BIONETS. Application of genetic exchange rules to pervasive networks and parallels from system science

Imrich Chlamtac, joint work with Iacopo Carreras and Hagen Woesner
CreateNet Research Association

Abstract. As the trend toward ubiquitous and pervasive computing continues to gain momentum, the number of nodes is expected to grow by multiple orders of magnitude as tags, sensors, body networks and myriad of other miniaturized devices get fully integrated into the global communication superstructure. Not only will the amount of information in these all-embracing pervasive environments be enormous and to a large degree localized, but also the ambience within which these nodes will act will be intelligent, mobile, self-cognitive, knowledge based, and, in a sense, "almost alive".Current Internet protocol conceived almost fourty years ago, were never planned for these emerging pervasive environments.

Besides their many limitations with respect to agility, mobility, scalability, etc., they are also too heavyweight in the computing and communication requirements they impose on the small and energy-limited nodes. Hence a totally new way of controlling these emerging systems is much needed.

Realizing that the raison d'etre for the protocol layers is the support of better services to the user, we consider an approach to the pervasive environments that is derived from the network's original goal - services.  Given, as observed above, that the pervasive systems are dynamic, growing,  self aware and evolving, it is not surprising they can be looked at as biological entities. Considering the behaviour and hence control of these pervasive environments as evolving organism, leads us to applying the rules of combination of genetic material and evolution to determine the information exchange, fully replacing the concept of end to end Internet oriented protocols. By information exchange occuring in these systems only as an individual need of each entity in the biosystem to reproduce, the rules of control become localized, minimized and occur only when and where needed, making it possible for the individual nodes to work with minimal energy, with computing or communication effort expended only as it  benefits their own goals.
Even more importantly, by defining themselves as the end result of the evolution process, the services here become characterized by their ability to mutate and select the fittest to survive, this way constantly evolving and self-optimizing leading to a new concept of self aware, self optimizing bionetwork, a candidate for the new world of omnipresent low cost pervasive environments of the future.

---

An Infrastructure-based Approach to Support Dynamic Networks with Mobile Agents
A. Dühler, C. Erfurth, W. Rossak (Friedrich-Schiller Uni Jena)

Abstract. With the growing size of distributed systems and the higher number of available resources and services in networks dynamical aspects become more and more important in systems engineering. We believe that there is a real need
for decentral, self-organizing structures to cope with the upcoming challenges. In this paper we describe an framework which provides a self-organizing infrastructure that allows to link otherwise autonomous elements in a flexible way and adapts dynamically to changes in the underlying network. This framework is implemented as an extension of the mobile agent system Tracy, which is also a product of our university. The Tracy Domain Management module is part of the framework and provides the basis for segmenting the infrastructure.  Another module we are going to discuss in this paper facilitates autonomous and proactive routing of mobile agents. Agents form the application layer of the system. Routing is triggered by the needs an agent inherits from its owner and then matched to the resources and services available in the network in an iterative fashion. We describe concepts, design issues and first results of our work with Tracy and the use of these additional Tracy modules.

---

Some Requirements for Autonomic routing in self-organized networks
F. Legendre, M. Dias de Amorim, S. Fdida (LIP6/CNRS)

Abstract. This paper addresses some requirements of self-organizing networks as well as interoperability problems due to merges and splits phenomena. In a mobile environment, merges and splits characterize the spatial overlap between two self-organized networks. While merge refers to the time when two disjoint networks meet and overlap, split refers to the time of partition. In a dynamic environment, AutoComm (AC) principles bring a new support for interoperability since current protocol heterogeneity is observed at all stack layers from the radio interface to applications. In this paper, we reconsider the formalization of a community and its requirements. We then characterize the split and merge phenomena and their implications. We give some requirements that must fulfill solutions to merging (high context-awareness) in order for AC groups to self-scale. Finally, we propose a merging solution for overlapping wireless self-organized networks using heterogeneous routing protocols.

