International audience

As the ever growing mobile data traffic challenges the economic viability and performance of cellular networks, innovative solutions that harvest idle user-owned network resources are gaining increasing interest. In this work, we propose leasing wireless bandwidth and cache space of residential 802.11 (WiFi) access points (APs) for offloading mobile data. This solution not only reduces cellular network congestion, but, due to caching, improves also the user-perceived network performance without overloading the backhaul links of the APs. To encourage residential users to contribute their bandwidth and cache resources, we design monetary incentive (reimbursement) schemes. The offered reimbursements directly determine the amounts of available bandwidth and cache space in every AP, which in turn affect the caching policy (where to cache each content file) and the routing policy (where to route each mobile data request). In order to reduce operator's total cost for serving mobile data requests and leasing resources, we introduce a framework for the joint optimization of incentive, caching, and routing policies. Using a novel WiFi usage dataset collected from 167 residences, we show that in densely populated areas with relatively costly network capacity upgrades, our proposal can halve operator's total cost, while reimbursing up to 9€ per month each residential user.

In this paper, we analyze a large dataset of passive wireless measurements and obtain insights about wireless performance. We monitor 167 homes continuously for 4 months from the vantage point of the gateway, which allows us to capture all the activity on the home wireless network. We report on the makeup of the home wireless network, traffic activity, and performance characteristics. We find that in most homes, a small number of devices account for most of the observed traffic volume and the bulk of this traffic activity occurs in the evenings. Studying link performance, we find that overall, the vast majority of transmissions are carried out at high data rates and the wireless networks have good coverage. We find a small number of episodes where performance is poor; a few homes have a disproportionate number of poor performance reports. Investigating further, we observe that most of these are not caused by poor coverage (pointing to network interference). Our results significantly add to the understanding of home wireless networks and will help ISPs to understand their subscriber networks.

The success of broadband residential Internet access is changing the way home users consume digital content and services. Currently, each home service requires the installation of a separate physical box (for instance, the NetFlix box or IPTV set-top-boxes). Instead, we argue for deploying a single box in the home that is powerful and flexible enough to host a variety of home services. In addition, this box is managed by the Internet Service provider and is able to provide service guarantees. We call such a box a service-hosting gateway (SHG), as it combines the functionalities of the home gateway managed by the network service provider with the capability of hosting services. Isolation between such services is ensured by virtualization.

Anomaly detection methods typically operate on preprocessed traffic traces. Firstly, most traffic capturing devices today employ random packet sampling, where each packet is selected with a certain probability, to cope with increasing link speeds. Secondly, temporal aggregation, where all packets in a measurement interval are represented by their temporal mean, is applied to transform the traffic trace to the observation timescale of interest for anomaly detection. These preprocessing steps affect the temporal correlation structure of traffic that is used by anomaly detection methods such as Kalman filtering or PCA, and have thus an impact on anomaly detection performance. Prior work has analyzed how packet sampling degrades the accuracy of anomaly detection methods; however, neither theoretical explanations nor solutions to the sampling problem have been provided. This paper makes the following key contributions: (i) It provides a thorough analysis and quantification of how random packet sampling and temporal aggregation modify the signal properties by introducing noise, distortion and aliasing. (ii) We show that aliasing introduced by the aggregation step has the largest impact on the correlation structure. (iii) We further propose to replace the aggregation step with a specifically designed low-pass filter that reduces the aliasing effect. (iv) Finally, we show that with our solution applied, the performance of anomaly detection systems can be considerably improved in the presence of packet sampling.

In opportunistic networks, end-to-end paths between two communicating nodes are rarely available. In such situations, the nodes might still copy and forward messages to nodes that are more likely to meet the destination. The question is which forwarding algorithm offers the best trade off between cost (number of message replicas) and rate of successful message delivery. We address this challenge by developing the PeopleRank approach in which nodes are ranked using a tunable weighted social information. Similar to the PageRank idea, PeopleRank gives higher weight to nodes if they are socially connected to important other nodes of the network. We develop centralized and distributed variants for the computation of PeopleRank. We present an evaluation using real mobility traces of nodes and their social interactions to show that PeopleRank manages to deliver messages with near optimal success rate (close to Epidemic Routing) while reducing the number of message retransmissions by 50% compared to Epidemic Routing.

