22 MHz .... 20 MHz

[RF_Buzzer]

I am transmitting a 22 MHz RF Signal using NI RFSG.
I want to receive this 22 MHz RF Signal using NI RFSA.

However NI-RFSA has a Real time Bandwidth of only 20 MHz.
So, I set the Acquistion Span to 20 MHz and started acquiring the data. I find it acquires data sometimes, but fails a lot of times.

Is the Reduced Bandwidth causing problems ??
What are the problems of acquiring a 22 MHz BW signal with a 20MHz RBW Hardware......

Alternately,
I knew NI-SCOPE supports for NI-RFSA supports > 20 MHz RBW.
So, I set the span to 25 MHz... I thought NI-RFSA acquires the data in a single shot..

But, what it actually does is get one acquistion of 20MHz data and next the remaining 5 MHz data... Spectrum itself is shown cut into two separate acquisitions.. ( Rising + Central part of 22MHz signal FIRST ... and Falling part of remaining 3 Mhz LAST...)

How do I solve this problem of acquiring 22 MHz data in a single shot with NI RFSA ??? Also, what are problems of acquiring a 22 MHz BW signal with a 20MHz RBW Hardware ???

Thanks,
RF Buzzer

[Andy_Hinde]

The 5600 RF downconverter has a passband bandwidth of 20 MHz. It does not matter what digitizer is acquiring the data - the data coming out of the 5600 has a real-time bandwidth of 20 MHz. There is no way to acquire 22 MHz of real-time data with the 5660. The problems with acquiring a 22 MHz wide signal with a 20 MHz wide input are that you will attenuate the 1 MHz of data on either end of othe BW.

---

@RF_Buzzer wrote:
I am transmitting a 22 MHz RF Signal using NI RFSG.
I want to receive this 22 MHz RF Signal using NI RFSA.

However NI-RFSA has a Real time Bandwidth of only 20 MHz.
So, I set the Acquistion Span to 20 MHz and started acquiring the data. I find it acquires data sometimes, but fails a lot of times.

Is the Reduced Bandwidth causing problems ??
What are the problems of acquiring a 22 MHz BW signal with a 20MHz RBW Hardware......

Alternately,
I knew NI-SCOPE supports for NI-RFSA supports > 20 MHz RBW.
So, I set the span to 25 MHz... I thought NI-RFSA acquires the data in a single shot..

But, what it actually does is get one acquistion of 20MHz data and next the remaining 5 MHz data... Spectrum itself is shown cut into two separate acquisitions.. ( Rising + Central part of 22MHz signal FIRST ... and Falling part of remaining 3 Mhz LAST...)

How do I solve this problem of acquiring 22 MHz data in a single shot with NI RFSA ??? Also, what are problems of acquiring a 22 MHz BW signal with a 20MHz RBW Hardware ???

Thanks,
RF Buzzer

---

[RF_Buzzer]

Hi Andy,
Thanks for replying.
I had one more doubt regarding the acquisition.

Suppose I found a way of fitting this 22 MHz real time data
into 20 MHz by using Signal Processing methods, Does this remove the problem ??

My real concern is that most Hardware devices have some tolerance levels ...say if RBW is 20 MHz ...Actual RBW is 20MHz +/- 2MHz...
Are there any such tolerance levels which disturb my acquiring of data.... if I try to fit my data of 22MHz to 20 MHz...

Are u aware of any such methods of fitting Higher Bandwidth Data to Lower Bandwidth ???

Thanks,
RF_Buzzer

---

@Andy Hinde wrote:
The 5600 RF downconverter has a passband bandwidth of 20 MHz. It does not matter what digitizer is acquiring the data - the data coming out of the 5600 has a real-time bandwidth of 20 MHz. There is no way to acquire 22 MHz of real-time data with the 5660. The problems with acquiring a 22 MHz wide signal with a 20 MHz wide input are that you will attenuate the 1 MHz of data on either end of othe BW.

---

@RF_Buzzer wrote:
I am transmitting a 22 MHz RF Signal using NI RFSG.
I want to receive this 22 MHz RF Signal using NI RFSA.

However NI-RFSA has a Real time Bandwidth of only 20 MHz.
So, I set the Acquistion Span to 20 MHz and started acquiring the data. I find it acquires data sometimes, but fails a lot of times.

Is the Reduced Bandwidth causing problems ??
What are the problems of acquiring a 22 MHz BW signal with a 20MHz RBW Hardware......

Alternately,
I knew NI-SCOPE supports for NI-RFSA supports > 20 MHz RBW.
So, I set the span to 25 MHz... I thought NI-RFSA acquires the data in a single shot..

