Re: [Discuss-gnuradio] OFDM benchmark optimal parameter

[Diyar Muhammed]

Marcus, many thanks I will do it.

On Mon, Mar 21, 2016 at 11:01 AM, Marcus Müller 
wrote:

> I'd encourage you to either fix the Bit Error Rate block or write
> something that does your job. In fact, the unmodified ofdm_loopback example
> doesn't work as BER test, because all packets are identical, and if a
> packet has errors, the OFDM receiver will drop it, so you'd never see an
> error.
>
> Open rx_ofdm.grc ; it is a very similar example, but instead of having the
> black box "OFDM Receiver", you see how the OFDM receiver internally works.
> Play with the channel model; e.g. set the noise voltage really high (1.0)
> and the frequency offset to e.g. 2.0/fft_len. You'll see a lot of
>
> INFO: Detected and invalid packet at item 
>
> printed.
> Now, change these parameters.
> Your ratio of valid packets and invalid packets gives you a packet error
> rate.
>
> Best regards,
> Marcus
>
>
> On 21.03.2016 11:47, Diyar Muhammed wrote:
>
> Marcus,
> I look at ofdm_loopback.grc example, I made the same scenario but I had
> problem with Error Rate block I got error rate around 4 to 5, as my
> knowledge that is not right I think should be between 0 to 1.
> If there is a transceiver example with measure bit error rate that will be
> helpful for me.
> in advance thank you.
>
> On Mon, Mar 21, 2016 at 10:36 AM, Marcus Müller 
> wrote:
>
>> Note that the benchmark_rx/_tx example is really a bit old, and I always
>> try to steer people away from it towards the newer OFDM examples that are
>> far more flexible and behave a lot more like a real system would.
>>
>> Have a look at the ofdm_loopback.grc example; you can replace the
>> (channelmodel->throttle) by a USRP sink and source. Tadah! Live demo.
>>
>> Best regards,
>> Marcus
>>
>>
>> On 21.03.2016 11:34, Diyar Muhammed wrote:
>>
>> many thanks
>>
>> On Mon, Mar 21, 2016 at 10:31 AM, Marcus Müller <
>> marcus.muel...@ettus.com> wrote:
>>
>>> Diyar,
>>>
>>> > I look at tx benchmark help I could not find rates but there is packet
>>> size and megabytes to transmit.
>>>
>>> benchmark_tx --help should help you.
>>> You set the bandwidth, which sets the sampling rate; together with the
>>> occupied tones number related to the FFT length, you get a symbol rate.
>>> Together with the modulation you set, this gives you a
>>>
>>> Since only one program can use a USRP at a time, you can't use
>>> benchmark_tx and benchmark_rx at the same time.
>>> Instead, use benchmark_tx with the "--to-file" option to save the
>>> samples to a file, and build a quick GNU Radio flow graph in GRC that has a
>>> file source (reading that file), a USRP sink (fed from the file source), a
>>> USRP source, and a file sink (saving the samples from the USRP source to
>>> another file).
>>>
>>> Then use benchmark_rx with the --from-file option to read in these saved
>>> samples.
>>>
>>> Best regards,
>>> Marcus
>>>
>>>
>>> On 21.03.2016 11:17, Diyar Muhammed wrote:
>>>
>>> Dear Marcus,
>>> Thank you very much indeed for fast replying.
>>> I look at tx benchmark help I could not find rates but there is packet
>>> size and megabytes to transmit.
>>> so that, which one do you mean packet size or megabytes?
>>> it is okay to use USRP B210 for transmitting and receiving by using to
>>> benchmark file?
>>> because when I used one of them (tx or rx) and then I wanted to run
>>> another one the error come up (no device found for empty device address).
>>> in advance many thanks.
>>>
>>> On Mon, Mar 21, 2016 at 9:57 AM, Marcus Müller <
>>> marcus.muel...@ettus.com> wrote:
>>>
 Diyar,

 with the benchmark_ scripts, you **set** the rates, and you can only
 observe how many packets were successfully transmitted.
 The rest is really very basic math.

 Best regards,
 Marcus


 On 21.03.2016 10:50, Diyar Muhammed wrote:

 Dear SangHyuk,
 I would like to know how to measure Throughput and BER by using
 benchmark tx and rx?
 could you show or explain with real example as you used.
 in advance thanks.

 On Mon, Mar 21, 2016 at 8:09 AM, Marcus Müller <
 marcus.muel...@ettus.com> wrote:

> Hi,
>
> On 21.03.2016 01 <21.03.2016%2001>:37, SangHyuk Kim wrote:
> > I want to know other user's performance (avg performance).
> Yes, but what is "user's performance"? Is it more important to have
> higher throughput, or lower error rates? What about robustness?
>
> I mean, the OFDM rx_benchmark is a really static example.
> You might find a setting that maximizes troughput for a given channel,
> but imagine something happens that reduces your receiver's SNR by 3dB:
> Now your suddenly losing a lot of performance.
>
> Really "how can I parameterize this" can only be answered for a single,
> mathematically well-defined target, and for a well-defined channel.
>

Re: [Discuss-gnuradio] OFDM benchmark optimal parameter

[Marcus Müller]

I'd encourage you to either fix the Bit Error Rate block or write
something that does your job. In fact, the unmodified ofdm_loopback
example doesn't work as BER test, because all packets are identical, and
if a packet has errors, the OFDM receiver will drop it, so you'd never
see an error.

