted by this year WoWMoM. Some previous WoWMoM workshops have grown to standalone conferences or continue to exist in other major mobile & wireless networking venues.

In this paper we design, implement and evaluate a practical flow-based routing architecture for WMNs, called TALB, which exploits traffic measurements to balance traffic load over available mesh paths and gateways. To achieve throughput enhancements while mitigating inter-flow interference, TALB implements a Constrained Maximal Bottleneck Path computation algorithm to maximize the bottleneck capacity of the paths used to reach the Internet, while preserving a given degree of traffic locality. The distinctive aspect of our proposed scheme is that it relies on a traffic-aware definition of residual link capacities to properly take into account the heterogeneity of Internet applications (e.g., inelastic vs. elastic traffic) and their service requirements (e.g., fixed vs. adaptive demands). Then, each mesh node monitors the main features of long-lived traffic flows, and reports such information to a network manager, which decides upon the sub-optimal paths to allocate to each new session. Tests performed in a 10-node wireless mesh testbed show that TALB provides remarkably throughput gains over nearest gateway strategies in case of mixed TCP and UDP traffic scenarios, or low to moderate load conditions.

The performance of vehicular Internet access using roadside 802.11-based APs has been extensively studied in the literature. However, much less attention has been dedicated to analysing the maximum achievable throughput when multiple vehicles simultaneously share the bandwidth of the same roadside AP in a given mobility scenario. To fill such a gap, in this paper we develop an analytical framework to quantify the total amount of data transferred by a vehicle that drives through the coverage area of a roadside 802.11-based AP. The distinctive aspects of our analysis are the following: (i) it considers heterogeneous vehicular environments where vehicles may have different mobility characteristics; and (ii) it accurately takes into account critical traffic parameters, such as road capacity and vehicle density, when modelling MAC layer capacity. Our model is able to accurately characterize the unfairness that may arise due to differences in the relative speed of vehicles. In addition, our analysis and the supporting simulation results are useful to explain the complex relation ship that exists between the upload capacity per vehicle, the macroscopic characteristics of the traffic stream, the vehicular mobility model, and the 802.11 channel access rules.

Classical routing protocols for WMNs are typically designed to achieve specific target objectives (e.g., maximum throughput), and they offer very limited flexibility. As a consequence, more intelligent and adaptive mesh networking solutions are needed to obtain high performance in diverse network conditions. To this end, we propose a reinforcement learning-based routing framework that allows each mesh device to dynamically select at run time a routing protocol from a pre-defined set of routing options, which provides the best performance. The most salient advantages of our solution are: i) it can maximize routing performance considering different optimization goals, ii) it relies on a compact representation of the network state and it does not need any model of its evolution, and iii) it efficiently applies Q-learning methods to guarantee convergence of the routing decision process. Through extensive ns-2 simulations we show the superior performance of the proposed routing approach in comparison with two alternative routing schemes.

Most studies on WMNs have assumed that all network traffic from clients goes through a gateway node. However, point-to-point traffic is expected to significantly contribute to WMN network load in many scenarios. In this paper we propose a novel routing strategy that seeks to alleviate network congestion by avoiding paths that are already heavily loaded. Our scheme effectively takes into account the elastic nature of TCP traffic when calculating the paths that balance network load. Furthermore, to mitigate the flow unfairness that affects TCP-like traffic in multi-hop environments, we avoid that single-hop flows fully utilize link resources, starving other multi-hop flows sharing the same links. We have implemented and tested the proposed routing protocol on a 10-node wireless mesh testbed. Our experiments show that the proposed scheme outperforms OLSR both in terms of average throughput and fairness.

It gives us great pleasure to introduce the technical program of the third IEEE Workshop on Hot Topics in Mesh Networking (HotMESH 2011). Following the success of the previous editions, the HotMESH workshop is now a well-established, high profile forum for researchers, engineers, and practitioners to present state-of-the-art research in the field of wireless mesh networking. Although significant advances have taken place in this field in the last few years, fundamental open questions and technical challenges still remain to build highly scalable, reliable and cost-effective wireless backbones for mobile users. One of the issues of particular interest to this workshop edition is how wireless mesh networks could be integrates with and/or influence existing mobile and fixed broadband networks in order to create an integrated infrastructure that supports emerging services. Then, the main purpose of this workshop is to promote discussions on recent advances in the analysis design and implementation of systems, protocols and services for next generation mobile mesh networks. It also aims at increasing the synergy between academic and industry professionals working in this area.

