Dynamic networks like Online Social Networks or Disruption Tolerant Networks (DTNs), when considering their spatial, temporal and size complexity, even if partly wired, are exposed to nodes and links churns and failures which can be modeled with dynamic graphs with time varying edges and vertices. Recently, it has been shown that dynamic networks exhibit some regularity in their temporal contact patterns. The impact of this regularity on network performances has not been well studied and analyzed. One of the most interesting problem in research on dynamic networks is the issue of efficient navigation techniques in such networks. For dynamic networks, because there is still no widely developed theoretical background to understand deeply the problems, research traditionally tends to propose heuristic solutions. In the context of DTNs, these solutions tend to answer to some specific questions about navigating in a dynamic network (e.g., how to reduce energy consumption of routing, how to maximize the delivery probability) while usually ignoring and not leveraging on the profound structural properties of the dynamic network. In this work, we aim to contribute to understanding the impact of this dynamic structure on information routing and show how to exploit this structure for efficient navigation in such networks.

Supporting end-to-end quality-of-service (QoS) in Large-scale distributed interactive simulations (DIS) is hard due to the heterogeneity and scale of communication networks, transient behavior, and the lack of mechanisms that holistically schedule different resources end-to-end. This paper aims to cope with these problems in the context of wide area network (WAN)-based DIS applications that use the OMG Data Distribution Service (DDS) QoS-enabled publish/subscribe middleware. First, we show the design and implementation of the QoS framework, which is a policy-driven architecture that shields DDS-based DIS applications from the details of network QoS mechanisms by specifying per-flow network QoS requirements, performing resource allocation and validation decisions (such as admission control), and enforcing per-flow network QoS at runtime. Second, we evaluate the capabilities of the framework in an experimental large-scale multi-domains environment. The evaluation of the architecture shows that the proposed QoS framework improves the delivery of DDS services over heterogeneous IP networks, and confirms its potential impact for providing network-level differentiated performance.

As new transport protocols are being proposed and standardized, the choice of the best communication service to be used by applications for delivering their data when distributed is becoming too complex. Application developers need much knowledge on "how the protocol works" to decide whether or not it can be used to fulfill their requirements. Moreover, the performance of the service provided by a given communication protocol is highly dependent on the network context. The Autonomic Transport Protocol presented in this paper is aware of the application requirements and uses learning techniques to adapt the service it provides to best satisfy these requirements as the network conditions vary.

The small-world phenomenon first introduced in the context of static graphs consists of graphs with high clustering coefficient and low shortest path length. This is an intrinsic property of many real complex static networks. Recent research has shown that this structure is also observable in dynamic networks but how it emerges remains an open problem. In this paper, we propose a model capable of capturing the small-world behavior observed in various real traces. We then study information diffusion in such small-world networks. Analytical and simulation results with epidemic model show that the small-world structure increases dramatically the information spreading speed in dynamic networks.

This paper introduces a Swarm-Intelligence based Routing protocol (SIR) that aims to efficiently route information in content centric Delay Tolerant Networks (CCDTN) also dubbed pocket switched networks. First, this paper formalizes the notion of optimal path in CCDTN and introduces an original and efficient algorithm to process these paths in dynamic graphs. The properties and some invariant features of these optimal paths are analyzed and derived from several real traces. Then, this paper shows how optimal path in CCDTN can be found and used from a fully distributed swarm-intelligence based approach of which the global intelligent behavior (i.e. shortest path discovery and use) emerges from simple peer to peer interactions applied during opportunistic contacts. This leads to the definition of the SIR routing protocol of which the consistency, efficiency and performances are demonstrated from intensive representative simulations.

Positioning requirements for mobile nodes in wireless (sensor) networks are largely increasing. However for indoors environments most of the existing solutions are expensive in terms of technology or infrastructure. We propose an indoors positioning scheme in which a static user collects self-estimated positions of collaborating peer nodes within radio range as well as the error ranges of these estimations and determines its position with a linear matrix inequality (LMI) method coupled with a barycenter computation. The approach is evaluated with simulations where the input parameters are the maximum radio range, the number of peer nodes within range, their speed, and dynamic error estimation. The simulations show that the scheme works fine even when the error estimation linearly increases with time. To estimate its position within 1 meter accuracy, the user node may wait at the same position for 5 minutes, with 16 pedestrians moving around him and an error increase rate of 10%.

