Biologically Inspired Networking and Sensing: Algorithms and Architectures
Book Citation Index

Biologically Inspired Networking and Sensing: Algorithms and Architectures

Pietro Lio (University of Cambridge, UK) and Dinesh Verma (IBM, USA)
Indexed In: SCOPUS View 2 More Indices
Release Date: August, 2011|Copyright: © 2012 |Pages: 312
ISBN13: 9781613500927|ISBN10: 1613500920|EISBN13: 9781613500934|DOI: 10.4018/978-1-61350-092-7

Description

Despite their widespread impact, computer networks that provide the foundation for the World Wide Web and Internet have many limitations. These networks are vulnerable to security threats, break easily, and have a limited ability to respond to changing conditions. Recent research on overcoming these limitations has used biological systems for inspiration, resulting in the development of biologically-inspired computer networks. These networks are designed and developed using principles that are commonly found in natural and biological systems.

Biologically Inspired Networking and Sensing: Algorithms and Architectures offers current perspectives and trends in biologically-inspired networking, exploring various approaches aimed at improving network paradigms. Research contained within this compendium of papers and surveys introduces studies in the fields of communication networks, performance modeling, and distributed computing, as well as new advances in networking.

Topics Covered

The many academic areas covered in this publication include, but are not limited to:

  • Autonomic Data Dissemination in Ad Hoc Wireless Networks
  • Autonomously Evolving Communication Protocols
  • Biologically Inspired Congestion Control Mechanisms
  • Biologically Inspired Routing Protocols
  • Dendritic Cell Algorithm for Intrusion Detection
  • Insect Built Networks
  • Lotka Volterra Competition Model
  • MANET Routing
  • Networking Inspired by Cell Communication Mechanisms
  • Neural Networks in Cognitive Science

Table of Contents and List of Contributors

Search this Book:
Reset

Preface

Computer communication networks have transformed human civilization, and enabled information to be shared across the globe at the speed of a mouse-click. They have transformed the way society functions, and their effects can be seen in all aspects of our life. This transformation can truly be called a miracle.

In spite of their far-reaching impact, the computer networks that provide the foundation of the World Wide Web and the Internet have many limitations. The networks were not designed to accommodate mobile users, they are extremely vulnerable to security threats, they break relatively easily, requiring extensive manual labor to resolve many of these disruptions, and have very limited ability to respond to changing conditions like huge swings in their workloads.    

Researchers in the networking area are continuously striving to find ways to improve the attributes of computer communication networks and find ways to address the limitations. These new explorations are gradually helping to address the weaknesses of the network infrastructure. The investigations to improve the network include incremental improvements to the extant protocols and systems, as well as fundamentally different ways to looking at the networks.     

Some of the researchers exploring a fundamentally different way to resolve the limitations of modern day networks have been looking towards biological systems for inspiration. This has results in an exciting new area of biologically inspired computer networks. Such networks are designed and developed using principles that are commonly found in natural and biological systems.  

This book provides a current snapshot of some of those research activities. By bringing together the research activities from a variety of institutes around the globe, we hope to provide a good coverage of the various approaches that are being explored to improve the networking paradigms.  

Comparing Biological and Computer Networks

The impetus to draw inspiration from biological networks comes from the fundamental observation that biological systems just do a better job at many functions than the best designed electronic computers and computer networks.

Perhaps the most obvious example of a domain where biological networks have an advantage is the human immune system. The immune system is able to react to attacks from a variety of viruses and bacteria, including those that it may have never encountered before. It is able to identify the intruders, and take action against them in a very effective manner. Even though the number and varieties of the viruses and bacteria keep on multiplying due to mutations and natural evolution, the immune system is able to manage these variations with relative ease. In stark contrast, computer networks have a very difficult time identifying malware, intrusions, and other attacks, and struggle to cope up with the new security threats that keep on surfacing all the time. In some instances, the security mechanisms become a nuisance rather than a useful feature.

Another unique area where biological systems have an advantage is in their ability to adjust themselves in the face of a changing external environment. When the external temperature is hot, the body sweats to cool itself down, and when the external temperature is low, the body shivers to restore and gain some heat. Not counting some extreme situations, the human body (and many other biological systems) is able to adapt to an amazing degree. On the other hand, the computer networks of today are rarely able to cope with a dynamically changing workload, and their ability to deal with extreme external changes is very limited.   