Open rx_ofdm.grc ; it is a very similar example, but instead of having
the black box "OFDM Receiver", you see how the OFDM receiver internally
works.
Play with the channel model; e.g. set the noise voltage really high
(1.0) and the frequency offset to e.g. 2.0/fft_len. You'll see a lot of

INFO: Detected and invalid packet at item 

printed.
Now, change these parameters.
Your ratio of valid packets and invalid packets gives you a packet error
rate.

Best regards,
Marcus

On 21.03.2016 11:47, Diyar Muhammed wrote:
> Marcus,
> I look at ofdm_loopback.grc example, I made the same scenario but I
> had problem with Error Rate block I got error rate around 4 to 5, as
> my knowledge that is not right I think should be between 0 to 1.
> If there is a transceiver example with measure bit error rate that
> will be helpful for me.
> in advance thank you.
>
> On Mon, Mar 21, 2016 at 10:36 AM, Marcus Müller
> > wrote:
>
> Note that the benchmark_rx/_tx example is really a bit old, and I
> always try to steer people away from it towards the newer OFDM
> examples that are far more flexible and behave a lot more like a
> real system would.
>
> Have a look at the ofdm_loopback.grc example; you can replace the
> (channelmodel->throttle) by a USRP sink and source. Tadah! Live demo.
>
> Best regards,
> Marcus
>
>
> On 21.03.2016 11:34, Diyar Muhammed wrote:
>> many thanks 
>>
>> On Mon, Mar 21, 2016 at 10:31 AM, Marcus Müller
>> > wrote:
>>
>> Diyar,
>>
>> > I look at tx benchmark help I could not find rates but
>> there is packet size and megabytes to transmit.
>>
>> benchmark_tx --help should help you.
>> You set the bandwidth, which sets the sampling rate; together
>> with the occupied tones number related to the FFT length, you
>> get a symbol rate.
>> Together with the modulation you set, this gives you a
>>
>> Since only one program can use a USRP at a time, you can't
>> use benchmark_tx and benchmark_rx at the same time.
>> Instead, use benchmark_tx with the "--to-file" option to save
>> the samples to a file, and build a quick GNU Radio flow graph
>> in GRC that has a file source (reading that file), a USRP
>> sink (fed from the file source), a USRP source, and a file
>> sink (saving the samples from the USRP source to another file).
>>
>> Then use benchmark_rx with the --from-file option to read in
>> these saved samples.
>>
>> Best regards,
>> Marcus
>>
>>
>> On 21.03.2016 11:17, Diyar Muhammed wrote:
>>> Dear Marcus,
>>> Thank you very much indeed for fast replying.
>>> I look at tx benchmark help I could not find rates but there
>>> is packet size and megabytes to transmit.
>>> so that, which one do you mean packet size or megabytes?
>>> it is okay to use USRP B210 for transmitting and receiving
>>> by using to benchmark file?
>>> because when I used one of them (tx or rx) and then I wanted
>>> to run another one the error come up (no device found for
>>> empty device address).
>>> in advance many thanks.
>>>
>>> On Mon, Mar 21, 2016 at 9:57 AM, Marcus Müller
>>> >
>>> wrote:
>>>
>>> Diyar,
>>>
>>> with the benchmark_ scripts, you **set** the rates, and
>>> you can only observe how many packets were successfully
>>> transmitted.
>>> The rest is really very basic math.
>>>
>>> Best regards,
>>> Marcus
>>>
>>>
>>> On 21.03.2016 10:50, Diyar Muhammed wrote:
 Dear SangHyuk,
 I would like to know how to measure Throughput and BER
 by using benchmark tx and rx?
 could you show or explain with real example as you used.
 in advance thanks.

 On Mon, Mar 21, 2016 at 8:09 AM, Marcus Müller
 > wrote:

 Hi,

 On 21.03.2016 01 :37, SangHyuk
 Kim wrote:
 > I want to know other user's performance (avg
 performance).
 Yes, but what is "user's performance"? Is it more
 important to have
 

Re: [Discuss-gnuradio] OFDM benchmark optimal parameter

[Diyar Muhammed]

Marcus,
I look at ofdm_loopback.grc example, I made the same scenario but I had
problem with Error Rate block I got error rate around 4 to 5, as my
knowledge that is not right I think should be between 0 to 1.
If there is a transceiver example with measure bit error rate that will be
helpful for me.
in advance thank you.