In operations related to Quality of Service (QoS) management such as SLA negotiations, QoS-aware service invocations and so, it is necessary to specify the required or negotiated quality parameters. Currently, there are no standards to formally describe QoS specifications at the network service level. Several works propose a fixed list of QoS parameters at the network level (e.g., delay, jitter, packet loss). However, solutions offering an extensible list of QoS parameters to specify QoS are more appropriate to deal with the heterogeneity of applications and network technologies. This paper proposes an ontology as a formal and extensible way of specifying QoS and provide the mapping between parameters of different layers. Using our ontology, clients and providers can adopt different user-perceived, application or network QoS parameters and metrics during the service negotiation and invocation, and use ontology reasoning to compare different parameters. The usage of this ontology is illustrated during the establishment of a VoIP session, where a high level QoS specification in the application layer is mapped into a network layer Class of Service defined by the service provider.

End-to-end QoS can be achieved through resource reservations along the communication path between end points. Per flow or per aggregate signaling can be used to set up reservation states in routers or domains on the path. In a mobile environment this reservation process can be very cumbersome, given that (i) the time to set up the resources in the new path is not neglectable and (ii) the resources may not be available during the handover process. Advance reservations is a possible solution for this problem. This paper describes a procedure to compute a set of considered paths in a resource reservation, during a communication session when both hosts are mobile. We depart from a semi Markov user mobility model extracted from regular user movements. We combine both user models to generate a time evolution equation that allows computing the probability to be in a future path from a current path in a given time. We use this function to build an optimisation problem in a binary integer programming formulation. A target probability of lacking resources is used as a constraint to select a subset of paths based on a cost metric.

WIMAX (Worldwide Interoperability for Microwave Access) is a promising new networking technology that potentially offers high speed and wide area wireless access services that complement consistently the 3G and WiFi access networks capabilities. The standard proposes an adaptive modulation scheme which allows WiMax nodes to communicate from various modulation coding schemes according to the link quality. However, the standard does not define a detailed link adaptation algorithm and currently, the most largely used modulation adaptation technique is based on a channel quality lookup table. We argue that this method is not able to make the best adaptation decisions and delivers a sub-optimal goodput in numerous communication contexts. In this paper, we propose a novel cross layer based modulation adaptation mechanism which incorporates the use of adaptive erasure code with the physical layer information to significantly improve the goodput and transmission efficiency. Simulation results show that our proposal adapts more efficiently to real environments and achieves a significant gain on the goodput delivered to mobile nodes.

This work addresses the problem of topology discovery and aggregation. For topology discovery, current solutions concentrate on using probe techniques like the MIBs in routers and gateways, the Internet control message protocol (ICMP), the traceroute program, the domain name system (DNS), etc. In contrast, this paper presents the simple topology announcement and management protocol (STAMP), a simple mechanism which infers the network topology by injecting suitable "hello" adjacency messages between direct connected devices. The discovered topology covers both the data link and the network layers. STAMP can be applied on large as well as on small topologies. In order to overcome scalability issues, we designed some mechanisms to hierarchically structure the discovered topology using aggregation to pull up management functions into the higher levels of the hierarchy. In the first version of STAMP we are interested in the design of a scalable call admission control management function, which has an increasingly important role for quality of service delivery in the new networking systems like IMS (IP multimedia subsystem). The paper presents the key points behind STAMP. It describes the protocol design as well as its behavior and hierarchy. Finally, some preliminary results presenting the application of STAMP in the scope of call admission control are shown in the end of the paper.

Recent papers propose QoS solutions allowing the users to explicitly specify the quality level they request during a so-called explicit QoS service invocation that includes dynamic QoS offering and new mechanisms for authentication, authorization and accounting. In particular, organizations should be able to control their network services and authorize QoS services based on service parameters, as destination addresses, applications and employees’ roles. Therefore, user profile-based authorization policies for explicit QoS service invocations should be adopted. Each user profile indicates the services that the user is authorized to request, their consumption limits and scope. This paper deals with the authorization of explicit QoS services invocations and proposes a user profile model for profile-based authorization of QoS services.

This paper deals with the problem of QoS signaling across multiples domains in a context of topological changes due to mobility. Data path changes caused by mobility can degrade severely the service continuity to mobile terminals considering the end-to-end QoS inter-domain path reestablishment. We propose an efficient NSIS-based scheme to allow end-to-end QoS path maintenance under mobility. The domains are controlled by central resource managers acting as bandwidth brokers. The scheme addresses two open issues in the area: (a) the integration of NSIS with Hierarchical Mobile IP including anchor points collaboration to QoS signaling, and (b) the use of NSIS in an hybrid on and off -path context, exploring the common off-path points to improve signaling. The signaling procedure aims to reduce the impact of mobility in the time to setup new QoS paths and the time to tear down unused resources.