---

Policy Interoperability and Network Autonomics
S. Magrath, R. Braun, F. Cuervo (Uni of Technology, Sydney and Alcatel Canada)

Abstract. Autonomic behaviours in network operations will alleviate much of the labour intensive and error prone interventions of today's complex networks. The Service Provider must be able to manage the infrastructure and services at an abstract level, focusing on what the desired behaviour should be rather than how it might be specifically achieved. Policy-Based Network Management (PBNM) appears as one of the leading mechanisms to describe desired behaviours and abstract the programmability of an autonomic network infrastructure to the Service Provider. For massive-scale and complex networks, the current understanding of the Higher Level to Lower Level (HL -> LL) refinement process commonly used in PBNM today is not completely effective. One problem encountered is the need to provide a bind mechanism between Higher Level and Lower Level policy specifications such that cross-layer policy requests in the policy continuum can be made by lower policy layers in a dynamic policy refinement cycle (LL -> HL -> LL). In this paper, we illustrate the problem with a policy-based simple admission control (SAC) application. We then show that policy specifications with a join operator simplify the SAC specification. We also investigate the performance considerations of this enhancement in Internet size applications. Our future goal is to provide a policy inference engine that can support complex specifications appropriate for PBNM systems that support autonomic behaviours in large networks, made of Network elements with realistic memory and processing constraints.

Spatial Computing: an Emerging Paradigm for Autonomic Computing and Communication
F. Zambonelli, M. Mamei (DISMI - Università di Modena e Reggio Emilia)

Self-Deployment, Self-Configuration: Critical Future Paradigms for Wireless Access Networks
F. J. Mullany, L. T. W. Ho, L. G. Samuel, H. Claussen (Bell Labs Research, Lucent Technologies, UK)

Abstract. To combat the increasing significance of deployment and configuration costs, the concept of a self-deploying, self-configuring radio access network is discussed.  It is proposed that the basic sciences of complex systems (cellular automata, game theory, ecology modeling) can be exploited to design algorithms for such a system.  An example, taken from the field of cellular automata, is presented for a network capable of self-adaptation to achieve universal radio coverage in a simplified environment.

---

Content Distribution through Autonomic Content and Storage Management
N. Laoutaris, A. Panagakis, I. Stavrakakis (University of Athens)

Abstract.  Content Distribution has to date been addressed by a mix of centralized and uncoordinated distributed processes, such as server replication and traditional node caching mechanisms, respectively. It is an inherently distributed process that is also increasingly relying on entities that are not only increasingly distributed but also increasingly autonomous. Consequently, centralized { and typically targeting the \socially optimal" { decisions are rather unrealistic for a distributed environment of autonomic entities. Instead, a distributed management of the engaged autonomic entities, which take decisions dynamically, should be key to ecient content distribution. The latter is advocated in this paper in which two entities that are central to content distribution - specically the content and the node storage { are considered and it is discussed how their autonomic behavior drives the operation of a content distribution network. In the rst case, it is the content that manages itself by dynamically generating duplicate copies and pushing them to (seizing) the appropriate storage. In the second one, it is the node storage that is in charge, deciding on the content to be locally stored. The decisions taken by the distributed and autonomic entities may { in the extreme case { be driven by self-awareness and self-interest only, without any network state information and co-operativeness. Or, they may use (some) network information and take decisions in a more cooperative manner, despite their autonomic and self-interest-driven nature. An example is presented on the later case, showing the potential both social and individual benets.

---

A Unified Framework for the negotiation and Deployment of Network Services
S. Denazis, L. Yamamoto (Hitachi Europe)

Abstract. The Internet network technology today does not allow a sufficient degree of autonomy to express user choices, constraints and preferences in order to dynamically obtain the most suitable services. One of the goals of Autonomic Communication is to produce self-managing network elements able to provide the desired services in an automated way. In this context, we propose an architecture to automate user-provider and provider-provider relationships, by converting the Internet into an electronic market space where the commodities to be traded are network services. After an agreement has been reached via agent-based automated negotiation mechanisms, network elements must be automatically configured in order to enforce the agreed conditions. This is achieved by generating commands to programmable network elements via open interfaces. The ultimate goal is enable fully automatic installation, configuration and monitoring of protocols or service components involving multiple ownership domains, while taking into account the constraints and preferences of users and providers.