On Mon, Mar 21, 2016 at 10:36 AM, Marcus Müller 
wrote:

> Note that the benchmark_rx/_tx example is really a bit old, and I always
> try to steer people away from it towards the newer OFDM examples that are
> far more flexible and behave a lot more like a real system would.
>
> Have a look at the ofdm_loopback.grc example; you can replace the
> (channelmodel->throttle) by a USRP sink and source. Tadah! Live demo.
>
> Best regards,
> Marcus
>
>
> On 21.03.2016 11:34, Diyar Muhammed wrote:
>
> many thanks
>
> On Mon, Mar 21, 2016 at 10:31 AM, Marcus Müller 
> wrote:
>
>> Diyar,
>>
>> > I look at tx benchmark help I could not find rates but there is packet
>> size and megabytes to transmit.
>>
>> benchmark_tx --help should help you.
>> You set the bandwidth, which sets the sampling rate; together with the
>> occupied tones number related to the FFT length, you get a symbol rate.
>> Together with the modulation you set, this gives you a
>>
>> Since only one program can use a USRP at a time, you can't use
>> benchmark_tx and benchmark_rx at the same time.
>> Instead, use benchmark_tx with the "--to-file" option to save the samples
>> to a file, and build a quick GNU Radio flow graph in GRC that has a file
>> source (reading that file), a USRP sink (fed from the file source), a USRP
>> source, and a file sink (saving the samples from the USRP source to another
>> file).
>>
>> Then use benchmark_rx with the --from-file option to read in these saved
>> samples.
>>
>> Best regards,
>> Marcus
>>
>>
>> On 21.03.2016 11:17, Diyar Muhammed wrote:
>>
>> Dear Marcus,
>> Thank you very much indeed for fast replying.
>> I look at tx benchmark help I could not find rates but there is packet
>> size and megabytes to transmit.
>> so that, which one do you mean packet size or megabytes?
>> it is okay to use USRP B210 for transmitting and receiving by using to
>> benchmark file?
>> because when I used one of them (tx or rx) and then I wanted to run
>> another one the error come up (no device found for empty device address).
>> in advance many thanks.
>>
>> On Mon, Mar 21, 2016 at 9:57 AM, Marcus Müller <
>> marcus.muel...@ettus.com> wrote:
>>
>>> Diyar,
>>>
>>> with the benchmark_ scripts, you **set** the rates, and you can only
>>> observe how many packets were successfully transmitted.
>>> The rest is really very basic math.
>>>
>>> Best regards,
>>> Marcus
>>>
>>>
>>> On 21.03.2016 10:50, Diyar Muhammed wrote:
>>>
>>> Dear SangHyuk,
>>> I would like to know how to measure Throughput and BER by using
>>> benchmark tx and rx?
>>> could you show or explain with real example as you used.
>>> in advance thanks.
>>>
>>> On Mon, Mar 21, 2016 at 8:09 AM, Marcus Müller <
>>> marcus.muel...@ettus.com> wrote:
>>>
 Hi,

 On 21.03.2016 01 <21.03.2016%2001>:37, SangHyuk Kim wrote:
 > I want to know other user's performance (avg performance).
 Yes, but what is "user's performance"? Is it more important to have
 higher throughput, or lower error rates? What about robustness?

 I mean, the OFDM rx_benchmark is a really static example.
 You might find a setting that maximizes troughput for a given channel,
 but imagine something happens that reduces your receiver's SNR by 3dB:
 Now your suddenly losing a lot of performance.

 Really "how can I parameterize this" can only be answered for a single,
 mathematically well-defined target, and for a well-defined channel.

 In a real-world scenario, if using a transceiver with a fixed
 modulation, you usually wouldn't maximize throughput for a given
 setting, but you would define what "it still works sufficiently" means,
 and then you'd define "the worst channel I want the system to still work
 sufficiently".
 Then you'd come up with a metric that gives you a number for "the link
 quality on all considerable channels where this should be working", and
 then you'd try to maximize that metric under the outage constraints set
 before. Notice that this metric has to take things like error rate,
 throughtput, the "cost" of re-sending something (if you have a mechanism
 for that), available channel coding, how much you care about latency,
 computational complexity (that really gets important with iterative
 channel decoding),

 In other words:
 This is digital communications. If there was a single "best" solution,
 we'd all be using that and be done. Use your digital communications
 knowledge to analyze your requirements and challenges!

 Best 

Re: [Discuss-gnuradio] OFDM benchmark optimal parameter

[Marcus Müller]

Note that the benchmark_rx/_tx example is really a bit old, and I always
try to steer people away from it towards the newer OFDM examples that
are far more flexible and behave a lot more like a real system would.

Have a look at the ofdm_loopback.grc example; you can replace the
(channelmodel->throttle) by a USRP sink and source. Tadah! Live demo.