Opportunistic Routing is emerging as a promising paradigm to mitigate performance degradation in wireless multi-hop networks due to lossy links and varying channel conditions. Opportunistic routing protocols exploit the broadcast nature of the wireless medium to perform hop-by-hop route construction, and to take advantage of path diversity. However, most of the existing solutions impose a-priori constraints on the set of candidate forwarders that can be used when routing a packet. In this paper, we describe MaxOPP, a flexible and adaptive opportunistic routing algorithm able to select at each hop, and at run-time, the candidate forwarders that can maximize the opportunistic throughput gain. Thus, forwarding decisions in MaxOPP are dynamically adapted to variations of network conditions, ensuring an efficient trade-off between reliability and opportunistic benefit. Simulations conducted with NS-2 on a set of representative scenarios show that MaxOPP achieves higher throughput for bulk data transfers than traditional shortest-path routing.

Wireless mesh networks (WMNs) are comprised of nodes with multiple radio interfaces and provide broadband residential internet access or connectivity to temporal events. Our goal is to simplify the network deployment of such a mesh network, and towards that we are presenting procedures for automatic configuration and optimisation of the network. We first present an architecture framework that supports the integration of key mechanisms to ensure the optimisation of the performance of a wireless mesh network. Secondly, we present three key mechanisms, namely auto-configuration, channel assignment and quality of service (QoS) enforcement based on QoS routing. We provide a method for automatic mesh start-up, joining a node into an existing mesh network and automatic repair of temporary connectivity outage, targeting at simplifying the node configuration as much as possible. The second mechanism supports an efficient algorithm for joint channel selection and topology control, supporting different target objective expressed as utility functions. The third mechanism supports QoS, by allowing routing and admission control decisions, in order to ensure that all flows are handled with the demanded QoS. Finally, we give some simulation results that show the increased performance of our framework.

Traffic routing and gateway selection for Internet flows play a crucial role in determining the performance of WMNs. Optimal routing strategies have been derived under the assumption that the traffic demands are static and known a priori, or that can be accurately predicted. However, the effectiveness of most of the existing optimization-based routing solutions is still to be demonstrated in real-world deployments. In this paper we present a practical and robust framework for load-balanced routing and gateway selection in WMNs under variable traffic loads, which relies only on real-time measurements of network statistics. Then, we design, implement and evaluate two simple adaptive strategies to automatically select network paths and gateways for flows that dynamically arrive and depart, with the goal of achieving a high network utilization and improved load balancing. We have developed a full-fledged implementation of the proposed mechanisms as an extension of the OLSR protocol, and conducted an experimental study in a trial outdoor mesh network using video streaming traffic. Collected measurements show that our scheme can significantly increase the number of video sessions that can be supported with acceptable quality compared to traditional shortest path routing.

In this paper we address the issue of designing a routing scheme for ad hoc networks, which aims at maximizing the number of flows with satisfied bandwidth demands. In general, selecting a multi-hop wireless path for bandwidth-guaranteed flows is challenging because of the broadcast nature of the wireless medium. Thus, in this study we formulate the route selection problem by considering a synchronized multi-rate TDMA access scheme. Once the route is found, the radio resources are reserved as time slots in the TDMA frames along the path the flow takes. We demonstrate that selecting routes so as to maximize the number of accepted flows is an APX-complete problem (i.e., there are no polynomial-time approximation schemes), even under simplified rules for bandwidth reservation. This result is stronger than previously established in other studies. Guided by our analysis, we propose a new cumulated available resources and Topology Aware (CARTA) routing heuristic, which selects routes bypassing heavily loaded and highly interfered network regions. Simulations performed with random topologies of up to 100 nodes and various traffic configurations show that CARTA obtains a more balanced utilization of the network resources, and up to 25% capacity increase than the second best metric from the literature we tested, at the cost of a limited increase in path lengths.

In wireless mesh networks (WMNs), gateway nodes may become a severe bottleneck for Internet flows. Indeed, if traffic is routed in the mesh without considering traffic distribution, as well as link capacities, some gateways or intermediate mesh routers may rapidly get overloaded due to unevenly utilization of network resources. To address this issue, in this paper we firstly develop a multi-class queuing network model to analyze feasible throughput allocations in heterogeneous WMNs, as well as to predict the residual capacity of network paths. Guided by our analysis, we design a Capacity-Aware Route Selection algorithm (CARS), which allocates network paths to downstream and upstream Internet flows so as to ensure a more balanced utilization of wireless network resources and gateways' Internet connections. Through simulations in a number of different network scenarios we show that CARS significantly outperforms shortest path routing using routing metrics that capture only inter-flow interference.