Congestion control is a fundamental mechanism for the stability of the Internet and is a central mechanism for TCP. However, this congestion control mechanism focuses mainly on the core network state and is blind to the characteristics of wireless and mobile access networks. Moreover, TCP window bas

Based on an identification and evaluation of the subtle counterproductive interactions between the WLANs MAC layer and the transport layer, this paper shows a new approach towards congestion control for WLANs. We introduce a specialization of TFRC (MTFRC: mobile TFRC), which is adapted to wireless access networks. This TFRC specialization requires only slight changes to the standard TFRC protocol. Simulation results show substantial improvements for applications over TFRC in scenarios where the bottleneck situates on the MAC layer of the mobile nodes.

The features of mobility, which enormously impact on how communication is evolving into the future, represent a particular challenge in todaypsilas wireless networking research. After an identification and evaluation of the gap between the discontinuities of the communication service inherent to the physical layer of mobile networks and the continuity requirements issue from the stream centric multimedia applications, we propose a novel middleware 3MOI (Middleware for the Mobility Management Over the Internet) which can support efficient and context-aware mobility management and satisfy new mobility requirements such as dynamical location management, fast handover, and continuous connection support.

The expansion of the networks in the last decade, in size and number of services, accentuated several problems. An important one is the management and the control of available resources, in particular, to provide QoS. To deal with this problem, networks are often structured hierarchically and each domain keeps its own detailed state and an aggregated view about the other domains. After aggregation, the amount of information is outstandingly reduced but we can not speak about relevance only if the distortion from the real network state is minimal. This distortion can result from a strict algorithm which leads to reduce unfeasible connections or a generous algorithm allowing to enlarge the amount of feasible connections even if the real network state does not allow it. Although it under-estimates the network capacities, the telecom operators are in favour of the first aggregation algorithm, because it guarantees better the quality of service. In this paper, we propose a new model for network state aggregation bringing a trade off between strict and generous algorithms. Contrary to traditional approaches where the probability of crank back is large, our solution, based on the concept of "supported services guarantee", does not over-estimate the network capacities and keep the admissibility amount reasonable. Our simulations show that the new approximation model achieves the best performance concerning the crank back reduction compared to the other aggregation methods.

Call admission control (CAC) is a crucial element for providing quality of service (QoS) guarantees in a network. Performance deteriorates rapidly within large scale networks. Compared to measurement-based admission control approaches, the performance degradation can be important in the case of parameter- based approaches due to their dependence to the network topology. So, as networks grow in size, the collection and the maintenance of the resources reservation status which must be maintained coherent in the database become difficult, if not impossible. In this paper, we propose a novel admission control approach based on an aggregated scheme of the topology. This approach improves the traditional parameter-based admission control techniques. Simulation results show that our proposed admission control scheme achieves very good performances in the reduction of the memory space utilization and the algorithm complexity.

More and more applications and protocols are now running on wireless networks. Testing such applications and protocols is a real challenge as the position of the mobile terminals and environmental effects strongly affect the overall performance. Network emulation is often perceived as a good trade-off between experiments on operational wireless networks and discrete-event simulations on Opnet or ns-2. However, ensuring repeatability and realism in network emulation while taking into account mobility in a wireless environment is very difficult. This paper proposes a network emulation architecture based on off-line computations preceding online pattern-based traffic shaping. The underlying concepts of repeatability, dynamicity and accuracy are defined in the emulation context. Three different simple case studies illustrate the validity of our approach with respect to these concepts.

In the next generation Internet, the network should evolve from a plain communication medium into an endless source of services available to the end-systems. We name these services "overlay applications". They would be composed of multiple cooperative distributed application elements that would build a dynamic communication mesh, namely "overlay association". In a former contribution, we proposed an unstructured super-peer architecture (SPAD) that provides enhanced quality of service (QoS) between end-points within an overlay association. This architecture aims at discovering and utilizing composite alternate end-to-end paths that experience better QoS than the path given by the default IP routing mechanisms. This paper presents a proactive information dissemination scheme that complements SPAD's mechanisms and significantly improves its performances.