TurfNet: An Architecture for Dynamically Composable Networks
S. Schmid, L. Eggert, M. Brunner, J. Quittek (NEC Europe)

Abstract. The Internet architecture is based on design principles such as end-to-end addressing and global routeability. It suits relatively static, well-managed and flat network hierarchies. Recent years have shown, however, that the Internet is evolving beyond what the current architecture can support. The Internet architecture struggles to support increasingly conflicting requirements from groups with competing interests, such as network, content and application service providers, or end-users of fixed, mobile and ad hoc access networks.

This paper describes a new internetworking architecture, called TurfNet. It provides autonomy for individual network domains, or Turfs, through a novel inter-domain communication mechanism that does not require global network addressing or a common network protocol. By minimizing inter-domain dependencies, TurfNet provides a high degree of independence, which in turn facilitates autonomic communications. Allowing network domains to fully operate in isolation maximizes the scope of autonomic management functions. To accomplish this, TurfNet integrates the emerging concept of dynamic network composition with other recent architectural concepts such as decoupling locators from identifiers and establishing end-to-end communication across heterogeneous domains.

---

A systems architecture for sensor networks based on hardware/software co-design
A. Nisbet, S. Dobson (Manchester Metropolitan Uni., Trinity College Dublin)

Invited talk: Dynamics, information and control in physical systems
A. Fradkov, Russian Academy of Sciences

Challenges in communication research beyond the VICOM project
F. Vatalaro, G. Cortese, F. Davide, M. Leo, P. Loreti, G. Riva
(Uni of Rome "Tor Vergata", CNIT, Telecom Italia, Istituto Auxologico Italiano)

Abstract. The VICOM (Virtual Immersive COMmunications) project is a three-year project funded by the Italian Ministry of Instruction University and Research aiming at investigating innovative communication paradigms. The project represents a wide coordinated effort focused on integration of immersive and wireless technologies in view of the fourth generation of mobile communications. The main goal of the project consists of the design of a wideband system architecture for immersive services and of its validation through two distributed large test-beds. Starting from VICOM ongoing experiences some future challenges and objectives for the future situated and autonomic communications technologies are envisaged in the paper.

A Framework for self-organizing network composition
C. Kappler, P. Mendes, C. Prehofer, P. Poyhonen, D. Zhou
(Siemens AG, Deutschland and Austria, DoCoMo EuroLabs, Nokia Research)

Abstract. This paper discusses a framework for a flexible, self-organized control plane for future mobile and ubiquitous networks. The current diversification of control planes requires a manual configuration of network interworking. The problem will increase in the future, with more dynamic topologies and integration of heterogeneous networks in a ubiquitous, reactive environment. In this paper we introduce the concept of network composition, a basic, scalable and dynamic network operation to achieve autonomic control plane interworking between Ambient Networks - our approach for next generation networks. We show the architectural components of a generic control plane and its flexible interfaces. With an example on seamless mobility we illustrate how composition can simplify and improve the interworking of future networks

Semantic-based Policy Engineering for Autonomic Systems
D. Lewis, K. Feeney, K. Carey, T. Tiropanis, S. Courtenage
(Trinity College Dublin, Athens Uni. of Technology, Uni. of Westminster)

Abstract. This paper presents some important directions in the use of ontology-based semantics in achieving the vision of Autonomic Communications. We examine the requirements of Autonomic Communication with a focus on the demanding needs of ubiquitous computing environments, with an emphasis on the requirements shared with Autonomic Computing. We observe that ontologies provide a strong mechanism for addressing the heterogeneity in user task requirements, managed resources, services and context. We then present two complimentary approaches that exploit ontology-based knowledge in support of autonomic communications: service-oriented models for policy engineering and dynamic semantic queries using content-based networks. The paper concludes with a discussion of the major research challenges such approaches raise.

