UmTRX October 2012

umtrx@lists.osmocom.org

23 participants
25 discussions

Re: Open-source telecom t-shirts
by Alexander Chemeris 03 Dec '12

03 Dec '12

(forwarding to the UmTRX mailing list) On Fri, Oct 19, 2012 at 7:33 PM, Ivan Kluchnikov <Ivan.Kluchnikov(a)fairwaves.ru> wrote: > Just sketch :) Nice idea. The last symbol would be better to have in the form of a SIM-card, imho. -- Regards, Alexander Chemeris. CEO, Fairwaves LLC / ООО УмРадио http://fairwaves.ru

8 34

Re: UmTRX Digest, Vol 4, Issue 45
by Darby 11 Nov '12

11 Nov '12

> Decreasing power consumption to 10-11W would allow us to use lighter > enclosure for the lab version, which is nice. at 20w of power consumption, this comes to roughly 68btu of generated heat -Is there a reason convection cooling in a sheetmetal enclosure wouldent work for this board?

3 10

Re: UmTRXv2 samples expected shortage
by Jean-Samuel Najnudel - BJT PARTNERS SARL 02 Nov '12

02 Nov '12

Hi Andrey, Do you think it will be possible to decrease the power consumption and to keep the best possible performances for industrial applications ? I know you cannot be sure about this but I would like to know how confident you feel about this ? Best regards. Jean-Samuel. :-) On Wed, Oct 31, 2012 at 7:40 PM, Andrey Sviyazov <andreysviyaz(a)gmail.com>wrote: > Hi Jean-Samuel, Alexander. > > Tomorrow I'll let you know what possible to make on time. > And we should set time limit for this. > For example, next morning. > > Best regards, > Andrey Sviyazov. > (Sent from my mobile client) > 31.10.2012 22:24 пользователь "Jean-Samuel Najnudel - BJT PARTNERS SARL" < > jsn(a)bjtpartners.com> написал: > > Hi Alexander, >> >> Thank you very much for your reply. >> >> Why skip/populate LMS power supply block ? >> As I understand we always need the LMS power supply block. I probably >> missed something. Could you explain this in more details to let me better >> understand ? >> >> Thanks a lot for your help. >> >> Best regards. >> >> Jean-Samuel. >> :-) >> >> >> On Wed, Oct 31, 2012 at 7:14 PM, Alexander Chemeris < >> alexander.chemeris(a)gmail.com> wrote: >> >>> Yes, lesser input voltage range makes sense only if it saves >$10 >>> and/or considerably increases power efficiency. I think this not the >>> case, and then only these changes will be needed: >>> >>> * traditional power connector >>> * MCX RF connector, because they sustain much more connect-disconnect >>> cycles >>> * lower power consumption mod >>> >>> I would appreciate if we could keep the same PCB for both versions and >>> populate the proper version of power connector/RF connector and >>> skip/populate LMS power supply block. >>> >>> On Wed, Oct 31, 2012 at 9:56 PM, Jean-Samuel Najnudel - BJT PARTNERS >>> SARL <jsn(a)bjtpartners.com> wrote: >>> > Hi Alexander, >>> > >>> > This would be much easier to have the same board for both lab and >>> > deployment. Only the UmSEL would make the difference. >>> > For deployment, I really need wide input voltage range as I plan to >>> power >>> > the whole system (UmTRX + PA) with a single 28V supply. >>> > Even if this can save a few euros, I would really prefer we do not >>> make the >>> > input voltage range smaller. >>> > By the way, even for lab, it might be convenient and it can avoid >>> damages in >>> > case of wrong voltage supply. >>> > >>> > Best