Best regards,
Marcus

On 21.03.2016 11:34, Diyar Muhammed wrote:
> many thanks 
>
> On Mon, Mar 21, 2016 at 10:31 AM, Marcus Müller
> > wrote:
>
> Diyar,
>
> > I look at tx benchmark help I could not find rates but there is
> packet size and megabytes to transmit.
>
> benchmark_tx --help should help you.
> You set the bandwidth, which sets the sampling rate; together with
> the occupied tones number related to the FFT length, you get a
> symbol rate.
> Together with the modulation you set, this gives you a
>
> Since only one program can use a USRP at a time, you can't use
> benchmark_tx and benchmark_rx at the same time.
> Instead, use benchmark_tx with the "--to-file" option to save the
> samples to a file, and build a quick GNU Radio flow graph in GRC
> that has a file source (reading that file), a USRP sink (fed from
> the file source), a USRP source, and a file sink (saving the
> samples from the USRP source to another file).
>
> Then use benchmark_rx with the --from-file option to read in these
> saved samples.
>
> Best regards,
> Marcus
>
>
> On 21.03.2016 11:17, Diyar Muhammed wrote:
>> Dear Marcus,
>> Thank you very much indeed for fast replying.
>> I look at tx benchmark help I could not find rates but there is
>> packet size and megabytes to transmit.
>> so that, which one do you mean packet size or megabytes?
>> it is okay to use USRP B210 for transmitting and receiving by
>> using to benchmark file?
>> because when I used one of them (tx or rx) and then I wanted to
>> run another one the error come up (no device found for empty
>> device address).
>> in advance many thanks.
>>
>> On Mon, Mar 21, 2016 at 9:57 AM, Marcus Müller
>> > wrote:
>>
>> Diyar,
>>
>> with the benchmark_ scripts, you **set** the rates, and you
>> can only observe how many packets were successfully transmitted.
>> The rest is really very basic math.
>>
>> Best regards,
>> Marcus
>>
>>
>> On 21.03.2016 10:50, Diyar Muhammed wrote:
>>> Dear SangHyuk,
>>> I would like to know how to measure Throughput and BER by
>>> using benchmark tx and rx?
>>> could you show or explain with real example as you used.
>>> in advance thanks.
>>>
>>> On Mon, Mar 21, 2016 at 8:09 AM, Marcus Müller
>>> >
>>> wrote:
>>>
>>> Hi,
>>>
>>> On 21.03.2016 01 :37, SangHyuk Kim
>>> wrote:
>>> > I want to know other user's performance (avg performance).
>>> Yes, but what is "user's performance"? Is it more
>>> important to have
>>> higher throughput, or lower error rates? What about
>>> robustness?
>>>
>>> I mean, the OFDM rx_benchmark is a really static example.
>>> You might find a setting that maximizes troughput for a
>>> given channel,
>>> but imagine something happens that reduces your
>>> receiver's SNR by 3dB:
>>> Now your suddenly losing a lot of performance.
>>>
>>> Really "how can I parameterize this" can only be
>>> answered for a single,
>>> mathematically well-defined target, and for a
>>> well-defined channel.
>>>
>>> In a real-world scenario, if using a transceiver with a
>>> fixed
>>> modulation, you usually wouldn't maximize throughput for
>>> a given
>>> setting, but you would define what "it still works
>>> sufficiently" means,
>>> and then you'd define "the worst channel I want the
>>> system to still work
>>> sufficiently".
>>> Then you'd come up with a metric that gives you a number
>>> for "the link
>>> quality on all considerable channels where this should
>>> be working", and
>>> then you'd try to maximize that metric under the outage
>>> constraints set
>>> before. Notice that this metric has to take things like
>>> error rate,
>>> throughtput, the "cost" of re-sending something (if you
>>> have a mechanism
>>> for that), available channel coding, how much you care
>>> about latency,

Re: [Discuss-gnuradio] OFDM benchmark optimal parameter

[Diyar Muhammed]

many thanks

On Mon, Mar 21, 2016 at 10:31 AM, Marcus Müller 
wrote:

> Diyar,
>
> > I look at tx benchmark help I could not find rates but there is packet
> size and megabytes to transmit.
>
> benchmark_tx --help should help you.
> You set the bandwidth, which sets the sampling rate; together with the
> occupied tones number related to the FFT length, you get a symbol rate.
> Together with the modulation you set, this gives you a
>
> Since only one program can use a USRP at a time, you can't use
> benchmark_tx and benchmark_rx at the same time.
> Instead, use benchmark_tx with the "--to-file" option to save the samples
> to a file, and build a quick GNU Radio flow graph in GRC that has a file
> source (reading that file), a USRP sink (fed from the file source), a USRP
> source, and a file sink (saving the samples from the USRP source to another
> file).
>
> Then use benchmark_rx with the --from-file option to read in these saved
> samples.
>
> Best regards,
> Marcus
>
>
> On 21.03.2016 11:17, Diyar Muhammed wrote:
>
> Dear Marcus,
> Thank you very much indeed for fast replying.
> I look at tx benchmark help I could not find rates but there is packet
> size and megabytes to transmit.
> so that, which one do you mean packet size or megabytes?
> it is okay to use USRP B210 for transmitting and receiving by using to
> benchmark file?
> because when I used one of them (tx or rx) and then I wanted to run
> another one the error come up (no device found for empty device address).
> in advance many thanks.
>
> On Mon, Mar 21, 2016 at 9:57 AM, Marcus Müller 
> wrote:
>
>> Diyar,
>>
>> with the benchmark_ scripts, you **set** the rates, and you can only
>> observe how many packets were successfully transmitted.
>> The rest is really very basic math.
>>
>> Best regards,
>> Marcus
>>
>>
>> On 21.03.2016 10:50, Diyar Muhammed wrote:
>>
>> Dear SangHyuk,
>> I would like to know how to measure Throughput and BER by using benchmark
>> tx and rx?
>> could you show or explain with real example as you used.
>> in advance thanks.
>>
>> On Mon, Mar 21, 2016 at 8:09 AM, Marcus Müller <
>> marcus.muel...@ettus.com> wrote:
>>
>>> Hi,
>>>
>>> On 21.03.2016 01 <21.03.2016%2001>:37, SangHyuk Kim wrote:
>>> > I want to know other user's performance (avg performance).
>>> Yes, but what is "user's performance"? Is it more important to have
>>> higher throughput, or lower error rates? What about robustness?
>>>
>>> I mean, the OFDM rx_benchmark is a really static example.
>>> You might find a setting that maximizes troughput for a given channel,
>>> but imagine something happens that reduces your receiver's SNR by 3dB:
>>> Now your suddenly losing a lot of performance.
>>>
>>> Really "how can I parameterize this" can only be answered for a single,
>>> mathematically well-defined target, and for a well-defined channel.
>>>
>>> In a real-world scenario, if using a transceiver with a fixed
>>> modulation, you usually wouldn't maximize throughput for a given
>>> setting, but you would define what "it still works sufficiently" means,
>>> and then you'd define "the worst channel I want the system to still work
>>> sufficiently".
>>> Then you'd come up with a metric that gives you a number for "the link
>>> quality on all considerable channels where this should be working", and
>>> then you'd try to maximize that metric under the outage constraints set
>>> before. Notice that this metric has to take things like error rate,
>>> throughtput, the "cost" of re-sending something (if you have a mechanism
>>> for that), available channel coding, how much you care about latency,
>>> computational complexity (that really gets important with iterative
>>> channel decoding),
>>>
>>> In other words:
>>> This is digital communications. If there was a single "best" solution,
>>> we'd all be using that and be done. Use your digital communications
>>> knowledge to analyze your requirements and challenges!
>>>
>>> Best regards
>>> Marcus
>>>
>>> ___
>>> Discuss-gnuradio mailing list
>>> Discuss-gnuradio@gnu.org
>>> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>>>
>>
>>
>>
>> --
>> Regards,
>> Diyar Muhammed
>> Ministry of Higher Education and Scientific Research
>> Duty: Network Administration and Design
>> Website:   www.mhe-krg.org
>> Cell Phone: 009647504690060
>> Office Phone:   00964662554683
>>
>>
>>
>
>
> --
> Regards,
> Diyar Muhammed
> Ministry of Higher Education and Scientific Research
> Duty: Network Administration and Design
> Website:  www.mhe-krg.org
> Cell Phone: 009647504690060
> Office Phone:   00964662554683
>
>
>


-- 
Regards,
Diyar Muhammed
Ministry of Higher Education and Scientific Research
Duty: Network Administration and Design
Website:  www.mhe-krg.org
Cell Phone: 009647504690060
Office Phone:   00964662554683
___
Discuss-gnuradio 

Re: [Discuss-gnuradio] OFDM benchmark optimal parameter

[Marcus Müller]

Diyar,

> I look at tx benchmark help I could not find rates but there is packet
size and megabytes to transmit.

benchmark_tx --help should help you.
You set the bandwidth, which sets the sampling rate; together with the
occupied tones number related to the FFT length, you get a symbol rate.
Together with the modulation you set, this gives you a

Since only one program can use a USRP at a time, you can't use
benchmark_tx and benchmark_rx at the same time.
Instead, use benchmark_tx with the "--to-file" option to save the
samples to a file, and build a quick GNU Radio flow graph in GRC that
has a file source (reading that file), a USRP sink (fed from the file
source), a USRP source, and a file sink (saving the samples from the
USRP source to another file).

Then use benchmark_rx with the --from-file option to read in these saved
samples.

