Autonomic Cooperative Networking

Introduction

Since the number of devices interconnected worldwide is growing drastically, the question of the orchestration of the durability-orientated operation of networked systems becomes critically substantial. In fact, durability understood in context of network resilience translates into the provision of the key features of reliability, availability, safety, confidentiality, integrity, and maintainability. To this end, it is advocated for the integration of networking with the rationale behind autonomic computing in terms of self-configuration, self-optimisation, self-healing, and self-protection. This chapter goes one step further as it presents the concept of Autonomic Cooperative Networking (ACN), which not only capitalises on the above-mentioned concept but also incorporates the novel notion of Autonomic Cooperative Behaviour (ACB). ACB stems from and is being triggered by the idea of distributed cooperative transmission, aiming to enable collaboration among devices, enhanced with the integration of the relevant network layer routines. This way the devices are expected to share their computational capabilities and memory to perform joint data processing for the benefit of meeting the global performance indicators through increased resilience.

Obviously, as such a system grows, it appears necessary, from the local perspective, to instantiate cross-layer integration of certain aspects with the use of a network protocol in order to limit any unnecessary control overhead resulting from the exchange of data among the cooperating devices. Yet, as today’s distributed systems are becoming more and more complex, the transition from the local to a more global perspective additionally requires the incorporation of an autonomic overlay so the routing-enabled distributed cooperative system may self-manage. Going further, the paradigm of autonomic system design assumes that such a networked system should follow the operating principles of the human autonomic nervous system and, thus, be able to self-configure and, then, self-manage without any external intervention. Autonomic design has its own area of applicability and it should not be confused with autonomous or automated systems. While autonomous may, on the one hand, pertain to being stand-alone, and, on the other hand, to being cognitive, automated is mostly done by means of scripting. The difference in meaning does not exclude, however, certain dose of synergy so that autonomics, understood as self-management, could be supported through the inclusion of autonomous and automated routines.

The idea of Autonomic Cooperative Networking is expected to become one of the key emerging technologies behind the concept of Autonomic Future Internet. At this time it is undergoing the pre-standardisation path within the Industry Specification Group (ISG) on Autonomic network engineering for the self-managing Future Internet (AFI), functioning under the auspices of the European Telecommunications Standards Institute (ETSI). In particular, it is applicable to ad-hoc and mesh networks and the author of this chapter is serving as a Rapporteur of ETSI AFI on the Autonomicity-Enabled Ad-hoc and Mesh Network Architecture which is preparing a Group Specification (GS) in this area (Wódczak, Meriem, Chaparadza, Quinn, Lee, Ciavaglia, Tsagkaris, Szott, Zafeiropoulos, Radier, Kielthy, Liakopoulos, Kousaridas & Duault, 2011). To account for the above, the chapter is organised as follows. First, the concept of cooperative transmission is described through the introduction of the idea of Virtual Antenna Arrays (VAA) in order to pave the ground for further analysis. Following, it is complemented with the incorporation of the relevant routines of the network layer with special emphasis on the Optimised Link State Routing Protocol (OLSR) and its Multi-Point Relay (MPR) selection heuristics. Finally, the routing enabled cooperation is translated into Autonomic Cooperative Behaviour (ACB) being integrated directly into the Generic Autonomic Network Architecture (GANA).