Hi Jason,
if I understand your problem correctly, this is a bit like the problem that, say,
independent user equipment (UE) devices (==mobile phones) need to solve in a cellular
network: they all need to align their clocks to the network-provided time, based on some
synchronization signal.
You *can* do that, but I don't think it has much to do what you mention here. Let me
sketch what I think you should consider:
1. let's assume for the time being that your two UEs and the network have the
same notion
of *frequency*, so f₁ = f₂, and because they're USRP, that also means their
sampling
rates are identical.
2. However, both UE have totally independent notions of time, because both
started
running at different times (t₁ ≠ t₂) (i.e., relative to network time, UE1
and UE2 had
their zeroth sample late by t₁ and t₂, respectively);
also, the network time t₀ is also different (t₁ ≠ t₀ ≠ t₂; wlog t₀ = 0)
However, since the sampling rates r are identical, a time interval T means
the same
exact time difference to all three.
3. Now, on UE1, a synchronization burst is observed. This gives us the
information that
at sample n₁, which, in terms of time for UE happened at sample r·t₁+x₁=n₁;
it
however, knows from the type or content of the burst that it happend at
n₀=r·t₀+y on
the network side.
4. UE1 now calculates the difference between the known network time of the
burst and the
observed time: r·(t₀-t₁)=y-x₁; since r, y and x₁ are know to UE1,this gives
the time
by which the UE lags t₀. That means we need to "jump" the time by -(t₀-t₁).
5. UE2 does the same calculation, but for-(t₀-t₁.
6. Both USPRs use a fixed time m₁, m₂ *by their own* counting, to set the
internal time
to m₁-(t₀-t₁), and m₂-(t₀-t₁), respectively.
There's ways to implement 6.: I can't remember whether the X310 in absence of an external
PPS source synthesizes an internal PPS every (master clock rate) clock ticks. If that's
the case: just loop till get_time_last_pps() changes, then add 1-(t₀-t₁) to it, and set
the result set_time_next_pps.
If that's not the case, you can either modify the FPGA image to make that happen, or just
use the GPIO API, which you can use with timed commands, and a cable from the front panel
GPIO back to the PPS input, to emulate the same.
Best regards,
Marcus M
DISCLAIMER: Any attached Code is provided As Is. It has not been tested or
validated as a product, for use in a deployed application or system, or for use
in hazardous environments. You assume all risks for use of the Code. Use of the
Code is subject to terms of the licenses to the UHD or RFNoC code with which
the Code is used. Standard licenses to UHD and RFNoC can be found at
https://www.ettus.com/sdr-software/licenses/.
NI will only perform services based on its understanding and condition that the
goods or services (i) are not for the use in the production or development of
any item produced, purchased, or ordered by any entity with a footnote 1
designation in the license requirement column of Supplement No. 4 to Part 744,
U.S. Export Administration Regulations and (ii) such a company is not a party
to the transaction. If our understanding is incorrect, please notify us
immediately because a specific authorization may be required from the U.S.
Commerce Department before the transaction may proceed further.
On 08.12.21 21:14, Jason Merlo wrote:
Hi All,
I’m currently working to synchronize multiple X310’s clocks without a PPS input, however
right now the best method I can find to update the clock from a host PC (using the C++
API) is to query the current time from the USRP device (using
usrp::multi_usrp::get_time_now), add a time delta to the current time, then send back
the new clock time to the USRP device (using usrp::multi_usrp::set_time_now).
Unfortunately, this method introduces large timing errors due to the nondeterministic
nature of packet processing on both he CPU and network stack.
I’m wondering if anyone knows of any other techniques for an "in-place" time update.
I.e., is there a method for the host PC to send a time delta to the USRP which would be
added to the clock register in a single operation?
I see there are other get/set_time_now functions in the rfnoc::mb_control and
rfnoc::radio_control classes, but I’m not sure if these would allow me to accomplish
this using only the C++ API. I can’t seem to find much documentation on this aside from
the examples in the “uhd/host/examples/rfnoc*” folder.
If it’s not possible to accomplish this using a purely C++ approach, is it possible to
do this through a custom RFNoC block? I don’t have experience with RFNoC at the moment
and I’m not sure if that register is exposed to user blocks, or if so, if the register
update would be deterministic in time, but if there’s motivation I would be willing go
down the RFNoC path.
Thanks in advance,
Jason
_______________________________________________
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com
_______________________________________________
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com
