Adaptive Control of Sensor Networks

In recent years many algorithms and protocols for applications in wireless sensor networks (WSN) have been introduced. These include, e.g, solutions for routing and event notifications. Common among them is the need to adjust the basic operation to particular operating conditions by means of changing algorithmic parameters. In most applications, parameters have to be set carefully before nodes are deployed to a particular environment. But what happens to the system performance, if the operating conditions change to unforeseen situations at runtime? In this paper, we present the Organic Network Control (ONC) system and its application to WSNs. ONC is a system for adapting network protocols in response to environmental changes at runtime. Being generic in nature, ONC regards existing protocols as black box systems with an interface to changeable protocol parameters. ONC detects environmental changes locally at each node and applies changes to the protocol parameters by means of lightweight machine learning techniques. More complex exploration of possible parameters is transferred to powerful nodes, such as sink nodes. As an example we show how ONC can be applied to an exemplary WSN protocol for event detection and how performance in the ONC controlled system increases over fixed settings of the protocol.

[1]  Jörg Hähner,et al.  Towards an Organic Network Control System , 2009, ATC.

[2]  Hock-Beng Lim,et al.  An Adaptive Distributed Resource Allocation Scheme for Sensor Networks , 2006, MSN.

[3]  Michele Zorzi,et al.  Statistically assisted routing algorithms (SARA) for hop count based forwarding in wireless sensor networks , 2008, Wirel. Networks.

[4]  Jörg Hähner,et al.  Dynamic Control of Network Protocols - A New Vision for Future Self-organising Networks , 2009, ICINCO-ICSO.

[5]  Stewart W. Wilson.  Classifier Fitness Based on Accuracy , 1995, Evolutionary Computation.

[6]  Jessica Andrea Carballido,et al.  CGD-GA: A graph-based genetic algorithm for sensor network design , 2007, Inf. Sci..

[7]  Ramesh Govindan,et al.  Wireless sensor networks , 2003, Comput. Networks.

[8]  M. Marks,et al.  Two-Phase Stochastic Optimization to Sensor Network Localization , 2007, 2007 International Conference on Sensor Technologies and Applications (SENSORCOMM 2007).

[9]  N. Georganopoulos,et al.  A Framework for Dynamic Link and Network Layer Protocol Optimisation , 2007, 2007 16th IST Mobile and Wireless Communications Summit.

[10]  Hartmut Schmeck,et al.  A Characterization of Key Properties of Environment-Mediated Multiagent Systems , 2007, EEMMAS.

[11]  Nikolay Mehandjiev,et al.  Engineering Environment-Mediated Multi-Agent Systems , 2008 .

[12]  Anna Förster.  Machine Learning Techniques Applied to Wireless Ad-Hoc Networks: Guide and Survey , 2007 .

[13]  Thorsten Schöler,et al.  Design, Implementation and Validation of a Generic and Reconfigurable Protocol Stack Framework for Mobile Terminals , 2004, ICDCS Workshops.

[14]  Michael L. Littman,et al.  Packet Routing in Dynamically Changing Networks: A Reinforcement Learning Approach , 1993, NIPS.

[15]  Hartmut Schmeck,et al.  Towards a generic observer/controller architecture for Organic Computing , 2006, GI Jahrestagung.

[16]  Jörg Hähner,et al.  Dynamic Control of Mobile Ad-hoc Networks - Network Protocol Parameter Adaptation using Organic Network Control , 2010, ICINCO.

[17]  P. Stone,et al.  Towards autonomic computing: adaptive network routing and scheduling , 2004 .

[18]  Michael J. North,et al.  Experiences creating three implementations of the repast agent modeling toolkit , 2006, TOMC.

[19]  Ganesh K. Venayagamoorthy,et al.  Neural network based secure media access control protocol for wireless sensor networks , 2009, 2009 International Joint Conference on Neural Networks.

[20]  Antónia Lopes,et al.  Appia to R-Appia: Refactoring a Protocol Composition Framework for Dynamic Reconfiguration , 2007 .