Best regards,
Marcus

On 21.03.2016 11:17, Diyar Muhammed wrote:
> Dear Marcus,
> Thank you very much indeed for fast replying.
> I look at tx benchmark help I could not find rates but there is packet
> size and megabytes to transmit.
> so that, which one do you mean packet size or megabytes?
> it is okay to use USRP B210 for transmitting and receiving by using to
> benchmark file?
> because when I used one of them (tx or rx) and then I wanted to run
> another one the error come up (no device found for empty device address).
> in advance many thanks.
>
> On Mon, Mar 21, 2016 at 9:57 AM, Marcus Müller
> > wrote:
>
> Diyar,
>
> with the benchmark_ scripts, you **set** the rates, and you can
> only observe how many packets were successfully transmitted.
> The rest is really very basic math.
>
> Best regards,
> Marcus
>
>
> On 21.03.2016 10:50, Diyar Muhammed wrote:
>> Dear SangHyuk,
>> I would like to know how to measure Throughput and BER by using
>> benchmark tx and rx?
>> could you show or explain with real example as you used.
>> in advance thanks.
>>
>> On Mon, Mar 21, 2016 at 8:09 AM, Marcus Müller
>> > wrote:
>>
>> Hi,
>>
>> On 21.03.2016 01 :37, SangHyuk Kim wrote:
>> > I want to know other user's performance (avg performance).
>> Yes, but what is "user's performance"? Is it more important
>> to have
>> higher throughput, or lower error rates? What about robustness?
>>
>> I mean, the OFDM rx_benchmark is a really static example.
>> You might find a setting that maximizes troughput for a given
>> channel,
>> but imagine something happens that reduces your receiver's
>> SNR by 3dB:
>> Now your suddenly losing a lot of performance.
>>
>> Really "how can I parameterize this" can only be answered for
>> a single,
>> mathematically well-defined target, and for a well-defined
>> channel.
>>
>> In a real-world scenario, if using a transceiver with a fixed
>> modulation, you usually wouldn't maximize throughput for a given
>> setting, but you would define what "it still works
>> sufficiently" means,
>> and then you'd define "the worst channel I want the system to
>> still work
>> sufficiently".
>> Then you'd come up with a metric that gives you a number for
>> "the link
>> quality on all considerable channels where this should be
>> working", and
>> then you'd try to maximize that metric under the outage
>> constraints set
>> before. Notice that this metric has to take things like error
>> rate,
>> throughtput, the "cost" of re-sending something (if you have
>> a mechanism
>> for that), available channel coding, how much you care about
>> latency,
>> computational complexity (that really gets important with
>> iterative
>> channel decoding),
>>
>> In other words:
>> This is digital communications. If there was a single "best"
>> solution,
>> we'd all be using that and be done. Use your digital
>> communications
>> knowledge to analyze your requirements and challenges!
>>
>> Best regards
>> Marcus
>>
>> ___
>> Discuss-gnuradio mailing list
>> Discuss-gnuradio@gnu.org 
>> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>>
>>
>>
>>
>> -- 
>> Regards,
>> Diyar Muhammed
>> Ministry of Higher Education and Scientific Research
>> Duty: Network Administration and Design
>> Website:  www.mhe-krg.org 
>> Cell Phone: 009647504690060
>> Office Phone:   00964662554683
>
>
>
>
> -- 
> Regards,
> Diyar Muhammed
> Ministry of Higher Education and Scientific Research
> 

Re: [Discuss-gnuradio] OFDM benchmark optimal parameter

[Diyar Muhammed]

Dear Marcus,
Thank you very much indeed for fast replying.
I look at tx benchmark help I could not find rates but there is packet size
and megabytes to transmit.
so that, which one do you mean packet size or megabytes?
it is okay to use USRP B210 for transmitting and receiving by using to
benchmark file?
because when I used one of them (tx or rx) and then I wanted to run another
one the error come up (no device found for empty device address).
in advance many thanks.

On Mon, Mar 21, 2016 at 9:57 AM, Marcus Müller 
wrote:

> Diyar,
>
> with the benchmark_ scripts, you **set** the rates, and you can only
> observe how many packets were successfully transmitted.
> The rest is really very basic math.
>
> Best regards,
> Marcus
>
>
> On 21.03.2016 10:50, Diyar Muhammed wrote:
>
> Dear SangHyuk,
> I would like to know how to measure Throughput and BER by using benchmark
> tx and rx?
> could you show or explain with real example as you used.
> in advance thanks.
>
> On Mon, Mar 21, 2016 at 8:09 AM, Marcus Müller 
> wrote:
>
>> Hi,
>>
>> On 21.03.2016 01:37, SangHyuk Kim wrote:
>> > I want to know other user's performance (avg performance).
>> Yes, but what is "user's performance"? Is it more important to have
>> higher throughput, or lower error rates? What about robustness?
>>
>> I mean, the OFDM rx_benchmark is a really static example.
>> You might find a setting that maximizes troughput for a given channel,
>> but imagine something happens that reduces your receiver's SNR by 3dB:
>> Now your suddenly losing a lot of performance.
>>
>> Really "how can I parameterize this" can only be answered for a single,
>> mathematically well-defined target, and for a well-defined channel.
>>
>> In a real-world scenario, if using a transceiver with a fixed
>> modulation, you usually wouldn't maximize throughput for a given
>> setting, but you would define what "it still works sufficiently" means,
>> and then you'd define "the worst channel I want the system to still work
>> sufficiently".
>> Then you'd come up with a metric that gives you a number for "the link
>> quality on all considerable channels where this should be working", and
>> then you'd try to maximize that metric under the outage constraints set
>> before. Notice that this metric has to take things like error rate,
>> throughtput, the "cost" of re-sending something (if you have a mechanism
>> for that), available channel coding, how much you care about latency,
>> computational complexity (that really gets important with iterative
>> channel decoding),
>>
>> In other words:
>> This is digital communications. If there was a single "best" solution,
>> we'd all be using that and be done. Use your digital communications
>> knowledge to analyze your requirements and challenges!
>>
>> Best regards
>> Marcus
>>
>> ___
>> Discuss-gnuradio mailing list
>> Discuss-gnuradio@gnu.org
>> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>>
>
>
>
> --
> Regards,
> Diyar Muhammed
> Ministry of Higher Education and Scientific Research
> Duty: Network Administration and Design
> Website:  www.mhe-krg.org
> Cell Phone: 009647504690060
> Office Phone:   00964662554683
>
>
>


-- 
Regards,
Diyar Muhammed
Ministry of Higher Education and Scientific Research
Duty: Network Administration and Design
Website:  www.mhe-krg.org
Cell Phone: 009647504690060
Office Phone:   00964662554683
___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] OFDM benchmark optimal parameter

[Marcus Müller]

Diyar,

with the benchmark_ scripts, you **set** the rates, and you can only
observe how many packets were successfully transmitted.
The rest is really very basic math.

Best regards,
Marcus

On 21.03.2016 10:50, Diyar Muhammed wrote:
> Dear SangHyuk,
> I would like to know how to measure Throughput and BER by using
> benchmark tx and rx?
> could you show or explain with real example as you used.
> in advance thanks.
>
> On Mon, Mar 21, 2016 at 8:09 AM, Marcus Müller
> > wrote:
>
> Hi,
>
> On 21.03.2016 01 :37, SangHyuk Kim wrote:
> > I want to know other user's performance (avg performance).
> Yes, but what is "user's performance"? Is it more important to have
> higher throughput, or lower error rates? What about robustness?
>
> I mean, the OFDM rx_benchmark is a really static example.
> You might find a setting that maximizes troughput for a given channel,
> but imagine something happens that reduces your receiver's SNR by 3dB:
> Now your suddenly losing a lot of performance.
>
> Really "how can I parameterize this" can only be answered for a
> single,
> mathematically well-defined target, and for a well-defined channel.
>
> In a real-world scenario, if using a transceiver with a fixed
> modulation, you usually wouldn't maximize throughput for a given
> setting, but you would define what "it still works sufficiently"
> means,
> and then you'd define "the worst channel I want the system to
> still work
> sufficiently".
> Then you'd come up with a metric that gives you a number for "the link
> quality on all considerable channels where this should be
> working", and
> then you'd try to maximize that metric under the outage
> constraints set
> before. Notice that this metric has to take things like error rate,
> throughtput, the "cost" of re-sending something (if you have a
> mechanism
> for that), available channel coding, how much you care about latency,
> computational complexity (that really gets important with iterative
> channel decoding),
>
> In other words:
> This is digital communications. If there was a single "best" solution,
> we'd all be using that and be done. Use your digital communications
> knowledge to analyze your requirements and challenges!
>
> Best regards
> Marcus
>
> ___
> Discuss-gnuradio mailing list
> Discuss-gnuradio@gnu.org 
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>
>
>
>
> -- 
> Regards,
> Diyar Muhammed
> Ministry of Higher Education and Scientific Research
> Duty: Network Administration and Design
> Website:  www.mhe-krg.org 
> Cell Phone: 009647504690060
> Office Phone:   00964662554683

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

Re: [Discuss-gnuradio] OFDM benchmark optimal parameter

[Diyar Muhammed]

Dear SangHyuk,
I would like to know how to measure Throughput and BER by using benchmark
tx and rx?
could you show or explain with real example as you used.
in advance thanks.