Wireless mesh networks are primarily used to provide Internet access by sharing the Internet connections of a limited number of gateways. If traffic is routed in the mesh without considering load distribution, unevenly network congestion may arise and some gateways may rapidly get overloaded, which causes a significant reduction of the network capacity. To address this issue, in this paper we firstly develop a queuing network model that accurately predicts the residual capacity of paths in heterogeneous mesh networks, and precisely identify network bottlenecks. By taking advantage of this model, we design a novel Load-Aware Route Selection algorithm, named LARS, which improves the network capacity. This objective is obtained by allowing each mesh node to distribute the traffic load among multiple gateways in order to ensure evenly utilization of Internet connections. Simulation results show that LARS significantly outperforms shortest path routing using contentionaware link costs, achieving throughput improvements of up to 210% in the considered network scenarios.

We present our initial work on the design and of an opportunistic forwarding scheme for a wireless network. The main focus in our work is to take advantage of the interplay of the lower layer parameters and thus decide the design aspects of the forwarding decisions for an adaptive probabilistic opportunistic forwarding scheme. We have already shown that our opportunistic scheme can outperform single path routing for moderate channel error values, we have also identified a tradeoff between differentiating the forwarder's back-off value to avoid collisions and reducing delay.

There is a vast literature on the throughput analysis of the IEEE 802.11 media access control (MAC) protocol. However, very little has been done on investigating the interplay between the collision avoidance mechanisms of the 802.11 MAC protocol and the dynamics of upper layer transport protocols. In this paper, we tackle this issue from an analytical, simulative, and experimental perspective. Specifically, we develop Markov chain models to compute the distribution of the number of active stations in an 802.11 wireless local area network (WLAN) when long-lived transmission control protocol (TCP) connections compete with finite-load user datagram protocol (UDP) flows. By embedding these distributions in the MAC protocol modeling, we derive approximate but accurate expressions of the TCP and UDP throughput. We validate the model accuracy through performance tests carried out in a real WLAN for a wide range of configurations. Our analytical model and the supporting experimental outcomes show that 1) the total TCP throughput is basically independent of the number of open TCP connections and the aggregate TCP traffic can be equivalently modeled as two saturated flows; and 2) in the saturated regime, n UDP flows obtain about n times the aggregate throughput achieved by the TCP flows, which is independent of the overall number of persistent TCP connections.

It is widely recognized that a prerequisite for the commercial penetration of the ad hoc networking technologies is the integration with existing wired/wireless infrastructure-based networks to provide an easy and transparent access to the Internet and its services. However, most of the existing solutions for enabling the interconnection between IPv4-based mobile ad hoc networks and the Internet are based on complex and inefficient mechanisms, as Mobile-IP and IP tunnelling. In this paper, we describe an alternative approach to build multi-hop and heterogeneous proactive ad hoc networks, which can be used as flexible and low-cost extensions of traditional wired local area networks (LANs). Our proposed architecture provides transparent global Internet connectivity and address autoconfiguration capabilities to mobile nodes without requiring configuration changes in the pre-existing wired LAN, and relying on basic layer-2 functionalities. We have prototyped the core components of this architecture for Optimized Link State Routing-based ad hoc networks and we have conducted several experiments comparing the throughput performance obtained using our scheme and a well-known alternative (network address translation) NAT-based solution. The experimental outcomes show that our proposed technique ensures higher perconnection throughputs than the NAT-based solution in the considered network scenarios.

The growing popularity of the 802.11 standard for building local wireless networks has generated an extensive literature on the performance modelling of its MAC protocol. However, most of the available studies focus on the throughput analysis in saturation conditions, while very little has been done on investigating the interactions between the 802.11 MAC protocol and closed-loop transport protocols such as TCP. This paper addresses this issue by developing an analytical model to compute the stationary probability distribution of the number of backlogged nodes in a WLAN in the presence of persistent TCP-controlled download and upload data transfers, and embedding the network backlog distribution in the MAC protocol modelling. A large set of experiments conducted in a real network validates the model correctness for a wide range of configurations. A particular emphasis is devoted to investigate and explain the TCP fairness characteristics. Our analytical model and the supporting experimental outcomes demonstrate that using default settings for the capacity of devices' output queues provides a fair allocation of channel bandwidth to the TCP connections, independently of the number of downstream and upstream flows. Furthermore, we show that the TCP total throughput does not degrade by increasing the number of wireless stations.