regards. >>> > >>> > Jean-Samuel. >>> > :-) >>> > >>> > >>> > >>> > On Wed, Oct 31, 2012 at 6:50 PM, Alexander Chemeris >>> > <alexander.chemeris(a)gmail.com> wrote: >>> >> >>> >> Andrey, how much time do you need to create 2.1? If it's mire than a >>> day, >>> >> we should postpone this. I believe that enclosure is a much more >>> important >>> >> issue at this moment. >>> >> >>> >> I think we need following changes for the lab version: >>> >> * traditional power connector >>> >> * MCX RF connector, because they sustain much more connect-disconnect >>> >> cycles >>> >> * lower power consumption mod >>> >> * smaller input voltage range (only if this makes things cheaper) >>> >> >>> >> Sent from my Android device. >>> >> >>> >> -- >>> >> Regards, >>> >> Alexander Chemeris >>> >> CEO, Fairwaves LLC >>> >> http://fairwaves.ru >>> >> >>> >> 31.10.2012 14:05 пользователь "Andrey Sviyazov" < >>> andreysviyaz(a)gmail.com> >>> >> написал: >>> >> >>> >>> Hi Jean-Samuel. >>> >>> >>> >>> First of all, you didn't said about delay duration :) >>> >>> I can't delay this batch just due to my wishes that each next batch >>> >>> should work more and more ideally. >>> >>> >>> >>>> These modifications looks interesting. I think it is a good idea. >>> >>> >>> >>> Hope so. >>> >>> >>> >>>> >>> >>>> I just have a few question. >>> >>>> >>> >>>> Why do you call this board revision 2.1 "a special batch for labs" ? >>> >>>> Would these modifications make this revision 2.1 also more suitable >>> for >>> >>>> field deployment, at least as much as the revision 2.0 ? >>> >>> >>> >>> Because of I can fix only known issues. >>> >>> Also not all really required improvements are known yet. >>> >>> >>> >>>> Decreasing power consumption to 10..11 W would be great. Is there >>> any >>> >>>> drawbacks of this modification ? Would it decrease some >>> performances ? If >>> >>>> not, this modifiction to decrease power consumption is a >>> significant very >>> >>>> good modification. >>> >>> >>> >>> Decreasing of performances it is only spurs with DC/DC conversion >>> freq >>> >>> ~500kHz. >>> >>> Now I searching more good LC filter to suppress it better than >>> TI2012U601 >>> >>> can. >>> >>> >>> >>>> Do you have enough space on the board to replace some U_FL >>> connectors >>> >>>> with MCX connectors as you suggest ? >>> >>> >>> >>> Yes, but may be not all should be MCX. >>> >>> Of course I did not insist, but just asking whether there is a >>> reason to >>> >>> do it or not. >>> >>> >>> >>>> An external LNA would probably need around 5 Volts instead of 6V. A >>> >>>> small PA would probably need a little bit higher voltage. Do you >>> think it >>> >>>> would be possible to have a variable voltage low power connector ? >>> >>> >>> >>> On my opinion, variable voltage is not good idea. >>> >>> For example for LNA's better to place low noise LDO 6V to 5V near to >>> IC's >>> >>> to get Vcc clean too. >>> >>> >>> >>>> By the way, could you pelase also add the 2 LMS output matching >>> >>>> capacitors we need to improve output power figures in the 1800 band >>> ? >>> >>> >>> >>> Of course, because it is issue which should be fixed, but not >>> >>> improvement. >>> >>> >>> >>> Best regards, >>> >>> Andrey Sviyazov. >>> > >>> > >>> >>> >>> >>> -- >>> Regards, >>> Alexander Chemeris. >>> CEO, Fairwaves LLC / ООО УмРадио >>> http://fairwaves.ru >>> >> >>