On Mon, Mar 21, 2016 at 8:09 AM, Marcus Müller 
wrote:

> Hi,
>
> On 21.03.2016 01:37, SangHyuk Kim wrote:
> > I want to know other user's performance (avg performance).
> Yes, but what is "user's performance"? Is it more important to have
> higher throughput, or lower error rates? What about robustness?
>
> I mean, the OFDM rx_benchmark is a really static example.
> You might find a setting that maximizes troughput for a given channel,
> but imagine something happens that reduces your receiver's SNR by 3dB:
> Now your suddenly losing a lot of performance.
>
> Really "how can I parameterize this" can only be answered for a single,
> mathematically well-defined target, and for a well-defined channel.
>
> In a real-world scenario, if using a transceiver with a fixed
> modulation, you usually wouldn't maximize throughput for a given
> setting, but you would define what "it still works sufficiently" means,
> and then you'd define "the worst channel I want the system to still work
> sufficiently".
> Then you'd come up with a metric that gives you a number for "the link
> quality on all considerable channels where this should be working", and
> then you'd try to maximize that metric under the outage constraints set
> before. Notice that this metric has to take things like error rate,
> throughtput, the "cost" of re-sending something (if you have a mechanism
> for that), available channel coding, how much you care about latency,
> computational complexity (that really gets important with iterative
> channel decoding),
>
> In other words:
> This is digital communications. If there was a single "best" solution,
> we'd all be using that and be done. Use your digital communications
> knowledge to analyze your requirements and challenges!
>
> Best regards
> Marcus
>
> ___
> Discuss-gnuradio mailing list
> Discuss-gnuradio@gnu.org
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>



-- 
Regards,
Diyar Muhammed
Ministry of Higher Education and Scientific Research
Duty: Network Administration and Design
Website:  www.mhe-krg.org
Cell Phone: 009647504690060
Office Phone:   00964662554683
___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] OFDM benchmark optimal parameter

[Marcus Müller]

Hi,

On 21.03.2016 01:37, SangHyuk Kim wrote:
> I want to know other user's performance (avg performance).
Yes, but what is "user's performance"? Is it more important to have
higher throughput, or lower error rates? What about robustness?

I mean, the OFDM rx_benchmark is a really static example.
You might find a setting that maximizes troughput for a given channel,
but imagine something happens that reduces your receiver's SNR by 3dB:
Now your suddenly losing a lot of performance.

Really "how can I parameterize this" can only be answered for a single,
mathematically well-defined target, and for a well-defined channel.

In a real-world scenario, if using a transceiver with a fixed
modulation, you usually wouldn't maximize throughput for a given
setting, but you would define what "it still works sufficiently" means,
and then you'd define "the worst channel I want the system to still work
sufficiently".
Then you'd come up with a metric that gives you a number for "the link
quality on all considerable channels where this should be working", and
then you'd try to maximize that metric under the outage constraints set
before. Notice that this metric has to take things like error rate,
throughtput, the "cost" of re-sending something (if you have a mechanism
for that), available channel coding, how much you care about latency,
computational complexity (that really gets important with iterative
channel decoding),

In other words:
This is digital communications. If there was a single "best" solution,
we'd all be using that and be done. Use your digital communications
knowledge to analyze your requirements and challenges!

Best regards
Marcus

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

Re: [Discuss-gnuradio] OFDM benchmark optimal parameter

[SangHyuk Kim]

Hi,

I want to know other user's performance (avg performance).

When I operate ./benchmark_tx.py -f 1.5G -m qpsk -W 250
--fft-length=256 --tx-gain=31.5, it results up to 60Mbps. (USRP N210, CBX
40MHz, ANT500)

I empirically got fine parameters and generally (small fft length, high
bandwidth, high tx-gain, big occupied carriers, high bits per symbol
modulation) show better performance (throughput and BER)

Thanks.


2016-03-21 3:22 GMT+09:00 Marcus Müller :

> Trial and error. It depends on your environment, what you want to test,
> and what you want to achieve.
> Best regards,
> Marcus
>
> On 19.03.2016 14:15, SangHyoeg Kim wrote:
> > Hi, I am using ofdm/benchmark example with USRP N210 and CBX 40Mhz
> daughterboard.
> >
> > I tried to know optimal parameter(bandwidth, fft length, occupied tones,
> packet size) for maximum throughput and minimum error rates.
> >
> > How can I reach maximum throughput by adjusting parameter or modifying
> code ?
> >
> > Thanks.
> > ___
> > 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] OFDM benchmark optimal parameter

[Marcus Müller]

Trial and error. It depends on your environment, what you want to test,
and what you want to achieve.
Best regards,
Marcus

On 19.03.2016 14:15, SangHyoeg Kim wrote:
> Hi, I am using ofdm/benchmark example with USRP N210 and CBX 40Mhz 
> daughterboard. 
>
> I tried to know optimal parameter(bandwidth, fft length, occupied tones, 
> packet size) for maximum throughput and minimum error rates. 
>
> How can I reach maximum throughput by adjusting parameter or modifying code ?
>
> Thanks. 
> ___
> 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] OFDM benchmark optimal parameter

[SangHyoeg Kim]

Hi, I am using ofdm/benchmark example with USRP N210 and CBX 40Mhz 
daughterboard. 

I tried to know optimal parameter(bandwidth, fft length, occupied tones, packet 
size) for maximum throughput and minimum error rates. 

How can I reach maximum throughput by adjusting parameter or modifying code ?

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