In opportunistic ad-hoc networks, multi-hop data transfer over contemporaneous paths is unlikely since the devices are often disconnected from each other. However, data can still be stored and forwarded over time in an opportunistic hop-by-hop manner. Previous work has considered how the availability of various types of information such as social relationships can be used to guide forwarding algorithm to make better decisions and bring messages closer to the destination. This implicitly assumes that opportunistic contacts relate with the social property of node. However, the impact of such correlation between social and contact properties on social forwarding performances remains largely unexplored. In this paper we argue that the relevance of such social information (social inputs) is as important as designing a new social forwarding algorithms. We examine multiple datasets to determine the impact of correlation, if any, between social information of individuals and their mobility patterns on the forwarding performances. We propose methods which process the social inputs to improve the relevance of such social information to forwarding. We show that our processing methods could improve the success rate performances of many social forwarding algorithms by more than 30%.

The objective of autonomic networking is to enable the autonomous formation and parametrization of nodes and networks by letting protocols sense and adapt to the networking environment at run time. Besides its dynamic aspects, a core requirement of autonomic networking is to define a structured framework and execution environment that enables algorithms to operate in a continously changing environment. This paper presents the major design principles of the Autonomic Network Architecture (ANA) and reports on a first implementation. The guiding principle of ANA is to strive for flexibility and genericity at all levels of the architecture. In our approach we explicitly avoid to impose a "one-size-fits-all" architecture (where communication protocols and paradigms are fixed by the architecture). To this end, ANA introduces generic abstractions, for example "information dispatch points" instead of addressable endpoints, as well as communication primitives that support network heterogeneity, adaptability, and evolution. These core abstractions allow for the coexistance of multiple and diverse networking styles and protocols. With the public release of the ANA prototype, we aim at federating autonomics related networking projects, enabling different actors to share, compare, and build upon each other¿s work. The ANA runtime can host clean slate network designs as well as legacy Internet technology and serves as a platform for demonstrating autonomic communication principles.

Spatial Principal Component Analysis (PCA) has been proposed for network-wide anomaly detection. A recent work has shown that PCA is very sensitive to calibration settings. Unfortunately, the authors did not provide further explanations for this observation. In this paper, we fill this gap and provide the reasoning behind the found discrepancies. We revisit PCA for anomaly detection and evaluate its performance on our data. We develop a slightly modified version of PCA that uses only data from a single router. Instead of correlating data across different spatial measurement points, we correlate the data across different metrics. With the help of the analyzed data, we explain the pitfalls of PCA and underline our argumentation with measurement results. We show that the main problem is that PCA fails to capture temporal correlation. We propose a solution to deal with this problem by replacing PCA with the Karhunen-Loeve transform. We find that when we consider temporal correlation, anomaly detection results are significantly improved.

In this paper, we propose a fully distributed network model for rate control in application layer multicast based on the utility-price model to maximize the aggregate utilities, and accordingly design an original primal algorithm and a typical dual algorithm both with very small messaging overhead.

In this paper, we study the topic of distributed and optimal congestion control for scalable video streams in application-layer multicast (ALM). We propose a TCP-friendly, fully distributed synchronous algorithm based on the utility-price model which maximizes the global utilities for the streams in the application-layer multicast tree. With the help of numerical study, we show that our proposed algorithm optimizes the overall video quality for fine-grained scalable streams, while minimizing the messaging overhead in the application-layer multicast channel.

A majority of network architectures aim at solving specific shortcomings of the original Internet architecture. While providing solutions for the particular problems, they often lack in flexibility and do not provide general concepts for future networking requirements. In contrast, we introduce a network architecture that aims to be versatile enough to serve as a foundation for the future Internet. The main pillars of our architecture are communication pivots called information dispatch points (IDPs) which embed the concept of modularity at all levels of the architecture. IDPs completely decouple functional entities by means of indirection thus enabling evolving protocol stacks. Our architecture also provides a consistent application programming interface (API) to access node-local or network-wide functionality. In addition to the description of this architecture, we report about a working prototype of the architecture and we give examples of its application.

The development of mobile ad hoc systems have considerably emphasized the need for a better understanding of the factors that influence the systemspsila performance, i.e., mobility patterns, radio propagation, traffic characteristics and their interrelations. Depending on the context where the system is being used-whether at work, at home, or in some means of transportation-there will be various applications that provide benefit of using the system. These different applications in turn generate traffic with very heterogeneous characteristics. In addition, the radio propagation depends on the surrounding environment and the density of communicating nodes in the system and in other systems that compete for radio resources; hence the available data rates in the system are hard to ascertain. The uncertainty regarding both the traffic and the available data rates make it hard to evaluate the performance of the system. This uncertainty is due to the lack of deployed systems. We hence believe that research in mobile ad hoc networks has to couple theory and practice in a continuous feedback loop with experimentations running full-fledged applications. We show how this methodology has proved to be beneficial in order to overcome challenges and to determine main issues for future research in the context of our PodNet project. It has also allowed us to revisit assumptions and scepticism surrounding the feasibility of mobile ad hoc networks. PodNet is an architecture, instantiated in a prototype, that is dedicated to cooperative content distribution. It enables exchange of podcasts (for any kind of multimedia content) amongst mobile devices in a peer-to-peer fashion using IEEE 802.11 in ad hoc mode. PodNet has clear advantages over traditional content distribution approaches (e.g., newspapers, 3G) in terms of public availability, practicability, capacity, resilience to failures, and jamming.

