Autonomic Computing defines computational systems that are able to manage themselves with none or minimal human intervention. The concepts related to autonomic computing in computer networks leads to the development of a new concept: autonomic networks. This kind of network is able to perform self-management of its elements and data communication links. Smart environments represent the next evolutionary development step in building, utilities, industrial, home, shipboard, and transportation systems automation. In those scenarios, wireless transmission is complex and often problematic, where a number of conditions such as multipath interference, transmission collisions, and obstructions adversely impact the reliability, scalability and data throughput of high bandwidth communications. This tutorial will present a perspective on autonomic wireless networks to be employed in the scenarios described above. The goal is to present the main characteristics of protocols, applications and services for those scenarios and how autonomic wireless networks can help in the design process.

1.    Introduction

Autonomic Computing defines computational systems that are able to manage themselves with none or minimal human intervention. It is a new vision that promises to enable the computing to deliver much more automation than the sum of its individually self-managed parts. The implementation of the concepts related to autonomic computing in computer networks leads to the development of a new concept: autonomic networks. This kind of network is able to perform self-management of its elements and data communication links. The goal is to have network services and functions performed without the involvement of a human manager and in a transparent way to its users. Besides, the network should be able to learn from the actions performed by its elements and the analysis of all acquired results. The automatic execution of tasks and the learning possibility characterize the autonomic aspect of this kind of network.

Smart environments represent the next evolutionary development step in building, utilities, industrial, home, shipboard, and transportation systems automation. In those scenarios, wireless transmission is complex and often problematic, where a number of conditions such as multipath interference, transmission collisions, and obstructions adversely impact the reliability, scalability and data throughput of high bandwidth communications. In addition, the vast majority of applications and services for those scenarios were initially designed to run on wired networks, and thus require considerable technology adaptation for wireless implementation. Furthermore, in order to ensure widespread adoption, the wireless modules deployed must fit into the cost constraints of their environments. To cope with those design challenges, in the future, we can expect to have those networks to work in an autonomic way.

Since an autonomic wireless network can exist at many levels (see Figure 1) and using different wireless and wired communication technologies (see Figure 2), an autonomic system will need detailed knowledge of its components, current status, ultimate capacity, and all connections with other systems to govern itself. It will need to know the extent of its owned resources, those it can borrow or lend, and those that can be shared or should be isolated.

To really benefit people in smart environments, autonomic computing will need to deliver measurable advantages and opportunities by improving interaction with the underlying systems and the quality of the information they provide. Thus, people in those environments will focus entirely on the information services they want and will not be aware of the systems providing them.

This tutorial will present a perspective on autonomic wireless networks to be employed in the scenarios described above, i.e., building, utilities, industrial, home, shipboard, and transportation systems automation. The goal is to

 

Figure 1. Wireless networks at different levels

Figure 2. Wireless and wired communication technologies

 

present the main characteristics of protocols, applications and services for those scenarios and how autonomic wireless networks can help in the design process.

This tutorial will cover the following topics:

• Introduction to autonomic computing and autonomic wireless networks
  Presents an overview of these concepts.
• Autonomic Wireless Networks
  Describes the main autonomic features of different wireless networks.
• Smart environments
  Identifies the main characteristics of this class of ap- plications.
• Integrating smart environments and autonomic wireless networks
  Discusses how these applications can be supported by autonomic wireless networks. It presents how the different autonomic aspects must be considered together when designing a wireless network for smart environments.
• Case studies
  Presents some case studies of smart environments and autonomic wireless networks. Research agenda for autonomic wireless networks Presents some research topics related to autonomic wireless networks.
• Conclusions
  Discusses how a proper autonomic wireless network solution can benefit the productivity of smart applications.

2.    Authors' Biography

Antonio Loureiro is an Associate Professor of Computer Science at the Federal University of Minas Gerais (UFMG), Brazil. Professor Loureiro holds a PhD in Computer Science from the University of British Columbia, Canada, 1995. His main research areas are mobile computing, computer networks and distributed systems. In the last 10 years he has published over 70 papers in international conferences and journals. In August 2004, he presented a plenary session at the International Workshop on Telecommunications (IWT), and, in May 2005, the authors presented the tutorial entitled Management of Wireless Sensor Networks at the 9th IFIP/IEEE International Symposium on Integrated Network Management (presented again at IWT'07). Currently, the two authors are preparing a book entitled Management of Autonomic Wireless Sensor Networks to be published by an international publisher. He was the TPC Chair for LANOMS 2001 (Latin American Network Operations and Management Symposium, sponsored by IEEE Communica- tions Society) and for the 2005 ACM Workshop on Wireless Multimedia Networking and Performance Modeling, and will be for LAACS 2007 (Latin American Autonomic Computing Symposium).

Linnyer Ruiz is an Associate Professor of Computer Science at the State University of Londrina (UEL), Brazil. Her areas of interest and research include computer networks, telecommunications and computer network management, and wireless sensor networks. She received a PhD degree in Computer Science from the Federal University of Minas Gerais, an MSc degree in Electrical Engineering and Indus- trial Information from the Federal Center of Technological Education of Paran (CEFETPR), Brazil, in 1996, and a BSc degree in Computer Engineering from PUCPR. She held a post-doctoral position at the UFMG during 2004. She is an expert in telecommunications management network (TMN). Since 1993 she has participated and coordinated research groups on TMN. Currently, she is coordinating the Wireless Sensor Network group in the Department Computer Science of UEL and is leading the Manna research team.