Re: [Discuss-gnuradio] Pay You $$$ to Mentor Me - Tx OFDM 64 QAM and Rx Measure Error Vector Magnitude

[Marcus Müller]

Hi Ifly,

yes, I understood that you wanted to make the LTE-typical OFDM 
configuration, but again, you can't measure anything useful with only 
that! So, what *exactly* do you want to measure with your setup? I 
promise you, you'll get better results if you try to explain your 
overall application to us! The first thing I (and anyone else) would do 
if I were to take up the job to do consulting with you is to sit down 
and figure out what you're actually trying to do *in the bigger 
picture*, and then help you understand your own requirements, and only 
after that, help you implement these. In case you ever happen to write a 
paper, thesis or any other publication about this: [1] is a must-read.


Best regards,

Marcus

[1] 
https://www.microsoft.com/en-us/research/wp-content/uploads/2016/12/State-the-Problem-Before-Describing-the-Solution.pdf


On 09/25/2017 04:38 AM, Ifly88 wrote:
I want to create a spread of 15khz sub carriers ( like LTE ) that 
fills up bandwidth of 1.4 MHz, 3 MHz, 5 MHz, 10 MHz, 15 MHz, and 20 
MHz ( like LTE bandwidths ).  In the 1.4 MHz block there are 72 15 khz 
subcarriers, 180 in a 3 MHz block, 300 in a 5 MHz, 600 in a 10 MHz, 
900 in a 15 MHz, and 1200 in a 20 MHz bandwidth.


The subcarriers are to be modulated with a constant QPSK, or 16 QAM or 
64 QAM.  A BladeRF or HackRF One will be used to measure and generate 
these signals.  Preferably the HackRF because the BladeRF seems to be 
more unstable (locks up) in my experience.  May be bad code or 
hardware, not sure which one.


The goal is to test RF components to see if they are causing signal 
distortion.  In the past I had access to a LTE signal generator that 
would key up all sub-carriers with 64 quam and a LTE signal analyser 
that would measure EVM and that was a great way to see if a system 
could transmit 64 QAM without issue.  If the EVM was over a certain X 
for 64 QAM, 16 QAM or QPSK then the components needed to be addressed.


Below is the chart that shows Bandwidth and Subcarriers in an LTE signal.

Thanks,
Andrew





Sent with ProtonMail <https://protonmail.com> Secure Email.


 Original Message 
Subject: Re: [Discuss-gnuradio] Pay You $$$ to Mentor Me - Tx OFDM 64 
QAM and Rx Measure Error Vector Magnitude

Local Time: September 24, 2017 3:54 PM
UTC Time: September 24, 2017 8:54 PM
From: marcus.muel...@ettus.com
To: discuss-gnuradio@gnu.org


HI IFly88,

I don't fully understand your requirements – on one hand, you just 
say "I just want OFDM with 15 kHz subcarrier spacing and QAM/QPSK 
inside", on the other hand you mention LTE: an arbitrary OFDM signal 
with the right subcarrier spacing and constellation mapping does not 
constitute a validly measurable LTE signal. Your LTE receiver needs a 
lot of specific structure in the frames to be able to synchronize, 
and equalize, to an LTE transmission.


So, what exactly are you using to measure EVM? What is the higher 
level goal of this? Your transmit hardware will always also play a 
part in the signal quality you receive, so it's usually better to 
first sit down and contemplate what you want to measure, then design 
a system.


Best regards,
Marcus

On 09/23/2017 10:30 PM, Ifly88 wrote:
Name your price to give me pointers and help me kick start this 
project.  I want to emulate a 64qam, 16qam, and qpsk LTEs signal to 
test for signal impairments.  Basically take 15khz sub carriers over 
channel bandwidths of 1.4mhz to 20 mhz and modulate them with 64qam, 
16qam, and qpsk.  I want to use another receiver to Rx the signal 
and calculate error vector magnitude for each of those three 
modulations.  I'll be using HackRF One radios or will use BladeRF if 
necessary.


I'be been spinning my wheels for a few days trying to get this to 
work on GRC and need some guidance.


Please help.
ifly88 at protonmail dot com


Sent with ProtonMail <https://protonmail.com> Secure Email.



___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio





___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio


___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] Pay You $$$ to Mentor Me - Tx OFDM 64 QAM and Rx Measure Error Vector Magnitude

[Ifly88]

I want to create a spread of 15khz sub carriers ( like LTE ) that fills up 
bandwidth of 1.4 MHz, 3 MHz, 5 MHz, 10 MHz, 15 MHz, and 20 MHz ( like LTE 
bandwidths ).  In the 1.4 MHz block there are 72 15 khz subcarriers, 180 in a 3 
MHz block, 300 in a 5 MHz, 600 in a 10 MHz, 900 in a 15 MHz, and 1200 in a 20 
MHz bandwidth.

The subcarriers are to be modulated with a constant QPSK, or 16 QAM or 64 QAM.  
A BladeRF or HackRF One will be used to measure and generate these signals.  
Preferably the HackRF because the BladeRF seems to be more unstable (locks up) 
in my experience.  May be bad code or hardware, not sure which one.