Interference is an inherent property of wireless multihop networks. Adding interference-awareness to their control functions can significantly enhance the overall network performance. In this paper we present an analytical model for the probability that a transmission destined to an arbitrary network node is successful in the presence of interference from other nodes in the network. We introduce the concept of interference areas and interference zones to express this probability as a function of the network density, node transmission probability, radio propagation environment, and network card reception sensitivity. Our derivation includes a simpleMAC model, which captures the carrier sense function of many MAC protocols. Contrary to measurementbased models, our derivation only requires information that is locally available to the nodes, avoiding all measurement-related pitfalls. The validation of our model against experiments in a real testbed, set up for this purpose in our indoor office environment, shows good match of the experimental results with the analytical predictions. Interestingly our model predictions follow closely those of more elaborate state-of-the-art analytical models. Finally, to demonstrate the real utility of our model, we have implemented on our testbed a routing metric that explicitly takes interference into account via our derivation. The throughputs of the resulting routes compare favorably with those achieved by a well-known probe-based routing metric.

The eight papers in this special issue focus on delay and disruption tolerant wireless communication. The papers cover routing and network coding for spare mobile ad hoc networks, cross-layer design for sensor networks, satellite communication, and DTN architectural issues.

In this comments paper, we revisit the network model introduced in Cui, Y., et al (2006). We discuss the inaccuracy of the model and, to correct the network model, we propose to apply directed capacity constraints for directed flows. Based on a comparison of numerical results, we show that the corrected model leads to better accuracy than the original model.

In this paper, we present two new tree growing algorithms. The pairing algorithm allows for the local approximate implementation of global algorithms such as Prim's or Dijkstra 's algorithm. A node requires only information from its neighborhood acquired by local message exchange. Any global cost function that can be locally calculated can be used with this algorithm. The N-SafeLinks Algorithm establishes look-up tables with N possible links per node. Implementing an additional constraint it is guaranteed that each link leads to the sink, ruling out the possibility of loops. Therefore, if maximal N-l links per node are broken there is still a guaranteed connection to the destination node (sink) for every node. A proof of this property is presented. As demonstrated in simulations the trees are close to optimum. The algorithms can be utilized for static routing in wireless single-sink Ad-Hoc networks with safety critical applications where timeliness, robustness and energy efficiency is crucial.

Existing unicast routing protocols are not suited well for wireless mesh networks as in such networks, most traffic flows between a large number of mobile nodes and a few access points with Internet connectivity. In this paper, we propose HEAT, an anycast routing protocol for this type of communication that is designed to scale to the network size and to be robust to node mobility. HEAT relies on a temperature field to route data packets towards the Internet gateways, as follows. Every node is assigned a temperature value, and packets are routed along increasing temperature values until they reach any of the Internet gateways, which are modeled as heat sources. Our major contribution is a distributed protocol to establish such temperature fields. The distinguishing feature of our protocol is that it does not require flooding of control messages. Rather, every node in the network determines its temperature considering only the temperature of its direct neighbors, which renders our protocol particularly scalable to the network size. We analyze our approach and compare its performance with OLSR through simulations with Glomosim. We use realistic mobility patterns extracted from geographical data of large Swiss cities. Our results clearly show the benefit of HEAT versus OLSR in terms of scalability to the number of nodes and robustness to node mobility. The packet delivery ratio with HEAT is more than two times higher than OLSR in large mobile scenarios and we conclude that HEAT is a suitable routing protocol for city-wide wireless mesh networks.