There are some aspects of networking in which current computer networks outperform biological networks, e.g. the fidelity and speed at which information can be communicated in electronic networks is much more reliable and higher-speed than biological networks. The goal of biologically inspired networks is not to belittle those advantages, but to explore those aspects that can be made better by drawing inspiration from biology.

Some of the recent advances made in improving the design of networks using biologically inspired paradigms are compiled in this book. The next section explains the structure of the book and the content of the different chapters.

Structure of the Book

This book consists of thirteen chapters which provide a good overview of the current state of the art in biologically inspired computer networks. For organizational purposes, the work is divided into three different categories.

The first category consists of chapters that are proposing new architectures for computer networks that are based on biologically inspired techniques. These chapters include description of work that is trying to develop a new paradigm for computer communications.

The first chapter “A Networking Paradigm Inspired by Cell Communication Mechanisms” describes molecular communications - a new paradigm for networking in which information is encoded to and decoded from molecules, rather than electrons or electromagnetic waves. This paradigm is being used to explore new models for nano-networking and in synthetic biology. The chapter provides an overview of the current state of the art, and the models used in the current state of the art for molecular communications.

The second chapter, "Organic Network Control: Turning Standard Protocols into Evolving Systems" presents a new architecture that allows for automatic adaptation of protocol parameters in dynamically changing environment. It is based on an observer-controller paradigm and uses evolutionary algorithms for adaptation. The chapter provides some examples where such protocols can be used, and also surveys the current state of the art in the area.  

The third chapter "Robust Network Services with Distributed Code Rewriting" looks at a way to design distributed software systems that are based on continuous replication of a code base. They use the concept of quines – a piece of software that prints its own code, and leverage quines to create a system that dynamically rewrites itself in a regulated manner simultaneously exploiting competition as well as cooperation.  

The fourth chapter “Neural Networks in Cognitive Science – An introduction" provides an overview of an architecture for cognitive modeling that leverages neural networks. It is an instance of biologically inspired neural networks being used in various domains and applications.

The fifth and final chapter in this section, "The Dendritic Cell Algorithm for Intrusion Detection" is a new architecture to perform the security functions in computer networks. It uses an algorithm modelled after the body's immunity functions, and provides a new approach to detect anomalies in network traffic.

The next section of the book consists of chapters that are focused on resource optimization in computer networks. Any computer network operates under an environment of constrained resources such as bandwidth, power, and computation capacity at nodes. In different types of networks, different resources are the bottlenecks which need to be optimized. In the context of military or satellite networks, bandwidth is usually the bottleneck. In the context of commercial wireless sensor networks, battery power becomes the most constrained resource, while for high speed optical networks, the computation and switching capacity at intervening electronics is the bottleneck resource. Therefore, new approaches that allow optimal use of scarce resources are valuable to explore in all types of computer networks.  

The first chapter in this section, "TCP Symbiosis: Bio-inspired congestion control mechanism for TCP" looks at ways to improve the congestion control scheme used in the widely deployed TCP protocol using biologically inspired techniques. The authors use concepts borrowed from biophysics, such as the Lotka-Volterra competition model, to improve the congestion control behavior, and show that the new biologically inspired approach has better stability and scalability characteristics than the prevailing congestion control schemes.  

The second chapter in this section, "From local growth to global optimization in insect built networks" discusses how insect colonies optimize themselves in a completely distributed and decentralized manner. They provide an in-depth analysis of the local behaviors of insects that leads to an eventual overall optimization of the global network in the colony.

The next chapter, “Network Energy driven Wireless Sensor Networks” examines the subject of managing energy in wireless sensor networks. The approach proposed is that of scavenging energy available from unwanted radio frequency waves, a model inspired by the behavior of emperor penguins. In networks where energy is at a premium, such harvesting approaches can provide significant value.

The final chapter in this section, “Congestion Control in Wireless Sensor Networks based on the Lotka Volterra Competition Model” provides an alternative approach to congestion control in wireless sensor networks. The Lotka Volterra model is a mathematical model that characterizes the population of different species in an ecosystem. The model, when applied to the task of managing bandwidth resources and congestion during communication, provides an interesting paradigm to manage scarce resources.

