Bharat Doshi: Finding Problems in Things That Don't Exist

Murray Hill, N.J. (July, 1998) -- As a meticulous mathematician who oversees the work of 55 other mathematicians, computer scientists and electrical engineers, Bharat Doshi's job is simple to describe--he looks for problems in products that don't yet exist.

"I find that having a mix of engineering and mathematics is tremendously valuable in terms of the work I am doing today," says Doshi, whose department focuses on evaluating and optimizing the performance of technologies that have yet to be made.

"This is a technology company, not a mathematics research company, although mathematics does play a major role, and I can appreciate technology," he said.

Combining mathematics and engineering

He received a bachelor's degree in mechanical engineering from the Indian Institute of Technology in Bombay, and then attended Cornell University in New York, where he studied Operations Research, which essentially is applied mathematics.

Bharat Doshi

Doshi wrote his Ph.D. thesis on Markov Decision Theory. Named after the Russian mathematician who developed it, the theory is used for decision-making based on events having limited predictability. This is not unlike the work of Doshi's department. [See Working On 'Yet-to-be' Products and Technologies]

Solving real problems

After receiving his Ph.D., he taught in the statistics department of Rutgers University for about five years before coming to Bell Labs.

Doshi made the transition from academia, he says, because, "I eventually found I would like to have stimulation coming from real problems. I didn't want to create mathematics for mathematics sake. I wanted to create mathematics so it can be coupled with technology and this was the best place in the world for that. Here one can be a good mathematician, develop state-of-the-art techniques, and make sure it has value in real life."

When he started at Bell Labs in 1979, Doshi's first assignment was to work on the development of voice-messaging service. "My job was to look at the processing performance and make sure it met requirements," he said. "And," he added, "Our analysis and recommendations helped double the capacity of the system."

Designing adaptive networks

Doshi was also involved in designing a telecommunications network that could continue to operate during national emergencies.

In response to a request by the federal government, he explains, "We designed a virtual subnetwork that had its own special routing and identification protocol using special routing algorithms that can route around a region that goes down, and continue to provide service."

This system is still used during natural disasters such as floods and hurricanes.

Doshi's early work at Bell Labs also involved the development of overload controls for the 5ESS® switch, disc-drive scheduling algorithms, mathematical techniques for production scheduling, and the frame relay protocol.

Designing for tomorrow's networks

Today, Doshi's department is responsible for applying mathematical and computer modeling techniques to analyzing the performance of new technologies and products to ensure their quality. Some of these are related to wireless, Internet, ATM, SONET, and optical networking.

Besides work, Doshi enjoys traveling, photography and reading. As for where he finds time for all that, "I sleep very little," he says.