But, what it actually does is get one acquistion of 20MHz data and next the remaining 5 MHz data... Spectrum itself is shown cut into two separate acquisitions.. ( Rising + Central part of 22MHz signal FIRST ... and Falling part of remaining 3 Mhz LAST...)

How do I solve this problem of acquiring 22 MHz data in a single shot with NI RFSA ??? Also, what are problems of acquiring a 22 MHz BW signal with a 20MHz RBW Hardware ???

Thanks,
RF Buzzer

---

---

[Andy_Hinde]

Suppose I found a way of fitting this 22 MHz real time data
into 20 MHz by using Signal Processing methods, Does this remove the problem ??

The data has been filtered before it is acquired, so I don't know how you could use Signal Processing here.

My real concern is that most Hardware devices have some tolerance levels ...say if RBW is 20 MHz ...Actual RBW is 20MHz +/- 2MHz...
Are there any such tolerance levels which disturb my acquiring of data.... if I try to fit my data of 22MHz to 20 MHz...

Are u aware of any such methods of fitting Higher Bandwidth Data to Lower Bandwidth ???

Unfortunately I am unaware of any methods to help in this situation.

[RF_Buzzer]

Hi Andy,

Thanks for your reply.

What I meant when in my previous reply was how competitors overcome this problem.
To quote, Agilent Signal Analyzers expand their Actual RBW of 10MHz to a digitized bandwidth of
80 MHz using Signal Processing Methods ?? Can't NI do something like this...

However, u overlooked my other question ....
My real concern is that most Hardware devices have some tolerance levels ...say if RBW is 20 MHz ...Actual RBW is 20MHz +/- 2MHz...Are there any such tolerance levels for the NI-RFSA which disturb my acquiring of data.
What about the tolerance levels for NI RFSA ????

Regards,
RF_Buzzer

---

@Andy Hinde wrote:
Suppose I found a way of fitting this 22 MHz real time data
into 20 MHz by using Signal Processing methods, Does this remove the problem ??

The data has been filtered before it is acquired, so I don't know how you could use Signal Processing here.

My real concern is that most Hardware devices have some tolerance levels ...say if RBW is 20 MHz ...Actual RBW is 20MHz +/- 2MHz...
Are there any such tolerance levels which disturb my acquiring of data.... if I try to fit my data of 22MHz to 20 MHz...

Are u aware of any such methods of fitting Higher Bandwidth Data to Lower Bandwidth ???

Unfortunately I am unaware of any methods to help in this situation.

---

[Andy_Hinde]

Hi,
What I meant when in my previous reply was how competitors overcome this problem.
To quote, Agilent Signal Analyzers expand their Actual RBW of 10MHz to a digitized bandwidth of
80 MHz using Signal Processing Methods ?? Can't NI do something like this...

I am not familiar with what Agilent is doing here. When you use the term RBW are you referring to Resolution Bandwidth or Real-Time Bandwidth?

However, u overlooked my other question ....
My real concern is that most Hardware devices have some tolerance levels ...say if RBW is 20 MHz ...Actual RBW is 20MHz +/- 2MHz...Are there any such tolerance levels for the NI-RFSA which disturb my acquiring of data.
What about the tolerance levels for NI RFSA ????

Yes the passband of the PXI-5600 is spec'd at 20 MHz as that is the flat part of the bandpass filter used, but the rolloff is such that the magnitude reaponse is roughly 10 dB down at 5 MHz above and below the outsides of the 20 MHz flat passband. If you are capturing 22 MHz, you would be 1 MHz outside the passband on either side so you would have some rolloff but perhaps not a whole lot.

The main problem with this is that the software is not designed for this, although you can do it through low level SW. Basically the NI-RFSA SW operates both the PXI-5600 and PXI-5620 together for ease of use and assumes a 20 MHz max BW and operates as such. To do what you are talking about would require separate programming of the PXI-5600 and PXI-5620. I am attaching two examples to help you with this. One is a NI-Tuner (PXI-5600 driver) example used to set the RF Center Frequency and Attenuation of the PXI-5600. The other is a Spectral Measurements Toolkit example which uses just the PXI-5620 digitizer. These examples can be found at:

Start->Programs->National Instruments->NI-Tuner->Examples
Start->Programs->National Instruments->Spectral Measurements->LabVIEW Support->Spectral Measurements Example Folder->SMT Examples for niScope->Interactive Examples

[RF_Buzzer]

Hi Andy,
Thanks for replying.
It was great discussing this with you.