Packet sampling methods such as Cisco's NetFlow are widely employed by large networks to reduce the amount of traffic data measured. A key problem with packet sampling is that it is inherently a lossy process, discarding (potentially useful) information. In this paper, we empirically evaluate the impact of sampling on anomaly detection metrics. Starting with unsampled flow records collected during the Blaster worm outbreak, we reconstruct the underlying packet trace and simulate packet sampling at increasing rates. We then use our knowledge of the Blaster anomaly to build a baseline of normal traffic (without Blaster), against which we can measure the anomaly size at various sampling rates. This approach allows us to evaluate the impact of packet sampling on anomaly detection without being restricted to (or biased by) a particular anomaly detection method.We find that packet sampling does not disturb the anomaly size when measured in volume metrics such as the number of bytes and number of packets, but grossly biases the number of flows. However, we find that recently proposed entropy-based summarizations of packet and flow counts are affected less by sampling, and expose the Blaster worm outbreak even at higher sampling rates. Our findings suggest that entropy summarizations are more resilient to sampling than volume metrics. Thus, while not perfect, sampling still preserves sufficient distributional structure, which when harnessed by tools like entropy, can expose hard-to-detect scanning anomalies.

Existing anycast routing protocols solely route packets to the closest group member. In this paper, we introduce density-based anycast routing, a new anycast routing strategy particularly suitable for unstable networks. Instead of routing packets merely on proximity information to the closest member, density-based anycast routing considers the number of available anycast group members for its routing decision. To evaluate the benefits of densitybased routing, we present a unified model to analyze pure proximitybased, pure density-based, as well as combined routing strategies. With an extensive simulation study, we then evaluate these strategies in multiple mobile scenarios. The two main results are that (i) density-based routing increases the probability of successful packet delivery when the network is unstable; and (ii) for particular mobile scenarios, density-based routing finds even shorter routes compared to traditional proximity-based routing. Finally, we discuss implementation issues and propose a solution to dynamically adapt the protocol’s parameter settings. I.

End-to-end transport protocols perform poorly in mobile environments, primarily due to the frequent route breaks induced by node mobility. We present a framework for hop-by-hop transport protocols and propose a new reliable transport protocol: SaFT (store-and-forward transport). We evaluate its performance in mobile ad hoc networks vis-a-vis TCP. Overall, SaFT achieves up to 3 times faster delivery of messages in such networks. We conclude that store-and-forward protocols are suited well to provide reliable communication in mobile ad hoc networks

This paper analyzes the characteristics of a multi-hop 802.11b mobile ad hoc network. We present data gathered from a mobile network of 20 devices carried by test users over 5 days in an indoor environment. The data is analyzed with regard to (i) the number of reachable devices, (ii) the node degree, (iii) the average path length, (iv) the link lifetime, (v) and the route lifetime. Despite the relatively high node density and low node mobility in our setup, we observe frequent network partitioning and considerably high path lengths (as large as 7 hops). However, the usability of these long paths is questionable as their lifetime is short. We believe that our measurements are representative for typical indoor environments and that the results can and should be used for evaluating networking protocols as well as to validate existing or to derive new mobility models

This paper presents a media- and TCP-friendly rate-based congestion control algorithm (MTFRCC) for scalable video streaming in the Internet. The algorithm integrates two new techniques: i) a utility-based model using the rate-distortion function as the application utility measure for optimizing the overall video quality; and ii) a two-timescale approach of rate averages (long-term and short-term) to satisfy both media and TCP-friendliness. We evaluate our algorithm through simulation and compare the results against the TCP-friendly rate control (TFRC) algorithm. For assessment, we consider five criteria: TCP fairness, responsiveness, aggressiveness, overall video quality, and smoothness of the resulting bit rate. Our simulation results manifest that MTFRCC performs better than TFRC for various congestion levels, including an improvement of the overall video quality.

The manufacturing industry is nowadays subjected to an enormous competition pressure. The concept of the factory was introduced as the use of high-performance PCs has witnessed a striking expansion. It encompasses the employment of various tools that allow digitalizing, simulating and evaluating contemporaneously product design and production development. However, a lot of hurdles are still presented, when trying to apply results of digital simulations to the real production world. This justifies the need to focus research efforts in overcoming these inefficiencies. In this paper an approach based on a programming tool will be described. It aims at enabling the automatic generation of robot trajectory paths from different free forming surfaces of existing 3D-CAD models by means of the linear and circular interpolation. First trial tests have been carried out within a real robot based for windshields and rear windows sticking in co-operation with a big German automotive enterprise. These show a high system performance and stimulate further system improvements