3 7

LNA and Noise Figure improvement
by Alexander Chemeris 02 Nov '12

02 Nov '12

Forwarding to the mailing list. ---------- Forwarded message ---------- From: Jean-Samuel Najnudel - BJT PARTNERS SARL <jsn(a)bjtpartners.com> Date: Tue, Oct 30, 2012 at 6:25 PM Subject: LNA and Noise Figure improvement Hi Andrey, I understand you plan to prepare the UmTRXv3. I would need to discuss about the Rx Noise Figure improvement. With UmSEL, according to your calculations, NF would be around 0.7 dB (excluding duplexer IL) with the MGA-13116. It would be interesting to try to improve this figure with a very first stage LNA based on the MGA-633P8. This would let us reach a NF as low as about 0.4 dB. Adding this very first stage LNA in the Rx path brings a few questions. 1/ To improve isolation with the other RF parts, would we need to have this LNA in a separate shielded enclosure ? If yes, we would need a 5V power supply for this LNA. To have a better shielding between the voltage regulator circuitery and the LNA, it should be nice to have the 5V voltage regulator outside of the LNA enclosure. To make this possible, it would be very useful to add a 5V voltage regulator and connector on the UmTRXv3. If not, we do not need specific modifications on the UmTRXv3 but we would need to add an MGA-633P8/634P8 as a very first stage in the UmSEL. 2/ Whatever external or on the UmSEL, this extra LNA will increase the total gain. Considering your calculation, the UmSEL IIP3 is already as low as around 1 dBm. If we add a very first stage LNA, this IIP3 will go down to about -15 dBm. This could be a problem for inband blockers. Even if, in real life situation in the field, I am not sure it will be a very big problem, it will be a problem to pass the spec. Would you have another idea to improve the NF ? If not, could you please let me know what NF we could expect with the current design (UmSEL with MGA-13116/13216) in both 900 and 1800 bands ? Best regards. Jean-Samuel. :-) -- Regards, Alexander Chemeris. CEO, Fairwaves LLC / ООО УмРадио http://fairwaves.ru

3 4

GSM MAP for incoming roaming
by Dmitri Soloviev 01 Nov '12

01 Nov '12

Hi I'm preparing to implement a legacy GSM roaming for OpenBTS network. To start with, I'm trying to get a minimal list of mandatory MAP services to allow foreign subscribers to roam inside OpenBTS. CAMEL is not discussed now. Your help is appreciated. Regards, Dmitri MAP_UPDATE_LOCATION+ MAP_CANCEL_LOCATION+ MAP_SEND_IDENTIFICATION-? MAP_PURGE_MS+ MAP_UPDATE_GPRS_LOCATION MAP-NOTE-MM-EVENT MAP_PAGE MAP_SEARCH_FOR_MS MAP_PROCESS_ACCESS_REQUEST MAP_PREPARE_HANDOVER MAP_SEND_END_SIGNAL MAP_PROCESS_ACCESS_SIGNALLING MAP_FORWARD_ACCESS_SIGNALLING MAP_PREPARE_SUBSEQUENT_HANDOVER MAP_ALLOCATE_HANDOVER_NUMBER MAP_SEND_HANDOVER_REPORT MAP_AUTHENTICATE MAP_SEND_AUTHENTICATION_INFO+ MAP_AUTHENTICATION_FAILURE_REPORT? MAP_SET_CIPHERING_MODE MAP_CHECK_IMEI MAP_OBTAIN_IMEI MAP-INSERT-SUBSCRIBER-DATAneeded later? MAP-DELETE-SUBSCRIBER-DATAneeded later? MAP-PROVIDE-IMSI MAP-FORWARD-NEW-TMSI MAP_RESET+? MAP_FORWARD_CHECK_SS_INDICATION-? MAP_RESTORE_DATA+? MAP-ANY-TIME-INTERROGATION MAP-PROVIDE-SUBSCRIBER-INFO-? MAP-ANY-TIME-SUBSCRIPTION-INTERROGATION MAP-ANY-TIME-MODIFICATION MAP-NOTE-SUBSCRIBER-DATA-MODIFIED MAP-ACTIVATE-TRACE-MODE-? MAP-DEACTIVATE-TRACE-MODE-? MAP-TRACE-SUBSCRIBER-ACTIVITY MAP-SEND-IMSI MAP_SEND_ROUTING_INFORMATION MAP_PROVIDE_ROAMING_NUMBER+ MAP_RESUME_CALL_HANDLING-? MAP_PREPARE_GROUP_CALL MAP_PROCESS_GROUP MAP_FORWARD_GROUP_CALL_SIGNALLING MAP_SEND_GROUP_CALL_END_SIGNAL MAP_SEND_GROUP_CALL_INFO MAP_SET_REPORTING_STATE-? MAP_STATUS_REPORT-? MAP_REMOTE_USER_FREE-? MAP_IST_ALERT-? MAP_IST_COMMAND-? MAP_RELEASE_RESOURCES+? MAP_REGISTER_SS-? MAP_ERASE_SS-? MAP_ACTIVATE_SS-? MAP_DEACTIVATE_SS-? MAP_INTERROGATE_SS-? MAP_REGISTER_PASSWORD-? MAP_GET_PASSWORD-? MAP_PROCESS_UNSTRUCTURED_SS_REQUEST-? MAP_UNSTRUCTURED_SS_REQUEST-? MAP_UNSTRUCTURED_SS_NOTIFY-? MAP_SS_INVOCATION_NOTIFY MAP_REGISTER_CC_ENTRY MAP_ERASE_CC_ENTRY MAP-SEND-ROUTING-INFO-FOR-SM MAP-MO-FORWARD-SHORT-MESSAGE MAP-REPORT-SM-DELIVERY-STATUS MAP-READY-FOR-SM-? MAP-ALERT-SERVICE-CENTRE-? MAP-INFORM-SERVICE-CENTRE MAP-SEND-INFO-FOR-MT-SMS MAP-SEND-INFO-FOR-MO-SMS MAP-MT-FORWARD-SHORT-MESSAGE+ MAP-MT-FORWARD-SM-FOR-VGCS MAP_SEND_ROUTING_INFO_FOR_GPRS MAP_FAILURE_REPORT MAP_NOTE_MS_PRESENT_FOR_GPRS MAP-SEND-ROUTING-INFO-FOR-LCS MAP-PROVIDE-SUBSCRIBER-LOCATION MAP-SUBSCRIBER-LOCATION-REPORT