Obviously it RBW means Real Time Bandwidth,
More details on the Agilent Spectrum Analyzers
can be found here...
http://www.agilent.com/about/newsroom/presrel/2004/08jun2004b.html

As a concluding thought, What I would like to address to NI as a whole is that they should think of providing support
for REAL TIME BANDWIDTHs of GREATER than 25 MHz for enabling Designers to work on NI 5660 for Recent 802 Wireless Standards ,,,,

Regards,
Dharmendra Lingaiah

---

@Andy Hinde wrote:
Hi,
What I meant when in my previous reply was how competitors overcome this problem.
To quote, Agilent Signal Analyzers expand their Actual RBW of 10MHz to a digitized bandwidth of
80 MHz using Signal Processing Methods ?? Can't NI do something like this...

I am not familiar with what Agilent is doing here. When you use the term RBW are you referring to Resolution Bandwidth or Real-Time Bandwidth?

However, u overlooked my other question ....
My real concern is that most Hardware devices have some tolerance levels ...say if RBW is 20 MHz ...Actual RBW is 20MHz +/- 2MHz...Are there any such tolerance levels for the NI-RFSA which disturb my acquiring of data.
What about the tolerance levels for NI RFSA ????

Yes the passband of the PXI-5600 is spec'd at 20 MHz as that is the flat part of the bandpass filter used, but the rolloff is such that the magnitude reaponse is roughly 10 dB down at 5 MHz above and below the outsides of the 20 MHz flat passband. If you are capturing 22 MHz, you would be 1 MHz outside the passband on either side so you would have some rolloff but perhaps not a whole lot.

The main problem with this is that the software is not designed for this, although you can do it through low level SW. Basically the NI-RFSA SW operates both the PXI-5600 and PXI-5620 together for ease of use and assumes a 20 MHz max BW and operates as such. To do what you are talking about would require separate programming of the PXI-5600 and PXI-5620. I am attaching two examples to help you with this. One is a NI-Tuner (PXI-5600 driver) example used to set the RF Center Frequency and Attenuation of the PXI-5600. The other is a Spectral Measurements Toolkit example which uses just the PXI-5620 digitizer. These examples can be found at:

Start->Programs->National Instruments->NI-Tuner->Examples
Start->Programs->National Instruments->Spectral Measurements->LabVIEW Support->Spectral Measurements Example Folder->SMT Examples for niScope->Interactive Examples

---

[RF_Buzzer]

Hi Andy,

U forgot to post the examples.
Can u please post these two examples that u r talking of ?

One is a NI-Tuner (PXI-5600 driver) example used to set the RF Center Frequency and Attenuation of the PXI-5600. The other is a Spectral Measurements Toolkit example which uses just the PXI-5620 digitizer.

Thanks,
Dharmendra

---

@Andy Hinde wrote:
Hi,
What I meant when in my previous reply was how competitors overcome this problem.
To quote, Agilent Signal Analyzers expand their Actual RBW of 10MHz to a digitized bandwidth of
80 MHz using Signal Processing Methods ?? Can't NI do something like this...

I am not familiar with what Agilent is doing here. When you use the term RBW are you referring to Resolution Bandwidth or Real-Time Bandwidth?

However, u overlooked my other question ....
My real concern is that most Hardware devices have some tolerance levels ...say if RBW is 20 MHz ...Actual RBW is 20MHz +/- 2MHz...Are there any such tolerance levels for the NI-RFSA which disturb my acquiring of data.
What about the tolerance levels for NI RFSA ????

Yes the passband of the PXI-5600 is spec'd at 20 MHz as that is the flat part of the bandpass filter used, but the rolloff is such that the magnitude reaponse is roughly 10 dB down at 5 MHz above and below the outsides of the 20 MHz flat passband. If you are capturing 22 MHz, you would be 1 MHz outside the passband on either side so you would have some rolloff but perhaps not a whole lot.

The main problem with this is that the software is not designed for this, although you can do it through low level SW. Basically the NI-RFSA SW operates both the PXI-5600 and PXI-5620 together for ease of use and assumes a 20 MHz max BW and operates as such. To do what you are talking about would require separate programming of the PXI-5600 and PXI-5620. I am attaching two examples to help you with this. One is a NI-Tuner (PXI-5600 driver) example used to set the RF Center Frequency and Attenuation of the PXI-5600. The other is a Spectral Measurements Toolkit example which uses just the PXI-5620 digitizer. These examples can be found at:

Start->Programs->National Instruments->NI-Tuner->Examples
Start->Programs->National Instruments->Spectral Measurements->LabVIEW Support->Spectral Measurements Example Folder->SMT Examples for niScope->Interactive Examples

---

[Shea C]

As Andy stated in the previous post....

These examples can be found at:

Start->Programs->National Instruments->NI-Tuner->Examples
Start->Programs->National Instruments->Spectral Measurements->LabVIEW Support->Spectral Measurements Example Folder->SMT Examples for niScope->Interactive Examples


Shea C

[Andy_Hinde]

Sorry. I could have sworn I posted them. I'll try again...

Thanks,
Andy