[ntp:questions] NTP vs RADclock?

Terje Mathisen "terje.mathisen at tmsw.no" at ntp.org
Wed Jun 6 07:37:37 UTC 2012


unruh wrote:
> But then there is the comments about one way timing ( the server sends
> out the signals, and then sometime  the computers send out two way
> signals so as to estimate the trip time. What is the purpose of the one
> way signals? You certainly have zero estimate of the delay on that that
> particular signal.

This is obviously from their experimental support of the Ethernet hw 
time sync protocol, since that protocol is based on having _many_ time 
sync packets, usually 1 to 16 pr second.

These packets can then be broadcast on the local LAN, supporting lots of 
lab gear simultaneously, while each piece of client gear sends a two-way 
request every few seconds in order to determine the current propagation 
delay.

This protocol can also use single or dual-packet operation, where the 
dual-packet setup uses the second packet to transmit the actual sending 
time of the first packet in the pair.

Cell tower base stations use this protocol to achieve low ppb frequency 
offsets, in order to handle tower-to-tower handover at speed. (You need 
~20 PPB frequency precision to perform a handover of a cell phone that's 
on a high-speed (250-300 km/h) train.)

Terje

-- 
- <Terje.Mathisen at tmsw.no>
"almost all programming can be viewed as an exercise in caching"



More information about the questions mailing list