2 1

UmTRX issue
by Alexander Chemeris 31 Oct '12

31 Oct '12

Hi Thomas, Could you please describe symptoms which you observe on your UmTRX? We have to figure out the reason for the improper behavior and if this is a manufacturing issue - make sure this won't happen for the next batch. -- Regards, Alexander Chemeris. CEO, Fairwaves LLC / ООО УмРадио http://fairwaves.ru

2 1

Re: UmTRXv2 samples expected shortage
by Jean-Samuel Najnudel - BJT PARTNERS SARL 31 Oct '12

31 Oct '12

Hi Alexander, Thank you very much for your reply and detailled explainations. Now I understand. My point of view is we do not really care of a few watts power consumption saving for the lab version. At worst, we can still add a fan in the lab enclosure, as for the USRP which also dissipate up to about 15 Watts. On the other side, it makes a lot of sense to save some power consumption for the industrial version as, on the field, every watt counts and we also need to dissipate the heat without fan for better reliability. Considering this, I believe as follow. Either we are able to decrease the power consumption and keep good enough performances for industrial applications. In this case, we should make this modification for both lab and industrial versions. Either decreasing the power consumption will decrease the performances. In this case, I do not think this worth making this modification at all. In both cases, I would suggest we keep the same design for both lab and industrial versions, either with or without the modification. Moreover, if both lab and industrial versions are the same, except connectors may be, debug from lab users may help industrial users and debug from industrial users may help lab users. This would be really easier to get a reliable single board for both lab and industrial applications. Best regards. Jean-Samuel. :-) On Wed, Oct 31, 2012 at 7:30 PM, Alexander Chemeris < alexander.chemeris(a)gmail.com> wrote: > On Wed, Oct 31, 2012 at 10:23 PM, Jean-Samuel Najnudel - BJT PARTNERS > SARL <jsn(a)bjtpartners.com> wrote: > > Why skip/populate LMS power supply block ? > > As I understand we always need the LMS power supply block. I probably > missed > > something. Could you explain this in more details to let me better > > understand ? > > If I understand correctly, this is the modification which Andrey is > testing - avoid additional power conversion near LMS. We may need it > for better performance for industrial version, but we most likely do > not need it for the lab version. > > Decreasing power consumption to 10-11W would allow us to use lighter > enclosure for the lab version, which is nice. > > PS I'd love to move this discussion to the public UmTRX mailing list. > Is it fine with you? Actually, if you feel fine, you could start such > discussion at the public UmTRX mailing list from the beginning. > > -- > Regards, > Alexander Chemeris. > CEO, Fairwaves LLC / ООО УмРадио > http://fairwaves.ru >

