Community discussions

MUM Europe 2020
 
WirelessRudy
Forum Guru
Forum Guru
Topic Author
Posts: 3089
Joined: Tue Aug 08, 2006 5:54 pm
Location: Spain

Explaination of NV2 register table readings please..

Sun May 15, 2011 5:44 pm

How do I interpret the wireless tables ´TDMA Timing Offset´ readings?

I have them ranging from -1 (?) to 9
Is this ms? or µs?
How can I have a (only one station) negative one?
Is there an ideal reading?
´TDMA Timing "Offset"´ ; Where is it "Offset" from?

Can anyone (MT?) shine a light on this for full understanding?

Also:
"TDMA Tx Size"

What does it tell?
Is this ms? or µs? Or bytes? (Bits?)
What does it tell? Bigger better or worse?

To get the full benefit this forum should discuss these terms and thereafter we could update Wiki manual accordingly.
Many of us want to explore the new deep waters of NV2 but not all of us do already know how to swim it. But some probably do. Lets share the info so we all can use it to the best without drowning.... 8)
Show your appreciation of this post by giving me Karma! Thanks.

Rudy R. Puister

WISP operator based on MT routerboard & ROS.
 
Mplsguy
MikroTik Support
MikroTik Support
Posts: 226
Joined: Fri Jun 06, 2008 5:06 pm

Re: Explaination of NV2 register table readings please..

Wed May 18, 2011 12:22 am

tdma-timing-offset is proportional to distance, so there is no "ideal" reading - it is approximately two times the propagation delay. AP measures this so that it can tell clients what offset to use for their transmissions - clients then subtract this offset from their target transmission time such that propagation delay is accounted for and transmission arrives at AP when expected. You may occasionally see small negative value (like few usecs) there for close range clients because of additional unaccounted delay that may be produced in transmitter or receiver hardware that varies from chipset to chipset.

The only way how you can use this reading, except for calculating approximate distance to client (which is already done for you in distance reading), is watch if it is not changing over time for more than some usecs. Quick, significant changes in this reading usually signal that AP is not able to properly figure out the distance to client due to some problems - either in AP or client. In this situation it is advised to contact support with supout files.

tdma-tx-size is value in bytes that specifies the size of data unit whose loss can be detected (data unit over which CRC is calculated) sent by device. In general - the bigger the better, because overhead is less. On the other hand, small value in this setting can not always be considered a signal that connection is poor - if device does not have enough pending data that would enable it to use bigger data units (e.g. if you are just pinging over link), this value will not go up.
 
WirelessRudy
Forum Guru
Forum Guru
Topic Author
Posts: 3089
Joined: Tue Aug 08, 2006 5:54 pm
Location: Spain

Re: Explaination of NV2 register table readings please..

Wed May 18, 2011 3:00 am

tdma-timing-offset is proportional to distance, so there is no "ideal" reading - it is approximately two times the propagation delay. AP measures this so that it can tell clients what offset to use for their transmissions - clients then subtract this offset from their target transmission time such that propagation delay is accounted for and transmission arrives at AP when expected. You may occasionally see small negative value (like few usecs) there for close range clients because of additional unaccounted delay that may be produced in transmitter or receiver hardware that varies from chipset to chipset.

The only way how you can use this reading, except for calculating approximate distance to client (which is already done for you in distance reading), is watch if it is not changing over time for more than some usecs. Quick, significant changes in this reading usually signal that AP is not able to properly figure out the distance to client due to some problems - either in AP or client. In this situation it is advised to contact support with supout files.
Ok, thanks for the explanation. All my clients are at a distance from 10 to max 600mtr away from AP. (Distance is shown "1" for all). Still I see any possible value from -1 to 9?? Any explanation for this?
tdma-tx-size is value in bytes that specifies the size of data unit whose loss can be detected (data unit over which CRC is calculated) sent by device. In general - the bigger the better, because overhead is less. On the other hand, small value in this setting can not always be considered a signal that connection is poor - if device does not have enough pending data that would enable it to use bigger data units (e.g. if you are just pinging over link), this value will not go up.
OK, in the middle of the night I see almost all units having value 496 and only one with 1520. But this unit has no traffic while some other units have some traffic running?
Show your appreciation of this post by giving me Karma! Thanks.

Rudy R. Puister

WISP operator based on MT routerboard & ROS.
 
chadd
Member
Member
Posts: 347
Joined: Fri Dec 31, 2004 2:40 am

Re: Explaination of NV2 register table readings please..

Wed May 18, 2011 6:49 am

Mplsguy,

Do you have any tips for those of us who are having synchronization and medium-access timeouts?

Also I thought I remembered a thread on the forums that discussed some advanced/hidden settings that you could get to through the terminal that don't show up when you are looking at the /interface/wireless. Do you have any info you could share on that?
tdma-timing-offset is proportional to distance, so there is no "ideal" reading - it is approximately two times the propagation delay. AP measures this so that it can tell clients what offset to use for their transmissions - clients then subtract this offset from their target transmission time such that propagation delay is accounted for and transmission arrives at AP when expected. You may occasionally see small negative value (like few usecs) there for close range clients because of additional unaccounted delay that may be produced in transmitter or receiver hardware that varies from chipset to chipset.

