[ntp:questions] does this make sense?

Ruslan Nabioullin rnabioullin at gmail.com
Mon Aug 13 21:29:23 UTC 2018

On Thu, Apr 5, 2018 at 8:25 AM, Thomas Laus <lausts at acm.org> wrote:
> On 2018-04-04, Maria Iano <mianogannett at gmail.com> wrote:
> > I'm purchasing ntp appliances to put into three datacenters. Does it make sense to purchase two that use GPS and two that use WWVB, and configure them as peers?
> The USA Bureau of Time Standards has a link for timing receiver
> vendors:
> https://www.nist.gov/pml/time-and-frequency-division/time-services/manufacturers-time-and-frequency-receivers
> There are a few manufacturers of CDMA systems that receive the timing
> signals from the USA cellphone systems.  They have a high degree of
> accuracy and are a better choice than WWVB in most cases.

What's the point?  From what I know, cellular systems typically use
the following metrological hierarchy: 1. GPS, possibly redundant; and
2. onsite telecom-grade rubidium, possibly redundant.  Therefore,
using CDMA would partially defeat the purpose of seeking an alternate
source as a fallback/for corroboration (the only utility would be that
rubidium standard, which would be civilian and moreover telecom-grade,
and of course the GPS receiver is almost certainly civilian and
therefore be susceptible to spoofing).  Plus it's a for-profit
civilian commercial service (from which timing is extracted as a
byproduct of its operation), which will likely be decommissioned from
the entire US in a few years anyway...

WWVB should be a good fallback source to GNSS, due to the
significantly-different nature of the technology (terrestrial in lieu
of space, groundwave propagation in lieu of line-of-sight, and VLF in
lieu of L band), though I am concerned about it (and/or one's
receiver) employing civilian technology (that could be damaged or
taken out by nuclear EMPs) and the potential for *targeted* malicious
jamming/spoofing employed by adversaries (for general-coverage
malicious jamming/spoofing, obviously it would be more difficult due
to the need to erect VLF antennae and powerful transmitters, which
would be visible by authorities, one of the merits of terrestrial
hyperbolic radionavigation services, e.g., Loran).

It's quite unfortunate that WWV could be taken down (I am involved in
a number of projects which utilize WWV/CHU-derived timing for public
NTP services), though on the other hand I don't see too much of a
utility from it for T&F (it's skywave/groundwave in lieu of
groundwave, and HF in lieu of VLF) compared to VLF services (WWVB,
DCF77, et al.), due to its crude accuracy as a result of the skywave
propagation mode, apart from greater coverage during favorable
ionospheric conditions and there being WWV/CHU NTP decoders (try doing
that for WWVB...they changed the signal format in '12, thereby turning
virtually all metrological receivers into paperweights).  It's the
best GNSS alternative out there for NTP users (in North America, at
least) (just connect an HF AM receiver to the server's soundcard),
unless one resorts to purchasing a WWVB receiver (which to my
knowledge is made by only 2--3 corporations, one of which is not


Ruslan Nabioullin
Wittgenstein Laboratories
rnabioullin at gmail.com
(508) 523-8535
50 Louise Dr.
Hollis, NH 03049

More information about the questions mailing list