The third section of this book looks at the task of routing in computer networks. Routing is the process by which packets emanating from a source in the computer network are eventually delivered to their destination for unicast communication, or to multiple destinations in the case of multicast communications. The routing protocols for traditional networks like the Internet have become standardized and well-established, specially for the paradigm of unicast or point-to-point communications.  There is still a lot of room for routing innovation in other types of communication paradigms such as multicast or unicast. Furthermore, as new types of computer networks emerge, e.g. mobile ad-hoc networks, disruption tolerant networks, or nano-scale molecular networks, each with their own specific idiosyncrasies, new types of routing protocols need to be investigated for them.  

The first chapter in this section, “Autonomously Evolving Communication Protocols: The Case of Multi-Hop Broadcast” looks at the routing needs of broadcast networks which are relevant in tactical military environments, wild-life monitoring, and other instances of mobile ad-hoc networking. They propose an alternative approach for routing using autonomous machine intelligence built upon on-line evolutionary approaches such as natural selection and genetic programming. Creating a genetic programming language and a selection mechanism for multi-hop broadcast protocols allows them to create a system that outperforms traditional networks under some conditions.

The next chapter, “Application of Genetic Algorithms for Optimization of Anycast Routing in Delay and Disruption Tolerant Networks” provides another algorithm based on genetic programming, with the difference being the type of networks that are targeted. This chapter looks at the routing problem in disruption tolerant networks.

The third chapter in this section, “Data Highways: An Activator–Inhibitor–based Approach for Autonomic Data Dissemination in Ad Hoc Wireless Networks” uses the paradigm used in cell morphogenesis to create paths for information dissemination in ad-hoc networks. The concept provides a completely decentralized approach to establishing paths that lead to data sinks, a peculiar behavior that is commonly found in ad-hoc sensor networks.  

The last chapter in this section, “Scented Node Protocol for MANET Routing” provides an approach based on modified ant colony algorithms to create effective routes in mobile ad-hoc networks.

Taken together, the thirteen chapters in this book provide a current snap-shot of network research drawing its inspiration from biological systems.

Who is the Book For?

This book is intended for researchers in the academia, industry, and governments who want to understand the issues in networking, and obtain an overview of the recent advances in the field of networking that are inspired by biological systems. This book will introduce some new advances in networking. Researchers in the field of communication networks, performance modeling and distributed computing will find the chapters in this book to be of particular relevance.

Who is the Book Not For?

This book is not intended for a biologist or a researcher who is new to the principles of computer communications network. It does not provide a tutorial or introduction to the design of current computer networks, a topic that can take several books on its own. It also does not deal with incremental advances to existing deployed networks. Most of the ideas covered in this book will require a radical change in the networking infrastructure to implement.

This book is a compendium of research papers and surveys. As such, it is not a comprehensive introduction to the subject of biologically inspired communication networks. It is instead targeted for researchers who already have some understanding of the area and are looking for focused, detailed research papers on specific aspects of it.   

Author(s)/Editor(s) Biography

Pietro Lio is a Senior Lecturer in the Computer Laboratory which is the department of Computer Science of the University of Cambridge and a member of the Artificial Intelligence group of the Computer Laboratory. He has an interdisciplinary approach to research and teaching and holds a PhD in Complex Systems and Non Linear Dynamics (School of Informatics, dept of Engineering of the University of Firenze, Italy) and a PhD in (Theoretical) Genetics (University of Pavia, Italy). His current research interest is the investigation of biomedical processes employing a combination of techniques, ranging from machine learning to deterministic and stochastic models.
Dinesh Verma is a researcher and department group manager in the IT & Wireless Convergence area at IBM T J Watson Research Center, Hawthorne, New York. He received his doctorate in Computer Networking from University of California Berkeley in 1992, the Bachelor’s in Computer Science from Indian Institute of Technology, Kanpur, India in 1987, and a Master’s in Management of Technology from Polytechnic University, Brooklyn, NY in 1998. He holds over thirty US patents related to computer networks, and has authored over sixty papers and eight books in the area. He is the program manager for the US/UK International Technology Alliance in Network Sciences. He is a fellow of the IEEE, and has served in various program committees and technical committees. His research interests include topics in wireless networks, network management, distributed computing, and autonomic systems.

Indices

Editorial Board

  • Gregory Cirincione,U.S. Army Research Labs, USA
  • David Watson, IBM United Kingdom Ltd, UK
  • Ananthram Swami, U.S. Army Research Labs, UK
  • Don Towsley, University of Massachusetts, USA