This paper proposes a distributed and topology-independent contention control mechanism for IEEE 802.11-based multi-hop ad hoc networks, which aims at ensuring per-station fairness without degrading the total network throughput, and without requiring any information exchange between the nodes. The proposed algorithm dynamically controls the frequency each station accesses the channel according to the station's channel share. The statistics of the stations' channel shares are updated locally in each node based only on the standard physical carrier sensing activity. To avoid throughput degradations while seeking fair allocation of bandwidth, we have also developed a credit-based scheme that operates in cooperation with the backoff procedure. The credits are a virtual currency earned by stations refraining from transmitting, and spent to reclaim backoff-free transmission opportunities. Simulations conducted in several well-known problematic topologies show that our scheme can achieve a better fairness than standard IEEE 802.11, while increasing the total network throughput

This paper describes the results of an ns2-based comparative performance evaluation among three major solutions presented in the literature for forming multi-hop networks of Bluetooth devices (scatternet formation). The three protocols considered in this paper are BlueTrees, BlueStars, and the "Yao protocol". We observed that device discovery is the most time-consuming operation, independently of the particular protocol to which it is applied. By means of a thorough performance evaluation we have identified protocol parameters and Bluetooth technology features that affect the duration of this device discovery. We have also analyzed the effect of the different protocols operations on key metrics of the generated scatternets. The comparative performance evaluation showed that due to the simplicity of its operations and to its basic working requirements BlueStars is by far the fastest protocol for scatternet formation which also yields to scatternets with a lower number of piconets, average route length and number of roles per node. However, BlueStars produces scatternets with an unbounded, possibly large number of slaves per piconet, which imposes the use of potentially inefficient Bluetooth operations. A good compromise when interested informing scatternets whose piconets have a bounded number of slaves is obtained by combining BlueStars and the Yao protocol. Although latency and route lengths are longer than in BlueStars scatternets, with the combined solution we obtain an overall good protocol performance and scatternets with desired characteristics.

We address the throughput analysis of high-speed IEEE 802.11b WLANs from both an analytical and a simulative perspective. Specifically, we derive the throughput formula for the RTS/CTS access method of the p-persistent IEEE 802.11b MAC protocol. The accuracy of the proposed model is exhaustively validated via simulation results. By exploiting our formulas, we derive the theoretical upper bound for the throughput performance of the IEEE 802.11b protocol. Our analytical and simulative results indicated that the RTS/CTS mechanism produces very limited advantages in the standard IEEE 802.11 networks with respect to the basic access when no hidden stations are present. Finally, we extend a distributed backoff-tuning strategy firstly proposed for the basic access method, and we validate its effectiveness to approach closely the throughput limit of the IEEE 802.11b protocol.

Wireless technologies in the LAN environment are becoming increasingly important. The IEEE 802.11 is the most mature technology for wireless local area networks (WLANs). The limited bandwidth and the finite battery power of mobile computers represent one of the greatest limitations of current WLANs. In this paper, we deeply investigate the efficiency and the energy consumption of MAC protocols that can be described with a p-persistent CSMA model. As already shown in the literature, the IEEE 802.11 protocol performance can be studied using a p-persistent CSMA model (Cali et al. 2000). For this class of protocols, in the paper, we define an analytical framework to study the theoretical performance bounds from the throughput and the energy consumption standpoint. Specifically, we derive the p values (i.e., the average size of the contention window in the IEEE 802.11 protocol (Cali et al.)) that maximizes the throughput, p/sub opt//sup C/, and minimizes the energy consumption, p/sub opt//sup E/. By providing analytical closed formulas for the optimal p values, we discuss the trade-off between efficiency and energy consumption. Specifically, we show that power saving and throughput maximization can be jointly achieved. Our analytical formulas indicate that the optimal p values depend on the network configuration, i.e., number of active stations and length of the messages transmitted on the channel. As network configurations dynamically change, the optimal p values must be dynamically updated. In this paper, we propose and evaluate a simple but effective feedback-based distributed algorithm for tuning the p parameter to the optimal values, i.e., p/sub opt//sup E/ and p/sub opt//sup C/. The performance of the p-persistent IEEE 802.11 protocol, enhanced with our algorithm, is extensively investigated by simulation. Our results indicate that the enhanced p-persistent IEEE 802.11 protocol is very close to the theoretical bounds both in steady-state and in transient conditions.