Self-management of Autonomic Systems: The Reputation, Quality and Credibility (RQC)
A. Garg, R. Battiti, G. Costanzi (Uni. of Trento)

E Pluribus Unum: Deduction, Abduction and Induction, the Reasoning Services for Access Control in Autonomic Communication
H. Koshutanski, F. Massacci (Uni. of Trento)

We propose a logical framework and a Web-Service based implementation for reasoning about access control for Autonomic Communication. Our model is based on interaction and exchange of requests for supplying or declining missing credentials. We identify the formal reasoning services that characterise the  problem and sketch their implementation.

A Metabolic Approach to Protocol Resilience
C. Tschudin, L. Yamamoto (Uni of Basel, Hitachi Europe)

Abstract. The goal of this research is to create robust execution circuits for communication software which can distribute over a network and which continues to provide its service despite parts of the implementation being knocked out. Like packets that can be lost (which can be recovered by the appropriate protocols) we envisage an environment where parts of a protocol's execution can be lost. The remaining implementation elements should continue to operate and be able to recover by themselves for restoring full services again.  Based on a chemical execution model, we show a few initial examples of packet

processing functions that are robust against the knock-out of any single instruction. These examples illustrate how the model can be applied to implement resilient communication protocols, to which we add regulatory signals that can be used to steer the protocols' code basis.

Putting meaning into the network: some semantic issues for the design of autonomic communication systems
Simon Dobson (Trinity College Dublin)

Abstract. Traditional network abstractions follow a layered model in which a sub-system interacts with other network components through very narrow interfaces. We content that this model is weak both in providing clear models of end-to-end properties and allowing adaptation to the more abstract properties of systems. We propose instead a graph-centric, contextual abstract model in which sub-systems can relate to other components at a wide range of semantic levels. We explore the implications such a model would have for network technology, applications and users, and identify some of the major research challenges it poses.

A Role for Contextualised Knowledge in Autonomic Communications
R. Sterritt, M. Mulvenna, Agnieszka Lawrynowicz (Uni. of Ulster, Poznan Uni.)

Abstract. The conceptual architecture of autonomic communications requires a knowledge layer to offer effective, transparent and high level self-management capabilities. This knowledge plane can utilise the behaviour of autonomic communication regimes to monitor and intervene at many differing levels of network granularity. This paper introduces autonomic computing and autonomic communication, before outlining the role of knowledge in autonomic networks. Some research issues, in particular the concept of dynamic context as a method to acquire knowledge dynamically that helps to facilitate a successful realisation of the knowledge plane are explored and discussed.

---

Towards Adaptable Ad Hoc Networks: The routing Experience
C. Santivanez, I. Stavrakakis (BBN Technologies, Uni. of Athens)

to the deployment of a plethora of wireless networks for diverse applications: from rescue missions to military communications, from collaborative computing and sensor networks to web browsing and e-mail exchange to real time voice and video communications. Each with different constraints and requirements. And, for each type of application there is also a high degree of variability in the networking context: from a low mobile network of a few nodes to a highly mobile network with thousands of nodes.

This high degree of variability in the networking environment calls for a new design paradigm where network elements (nodes) should be able to adapt to totally different scenarios, engaging in a different behavior depending on the situation. Thus, next generation networks should be able to learn their environment/context and adapt their behavior accordingly inorder to achieve their goals. In this paper we introduce some key mechanisms required to enable broad adaptability. Although these mechanisms are general and common to a large variety of tasks/services (e.g. service discovery, location management, cooperative computing, clustering, etc.) we will discuss them in the context of the routing service, leveraging our past experience on the area. This will allow us to ground the discussion in concrete terms and the reader to better visualize the concepts.