Vpn, Voip, Wi-Fi

[dsp HISTORY]
Robert M. Gray

The 1974 Origins of VoIP

H

istory often unfolds in unforeseen ways. The story told here provides an example in which developments in digital signal processing (DSP)—speech coding, in particular—had a profound impact on the early development of the ARPANET, the ancestor of the Internet. I discovered the story while I was preparing a historical talk for the Special Workshop in Maui (SWIM). SWIM was intended to bring together a collection of pioneers in digital speech, especially in the development of linear predictive coding (LPC). Unfortunately for the workshop (but fortunately for me), two of the pioneers, John D. Markel and my brother A.H. “Steen” Gray, Jr., were not able to attend, so I was invited to tell the tale of their contributions to the early development of LPC. My connections were both filial and technical: I had begun working on LPC in the mid-1970s with John and Steen, so I knew many of the contributions and players. As I conducted interviews and tracked down sources for my Maui talk (see the link at the end of the article for more detail, references, and related material), the subject matter expanded from speech and DSP to the first successful attempts to transmit real-time packet speech. The story shows how packet speech, recently rediscovered and made popular as voice over IP (VoIP), was first successfully demonstrated in 1974 on the ARPANET and how the Internet protocol (IP) emerged largely as a result of that effort. HOW IT ALL BEGAN Two threads of the story began in 1966. In December 1966, Shuzo Saito of NTT and a young Nagoya University doctoral

EDITORS’ INTRODUCTION Robert M. Gray was born in November 1943 at North Island Naval Air Station, San Diego, California. He obtained his B.S. and M.S. degrees (1966) from the Massachusetts Institute of Technology and a Ph.D. degree (1969) from the University of Southern California. Since 1969, he has been with Stanford University, where he is the Lucent...