Face to the complexity involved in the wide deployment of QoS network services as well as the reduced set of services offered by existing transport protocols, this paper proposes a QoS-oriented transport protocol (QoSTP) as the adequate solution for providing soft QoS guarantees to common Internet users for next few years. The design of this QoSTP is based on a set of fundamental principles aimed at assuring the feasibly and efficient deployment of adequate mechanisms intended to satisfy different applications requirements. Moreover, this new protocol has to be designed within an extensible and compositional architectural aimed at integrating further mechanisms intended to satisfy new requirements and to operate under heterogeneous network environments. This paper presents the UML 2.0 specification model of this advanced architecture as well as experimental results intended to demonstrate the feasibility and the advantages of this approach.

Transport protocols implement mechanisms intended to satisfy the requirements of distributed applications making use of available network services. But the multiplicity of network services and the diversity of quality of service (QoS) requirements of current multimedia applications, make that efficient deployment of transport services becomes a complex function to be performed. The fully programmable transport protocol (FPTP) is a configurable and object-oriented protocol based on a compositional architecture of transport mechanisms suited to the provision, control, and management of the QoS. FPTP offers a partially ordered, partially reliable, congestion controlled and time controlled end-to-end communication service. FPTP services are intended to be easily used by legacy and specialized multimedia applications over heterogeneous transport and network services. In this paper, the unified modelling language (UML) and the specification and description language (SDL) have been used to design this transport protocol within a global QoS framework. Some study cases related to the configurable congestion control policies and aimed to demonstrate the advantages this proposal presented.

Wireless networks have become increasingly popular for the past few years, especially IEEE 802.11 as the most widespread wireless technology. This article proposes a new architecture for broadband 802.11 networks to support distributed multimedia applications. After discussing the differences between multimedia and traditional applications, we show how the 802.11b mechanism ensuring enhanced robustness connectivity for low signal to noise ratio (SNR) areas, penalizes multimedia stream transmissions. A new multimedia architecture disabling the dynamic rate shifting mechanism and using a performance enhancing proxy is then proposed.

This paper presents an analysis of the main requirements to support synchronous collaborative activities. Then, a collaborative platform, PLATINE, composed of a complete set of tools for synchronous group work support and collaboration session management, is presented. Focus is made on two important components: videoconference and session manager. Afterwards, a set of experiments is described. A co-design scenario in the electronic domain is detailed.

The paper assesses the potential of Java for developing both portable and efficient communication software. Specifically, this paper presents and compares implementation of a multimedia transport protocol, POC, using the Java and C languages. Performances measurement are applied to the transport of MPEG streams.

A partial order and partial reliable connection (POC) is an end-to-end transport connection authorized to deliver objects in an order that can differ from the transmitted one. The service that this connection provides is also authorized to lose a number of objects. The POC approach establishes a conceptual link between both connectionless best-effort (CL) and connection oriented (OC) protocols. This approach is motivated by the fact that heterogeneous CL networks such as the Internet are plagued by unordered delivery of packets and losses, which tar the performances of current protocols. Moreover, it has been showed, in several research works, that out of order delivery is able to alleviate the use of network resources such as memory and bandwidth, and reduces end-re-end transit delay. But, in order to take advantage of these benefits, applications must be able to relax some transport constraints. A temporal extension of POC, called TPOC (for Temporal POC), is introduced. TPOC offer a conceptual framework, which allow the QoS of distributed multimedia applications to be fully taken into account. The architecture for offering a TPOC transport service is introduced and evaluated for the transport of MPEG video streams. The authors demonstrate that POC connections fill not only the conceptual gap between CL protocols (such as UDP) and OC protocols (such as TCP) but also provides real performances improvements for the transport of multimedia streams such as MPEG video.

The paper presents the architecture and the design of a distributed videoconferencing application called N-TSVS. The structure of the cooperative groups supported by N-TSVS is defined and modelled by graphs: their vertices describe the group members and their edges represent the information sharing between the cooperative agents. The group dynamicity is modelled by valid configurations that are subgroups of members in which data exchanges can occur. A group membership service and an associated protocol that manage the dynamic cooperative groups are formally designed, validated and integrated inside N-TSVS. The multimedia data processing is developed in a separate way using multithreading functionalities and is coupled to the group membership service.

This paper presents an architecture that enforces time requirements and gives minimal end-to-end delays for multimedia applications. The layers and mechanisms allowing the system to fulfill the selected synchronization, i.e., the logical relationships and timed interval semantics, are presented. The proposed approach relies on the use of a formal model based on extended time Petri nets, i.e., the time stream Petri net model (TStreamPN), that allows the user to completely specify the time requirements of a given application. The architecture implements, in the application layer and on top of asynchronous environments, the requested quality of service (perceived by the user) with respect to time. At the transport layer, the use of a partial order transport service improves the reactive response of the communication transfers. Its principles are presented together with a presynchronization sublayer that makes the partial order transport service match the applicative synchronization requirements. Moreover, measurements on the implementation of a videoconference system show that the requirements of the quality of service are fulfilled.