The goal is to test RF components to see if they are causing signal distortion. 
 In the past I had access to a LTE signal generator that would key up all 
sub-carriers with 64 quam and a LTE signal analyser that would measure EVM and 
that was a great way to see if a system could transmit 64 QAM without issue.  
If the EVM was over a certain X for 64 QAM, 16 QAM or QPSK then the components 
needed to be addressed.

Below is the chart that shows Bandwidth and Subcarriers in an LTE signal.

Thanks,
Andrew

Sent with [ProtonMail](https://protonmail.com) Secure Email.

>  Original Message 
> Subject: Re: [Discuss-gnuradio] Pay You $$$ to Mentor Me - Tx OFDM 64 QAM and 
> Rx Measure Error Vector Magnitude
> Local Time: September 24, 2017 3:54 PM
> UTC Time: September 24, 2017 8:54 PM
> From: marcus.muel...@ettus.com
> To: discuss-gnuradio@gnu.org
>
> HI IFly88,
>
> I don't fully understand your requirements – on one hand, you just say "I 
> just want OFDM with 15 kHz subcarrier spacing and QAM/QPSK inside", on the 
> other hand you mention LTE: an arbitrary OFDM signal with the right 
> subcarrier spacing and constellation mapping does not constitute a validly 
> measurable LTE signal. Your LTE receiver needs a lot of specific structure in 
> the frames to be able to synchronize, and equalize, to an LTE transmission.
>
> So, what exactly are you using to measure EVM? What is the higher level goal 
> of this? Your transmit hardware will always also play a part in the signal 
> quality you receive, so it's usually better to first sit down and contemplate 
> what you want to measure, then design a system.
>
> Best regards,
> Marcus
>
> On 09/23/2017 10:30 PM, Ifly88 wrote:
>
>> Name your price to give me pointers and help me kick start this project.  I 
>> want to emulate a 64qam, 16qam, and qpsk LTEs signal to test for signal 
>> impairments.  Basically take 15khz sub carriers over channel bandwidths of 
>> 1.4mhz to 20 mhz and modulate them with 64qam, 16qam, and qpsk.  I want to 
>> use another receiver to Rx the signal and calculate error vector magnitude 
>> for each of those three modulations.  I'll be using HackRF One radios or 
>> will use BladeRF if necessary.
>>
>> I'be been spinning my wheels for a few days trying to get this to work on 
>> GRC and need some guidance.
>>
>> Please help.
>> ifly88 at protonmail dot com
>>
>> Sent with [ProtonMail](https://protonmail.com) Secure Email.
>>
>> ___
>> Discuss-gnuradio mailing list
>> Discuss-gnuradio@gnu.org
>>
>> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] Pay You $$$ to Mentor Me - Tx OFDM 64 QAM and Rx Measure Error Vector Magnitude

[Marcus Müller]

HI IFly88,

I don't fully understand your requirements – on one hand, you just say
"I just want OFDM with 15 kHz subcarrier spacing and QAM/QPSK inside",
on the other hand you mention LTE: an arbitrary OFDM signal with the
right subcarrier spacing and constellation mapping does not constitute a
validly measurable LTE signal. Your LTE receiver needs a lot of specific
structure in the frames to be able to synchronize, and equalize, to an
LTE transmission.

So, what exactly are you using to measure EVM? What is the higher level
goal of this? Your transmit hardware will always also play a part in the
signal quality you receive, so it's usually better to first sit down and
contemplate what you want to measure, then design a system.

Best regards,
Marcus


On 09/23/2017 10:30 PM, Ifly88 wrote:
> Name your price to give me pointers and help me kick start this
> project.  I want to emulate a 64qam, 16qam, and qpsk LTEs signal to
> test for signal impairments.  Basically take 15khz sub carriers over
> channel bandwidths of 1.4mhz to 20 mhz and modulate them with 64qam,
> 16qam, and qpsk.  I want to use another receiver to Rx the signal and
> calculate error vector magnitude for each of those three modulations.
>  I'll be using HackRF One radios or will use BladeRF if necessary.
>
> I'be been spinning my wheels for a few days trying to get this to work
> on GRC and need some guidance.
>
> Please help.
> ifly88 at protonmail dot com
>
>
> Sent with ProtonMail  Secure Email.
>
>
>
> ___
> Discuss-gnuradio mailing list
> Discuss-gnuradio@gnu.org
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

[Discuss-gnuradio] Pay You $$$ to Mentor Me - Tx OFDM 64 QAM and Rx Measure Error Vector Magnitude

[Ifly88]

Name your price to give me pointers and help me kick start this project.  I 
want to emulate a 64qam, 16qam, and qpsk LTEs signal to test for signal 
impairments.  Basically take 15khz sub carriers over channel bandwidths of 
1.4mhz to 20 mhz and modulate them with 64qam, 16qam, and qpsk.  I want to use 
another receiver to Rx the signal and calculate error vector magnitude for each 
of those three modulations.  I'll be using HackRF One radios or will use 
BladeRF if necessary.

I'be been spinning my wheels for a few days trying to get this to work on GRC 
and need some guidance.

Please help.
ifly88 at protonmail dot com

Sent with [ProtonMail](https://protonmail.com) Secure Email.___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio