The Sixth International Symposium on Wireless Communication Systems 2009 (ISWCS’09)
Siena-Tuscany, University of Siena, Italy, September 7–10, 2009
Technical Co-Sponsorship by IEEE Communications Society
For information: iswcs09@dii.unisi.it
Plenary Talks
"Next-generation Networking Technologies
from Bell Labs"
Abstract
The telecom industry has an aspiration of moving to an Internet core for all of its services. Even cellular networks, with billions of end points, are moving to an Internet core. This poses a number of challenges, especially with ever-increasing content traffic. This talk will present these challenges and some efforts at Bell Labs to deal with them. Some of these challenges are issues with Mobile IP, increased complexity of routers, and the need to deal with increasing opex expenses. Here are some solutions Bell labs is working on to enable migration of telecom networks to an Internet core.
- Making Mobility a core part of the Internet: Mobile IP is a patch on the original IP design. Traffic sent to a mobile client is first sent to a home agent, which in turn tunnels it to the client in its current location. This design should be fixed to avoid triangulation. Bell Labs has designed a new protocol to deal with this issue. In our new approach, a mobile host gets an address locally (using a lightweight DHCP). Then this address, along with the host’s unique ID, gets advertised on the network.
- Router designs: The networking community is looking at approaches to open up routers so that third parties can add new features. Bell Labs has developed an approach called softrouter in which routers are disaggregated into simple forwarding elements and shared control elements. This approach enables the easy addition of new functions to the IP networks.
- Self-management: Opex is beginning to dominate the total cost of ownership of networks, resulting in a high cost of service to the end-user. This is getting worse with the deployment of femto cells in homes and increasingly complex services. To change this trend, we need extensive automation of the deployment, configuration and optimization of networks. Scalability requires decentralized solutions where discovery and network integration tasks are performed with locally available information under local control. This is in contrast to the mostly centralized network management systems used today. Bell Labs has an extensive program in this area. This research uses various mathematical approaches such as linear programming and genetic programming.
Biography
Krishan Sabnani is Vice President of Networking Research at Bell Labs. He manages all networking research at Bell Labs, comprising nine departments in six countries: USA, France, Germany, UK, India, and Belgium. Krishan has conceived and launched numerous systems projects in the areas of internetworking and wireless networking.
Krishan received the 2005 IEEE Eric E. Sumner Award and the 2005 IEEE W. Wallace McDowell Award. He is a Bell Labs Fellow and a Fellow of the Institute of Electrical and Electronic Engineers (IEEE) and the Association of Computing Machinery (ACM). He received the Leonard G. Abraham Prize Paper Award from the IEEE Communications Society in 1991. Krishan received the 2005 Distinguished Alumni Award from Indian Institute of Technology (IIT), New Delhi, India. He also won the 2005 Thomas Alva Patent Award from the R&D Council of New Jersey. He holds 37 patents and has published more than 70 papers.
Krishan received his B. Tech. in electrical engineering from IIT Delhi in 1975, and a Ph.D. in electrical engineering from Columbia University, New York, in 1981. He joined Bell Labs in 1981.