The only way how you can use this reading, except for calculating approximate distance to client (which is already done for you in distance reading), is watch if it is not changing over time for more than some usecs. Quick, significant changes in this reading usually signal that AP is not able to properly figure out the distance to client due to some problems - either in AP or client. In this situation it is advised to contact support with supout files.

tdma-tx-size is value in bytes that specifies the size of data unit whose loss can be detected (data unit over which CRC is calculated) sent by device. In general - the bigger the better, because overhead is less. On the other hand, small value in this setting can not always be considered a signal that connection is poor - if device does not have enough pending data that would enable it to use bigger data units (e.g. if you are just pinging over link), this value will not go up.
 
User avatar
mramos
Member Candidate
Member Candidate
Posts: 230
Joined: Sun Nov 23, 2008 1:05 am
Location: S. B do Campo - SP - Brazil

Re: Explaination of NV2 register table readings please..

Wed May 18, 2011 4:27 pm

The only way how you can use this reading, except for calculating approximate distance to client (which is already done for you in distance reading), is watch if it is not changing over time for more than some usecs. Quick, significant changes in this reading usually signal that AP is not able to properly figure out the distance to client due to some problems - either in AP or client. In this situation it is advised to contact support with supout files.
I've been trying NV2 on 900MHz (1AP, 2 clients, 5MHz channel, XR9 based) and one of them locks at 4Km (real distance = 1.7Km), no problem at all. This link is LOS but some trees close to AP & client. Still unsuported X86, will be RB433 soon.

The second client have trees at AP side and a hill at the middle of the path (NLOS so). Real distance 2.4Km. Measured distance swings from 5Km to 11Km.

Seems stable with tdma period > 6ms. Less than that and the 5/11Km client (RB433UAH) disconnects often.

Interesting that S/N at 433 is 20dB and at X86 is 10dB.

So ... NV2 x XR9 x 5MHz ... all the down/up conversions at this cards ... or NV2 x NLOS is not the way to go?

Regards;
Marcus Ramos
Electronics Technician
(Microwave HW, RF, antennas, propagation)
S.Paulo - Brazil
 
uldis
MikroTik Support
MikroTik Support
Posts: 3427
Joined: Mon May 31, 2004 2:55 pm

Re: Explaination of NV2 register table readings please..

Wed May 18, 2011 5:05 pm

mramos, the distance calculation might be higher because of the channel width you use. Smaller channel width might effect the distance reading by making it 2 (for 10mhz channel) or 4 (for 5mhz channel) times bigger. We will try to check if we can optimize the distance calculation reading in future to show more accurate values when you use 5mhz or 10mhz channel width.
 
User avatar
mramos
Member Candidate
Member Candidate
Posts: 230
Joined: Sun Nov 23, 2008 1:05 am
Location: S. B do Campo - SP - Brazil

Re: Explaination of NV2 register table readings please..

Wed May 18, 2011 9:36 pm

the distance calculation might be higher because of the channel width you use. Smaller channel width might effect the distance reading by making it 2 (for 10mhz channel) or 4 (for 5mhz channel) times bigger.
Thanks sir. Helps if I set those distances manually at the CPEs (now using this channel bw compensation)? Or better to let it on automatic?

Don't take the setup I use as a standard one because still some weak points I need to deal with, such as a home made 2 way 900MHz splitter that was built on the fly with rough 75ohm cables (I rebuilt it on Rogers 4003 using microstrips but not finished yet) and the AP antenna structure is oscilating with strong wind (2m long yagis).

There is some interference at 433 CPE too. It's not a nearby cell tower but someone using 905 ... 925 MHz as a single 20MHz channel, may be UBNT TDMA. And our regulations split 902 ... 928 in two segments which left us room for a 5MHz on 917 or a 5/10MHz on 922. The other segment at the lower part of 900MHz band is useless, unless we use superchannel setup to take advantage of the 1MHz steps but not sure if XR9 handle it.

Regards;
Marcus Ramos
Electronics Technician
(Microwave HW, RF, antennas, propagation)
S.Paulo - Brazil
 
uldis
MikroTik Support
MikroTik Support
Posts: 3427
Joined: Mon May 31, 2004 2:55 pm

Re: Explaination of NV2 register table readings please..

Thu May 19, 2011 1:34 pm

mramos, is it possible to get support output files from your routers when you star to see those problems with the distance change?
 
User avatar
mramos
Member Candidate
Member Candidate
Posts: 230
Joined: Sun Nov 23, 2008 1:05 am
Location: S. B do Campo - SP - Brazil

Re: Explaination of NV2 register table readings please..

Thu May 19, 2011 2:51 pm

is it possible to get support output files from your routers when you star to see those problems with the distance change?
Yes, I did it now.

I'll forward 4 .rif files, 2 from AP (RB411AR), 2 from client (RB433UAH).

Difficult to trigger the snapshots because its a 1 sample a second on wireless monitor using terminal, 1 sample each 5 seconds on winbox wireless reg table.

This is a 5 minute graph from the calculated distance on RB433 client, 1 sample a second.

Image

Regards;
Marcus Ramos
Electronics Technician
(Microwave HW, RF, antennas, propagation)
S.Paulo - Brazil

Who is online

Users browsing this forum: No registered users and 20 guests