1 0

Multicarrier OpenBTS on UmTRX
by Thomas Tsou 30 Oct '12

30 Oct '12

Hi all, I spent some time yesterday testing UmTRX with MCBTS. Aside from some trivial fixes (inverted sample rate), the code ran fine with no issues. The default bandwidth of the LPF was insufficient, so that was widened. I tested with 9 handsets between 3 and 7 carriers on a Core2 Duo (P7570) laptop. git://github.com/ttsou/openbts-multi-arfcn.git umtrx Configuration is through sqlite database, which is unchanged from mainline; just set the number of ARFCN (7 max) and C1 channels. Because of the bandwidth involved, be sure to increase the maximum buffer sizes for UHD. The uhd_usrp_probe utility will warn if the target size cannot be set. http://files.ettus.com/uhd_docs/manual/html/transport.html#linux-specific-n… Design ===== There are three resampling stages. 1. Outer resampler converts from the device rate to a multiple of the 400 kHz channel spacing. 2. Channelizer demultiplexes input stream to M channels 3. Inner resampler converts from 400 kHz to a multiple of the GSM symbol rate. UmTRX rates: Channelizer Rate Device Rate Decimation 1.6 Msps 1.625 Msps 8 3.2 Msps 3.250 Msps 4 Options ====== To broadcast dummy bursts on all carriers for spectrum testing, uncomment the following preprocessor declaration in radioParams.h. #define ENABLE_ALL_CHANS With dummy bursts on all carriers: http://filebox.vt.edu/users/ttsou/http/7carrier_umtrx.PNG Calibration ======== Calibration remains a concern. With single carrier, the carrier leakage and and IQ imbalance appear as in-band distortion. For MCBTS, OpenBTS offsets C0 on to the lowest carrier. This makes the carrier and image quite visible, as shown in the following uncalibrated capture. I checked with a swept tone to verify that this issue was not a result of baseband DSP. The USRP also shows similar effects. http://filebox.vt.edu/users/ttsou/http/carrier_image.PNG Any idea how much carrier and image suppression we should expect with calibration on UmTRXv2? Thomas

5 9

UmTRXv2 Beta: Whats the maximum # of simultaneous callers
by SoftwareRadioGuy 30 Oct '12

30 Oct '12

Hello, I've read that the # of maximum callers is 15. Since there are "Beta" version coming out will these be able to handle 15 simultaneous callers too? SoftwareDefinesRadio

3 2

Superb potential! Re: UmTRXv2 Beta: Whats the maximum # of simultaneous callers
by SoftwareRadioGuy 30 Oct '12

30 Oct '12

Alexander, Superb potential of the device you and the Fairwaves team are bringing to fruition. I also look forward to your future potential product, UmTRXv3 4G/LTE. I know its just something your team and you have thought about publically, alot of us would love to see such. Thank you for your responses. SoftwareDefinesRadio On Tue, Oct 30, 2012 at 5:07 AM, Alexander Chemeris < alexander.chemeris(a)gmail.com> wrote: > Hi, > > On Tue, Oct 30, 2012 at 12:36 PM, SoftwareRadioGuy > <softwaredefinesradio(a)gmail.com> wrote: > > I've read that the # of maximum callers is 15. > > > > Since there are "Beta" version coming out will these be able to handle 15 > > simultaneous callers too? > > A short answer is "yes, after a software update". > > A longer answer is that hardware supports dual-TRX operation, but > software support is not there yet. We might be able to fix software > before we ship beta units, but even if not - you will be able to get > the software update later. > Also keep in mind, that UmTRX should be able to run multi-ARFCN > OpenBTS and in this case it'll be able to handle much more calls. We > haven't tested this mode of operation extensively yet, but > theoretically we might be able to support 14ARFCNs (2x7ARFCN) which > equals to (14*8-1)=111 calls. At this scale you're most likely limited > by computing power of your PC rather then UmTRX hardware. Whether we > could get UmTRX in Multi-ARFCN mode through certification or not is > still a question, but it should be fine for lab testing. > > > -- > Regards, > Alexander Chemeris. > CEO, Fairwaves LLC / ООО УмРадио > http://fairwaves.ru >

1 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

UmTRX October 2012