In mobile ad hoc networks, we envision a network where mobile users obtain services from close-by instances. The architecture of today's Internet was designed for fixed users that obtain services from stationary servers and is not well suited for such scenarios. The reason is that (i) the architecture combines identity and location in an IP address and thus forces mobile elements to change their identity when moving over subnet boundaries; and that (ii) the layered architecture implies a separation of service discovery/selection and routing, which is inflexible and also leads to protocol overhead. In this paper, we revise the existing Internet architecture and propose a novel architecture that is better suited for mobile ad hoc networks. There, clients bind to location-independent service identifiers and send packets that are routed to any instance of the desired service in proximity. The routing mechanism is based on the concept of (electrical) fields with which packets are forwarded towards a region with a high density of service nodes. As a result, this architecture increases the probability of successful packet delivery and leads to a robust routing substrate even in very unstable network conditions

Service discovery in mobile ad hoc networks is challenging because of the absence of any central intelligence in the network. Traditional solutions, as used in the Internet, are hence not well suited for mobile ad hoc networks. We present a novel decentralized service discovery mechanism for ad hoc networks. The basic idea is to distribute information about available services to the network neighborhood. We achieve this by using the analogy of an electrostatic field. A service is modeled by a (positive) point charge, and service request packets are seen as (negative) test charges which are attracted by the service instances. We map the physical model to a mobile ad hoc network in a way where each network element calculates a potential value and routes service requests towards the neighbor with the highest potential, hence towards a service instance. Our approach allows for differentiation of service instances based on their capacity. We define the required protocols and methods which we implemented in a network simulator. Using extensive simulations, we evaluate the performance and robustness of the mechanisms. The results indicate good performance and convergence, even in highly mobile environments. We believe that this technique can, and should, be further exploited, e.g., as a routing protocol in mobile networks.

This paper compares the performance of tail drop and three different flavors of the RED (random early detection) queue management mechanism: RED with a standard parameter setting, RED with an optimized parameter setting based on a model of RED with TCP flows, and finally a version of RED with a smoother drop function called "gentle RED ". The performance is evaluated under various load situations for FTP-like and Web-like flows, respectively. We use measurements and simulations to evaluate the performance of the queue management mechanisms and assess their impact on a set of operator oriented performance metrics. We find that in total (i) no performance improvements of RED compared to tail drop can be observed; (ii) fine tuning of RED parameters is not sufficient to cope with undesired RED behavior due to the variability in traffic load; and (iii) gentle RED is capable of resolving some of the headaches on RED but not all.

End-to-end congestion control mechanisms such as those in TCP are not enough to prevent congestion collapse in the Internet, and they must be supplemented by control mechanisms inside the network. The IRTF has singled out random early detection (RED) as one queue management scheme recommended for rapid deployment throughout the Internet. However, RED is not a thoroughly understood scheme-witness for example how the recommended parameter setting, or even the various benefits RED is claimed to provide, have changed over the past few years. In this paper, we describe simple analytic models for RED, and use these models to quantify the benefits (or lack thereof) brought about by RED. In particular, we examine the impact of RED on the loss and delay suffered by bursty and less bursty traffic (such as TCP and UDP traffic, respectively). We find that: (i) RED does eliminate the higher loss bias against bursty traffic observed with tail drop, but not by decreasing the loss rate of bursty traffic, rather by increasing that of non bursty traffic; (ii) the number of consecutive packet drops is higher with RED than tail drop, suggesting RED might not help as anticipated with the global synchronization of TCP flows; (iii) RED can be used to control the average queueing delay in routers and hence the end to end delay, but increases the jitter of non bursty streams. Thus, applications that generate smooth traffic, such as interactive audio applications, will suffer higher loss rates and require large playout buffers, thereby negating at least in part the lower mean delay brought about by RED.

Schemes based on the tagging of packets have been proposed as a low-cost way to augment the single class best effort service model of the current Internet by including some kind of service discrimination. Such schemes have a number of attractive features, however, it is not clear exactly what kind of service they would provide to applications. Yet quantifying such service is very important to understand the benefits and drawbacks of the different tagging schemes and of the mechanisms in each scheme (for example how much RED with input and output (RIO) contributes in the assured scheme), and to tackle key performance and economic issues (e.g. the difference in tariff between different service classes would presumably depend on the difference in performance between the classes). The goal in this paper is to obtain a quantitative description of the service provided by tagging schemes. Specifically, we describe and solve simple analytic models of two previously proposed schemes, namely the assured service scheme and the premium service scheme. We obtain expressions for performance measures that characterize the service provided to tagged packets, the service provided to non-tagged packets, and the fraction of tagged packets that do not get the better service they were supposed to. We use these expressions, as well as simulations and experiments from actual implementations, to illustrate the benefits and shortcomings of the schemes.