The paper introduces a unified formal model for the complete and accurate specification of both temporal and logical (i.e., link) synchronization within hypermedia distributed and weakly synchronous systems. This new model extends time Petri nets with hierarchical design capabilities and new firing rules. These new firing rules enlighten the notion of hypermedia synchronization through powerful combinations of temporal and logical synchronization. Moreover, the proposed model offers an easy and concise modeling technique of asynchronous events in hypermedia systems.

Investigates a partial-order connection (POC) service/protocol. Unlike classic transport services that deliver objects either in the exact order transmitted or according to no particular order, POC provides a partial-order service, i.e. a service that requires some, but not all objects to be received in the order transmitted. Two versions of POC are proposed: reliable, which requires that all transmitted objects are eventually delivered, and unreliable, which permits the service to lose a subset of the objects. In the unreliable version, objects are more finely categorized into one of three reliability classes depending on their temporal value. Two metrics based on e/sub i/(P), the number of linear extensions of partial-order P in the presence of i lost objects, are proposed as complexity measures of different combinations of partial order and reliability. Formulae for calculating e/sub i/(P) are derived when P is series-parallel. A formal specification of a POC protocol, written in Estelle, is presented and discussed. This specification was designed and validated using formal description tools and provides a basis for future implementations.>

Proposes the observer concept for designing self-checking distributed systems, i.e. systems that detect erroneous behaviors as soon as errors act at some observable output level. The approach provides a solution to build systems whose on-line behavior is checked against a formal model derived from a formal description. In other words, the actual implementation is continuously checked against a reference, this reference being a formal and verified model of some adequately selected aspects of the system behavior. The corresponding methodology, the software concepts and some applications of the observer are presented. General definitions are given first that theoretically define self-checking systems as systems that include and implement complete on-line validation. The basic concepts and the difficulties to implement self-checking validation are then given. In order to provide simple implementations, the previous definitions are weakened to design quasi-self-checking observers for LANs using a broadcast service. Three specific applications are given to illustrate the proposed approach: testing a virtual ring MAC protocol, checking the link and transport layers in an industrial LAN, and managing a complete OSI layering, from layer 2 to layer 6, in an open system architecture.>

This paper introduces a partial order connection (POC) protocol. Motivated in particular by multimedia applications, POC is an end-to-end connection that provides a partial order service, that is, a service that requires some, but not all objects to be received in the order transmitted. This paper discusses R-PO, a reliable version of POC which requires that all transmitted objects are eventually delivered. A metric based on the number of linear extensions of a partial order in the presence of no lost objects is proposed to quantify different partial orders. Means for its calculation is presented when P can be modeled as a combination of sequential and/or parallel compositions of Petri-nets. This metric allows one to compare and evaluate the complexity of different partial order services.>

A description and analysis of concurrent systems, such as communication systems, whose behavior is dependent on explicit values of time is presented. An enumerative method is proposed in order to exhaustively validate the behavior of P. Merlin's time Petri net model, (1974). This method allows formal verification of time-dependent systems. It is applied to the specification and verification of the alternating bit protocol as a simple illustrative example.>

This paper presents a fault-tolerant distributed system designed for real-time control applications (REBUS), which is one of the research basis of the industrial real-time system MODUMAT 800. It is made up of functional units, i.e., programmable multiloop regulators and operator displays, linked together by a communication structure. The communication hardware consists of a set of serial bus interface boards, one per functional unit, loosely coupled together by a double serial bus and linked to their functional units by a private parallel bus.

This correspondence deals with a unification and extension of some previous work on self-checking (SC) and fail-safe (FS) systems.

An APL simulation is developped in the case of parallel systems which communicate with the environment. This method using interpreted PETRI nets, helps the design verification.

Fail-safe sequential machines can be constructed in such a way that if a failure happens in the sequential part, the ulterior functioning must carry on outside the code chosen to represent the set of states. This paper presents a study of the failures in the input combinational circuit and of the feasibility conditions of sequential machines with states coded by a k-out-of-n code. The electronic circuit is realized in a classical way (on-set realization) and must obey two hypotheses, 1) no failure on clock line C, and 2) single fault (stuck at 0 or stuck at 1) on other connections than C.