Agilent Technologies would like to bring to Microwave Journal readers’ attention a recent set of statements that was made by National Instruments during NI week and through videos posted on YouTube. Agilent RF and microwave experts have reviewed the comparisons for ACPR dynamic range and speed, third order inter-modulation distortion, harmonics measurement, and price and believe that the NI statements misrepresent the performance and capabilities of the Agilent PXA Signal Analyzer. We are concerned that Microwave Journal readers who are involved in critical RF and microwave measurements may make decisions based on comparisons that we believe are inaccurate. Our intent is to provide a clarified summary of the inaccurate performance comparisons that were made and invite interested readers to review the additional references mentioned at the end of this statement.

1. Adjacent Channel Power Ratio (ACPR) dynamic range

In the YouTube posted videos, “Head to Head: NI PXIe-5665 Versus Traditional Boxed Instruments”, and NI Week presentation “NI 14 GHz Vector Signal Analyzer” NI makes the following comments about the Agilent PXA Signal Analyzer.
- NI states that like the PXIe-5665, the Agilent PXA achieves “about 80-81 dBc on the upper adjacent channel power measurement”.
- NI also states that they are “utilizing noise correction and also utilizing the maximum dynamic range of both instruments.” They also display a result of -79.1 dBc for the PXA.
- Both statement are INCORRECT:
o Without a good notch filter between the test signal and measurement input channel, NI appears to be measuring the adjacent channel noise of the RF test signal and not the dynamic range of the PXA as NI suggests.
o As shown below, the Agilent PXA can achieve an ACPR dynamic range of typically -88 dBc (nominal, -91.1dBc measured below) in swept mode with noise correction. This measurement is 7-10dB better than NI suggests.



Figure 1 : Measurement with deep notch filter for upper adjacent and alternate on PXA for 10 dB better performance

2. Adjacent Channel Power Ratio (ACPR) Speed

In the “Head to Head” YouTube posted video, NI makes the following comments about the Agilent PXA Signal Analyzer.
- NI claims that the “Agilent PXA takes 444ms to make the measurement” making NI’s PXIe-5665 “14-15x faster than the PXA”
- This statement is INCORRECT:
o The Agilent PXA has a FAST ACPR measurement that can achieve a .2 sigma variance in ~ 14 ms.
o In the Fast ACPR measurement mode, noise correction is not applied and the PXA is still able to achieve a dynamic range of greater than -81 dB (nominal, -82.8 dBc measured below).
o For W-CDMA, the correct terminology is ACLR (Adjacent Channel Leakage Ratio)



Figure 2: Sweep time selected influences averaging. For 0.2 sigma standard deviation (Sweep time is ~7.5 ms) and total meas. Time is 14 ms.

3. Third order Inter-modulation Distortion

In the “Head to Head” YouTube posted video, NI makes the following comments about the Agilent PXA Signal Analyzer.
- NI states that “the PXAs third order distortion product is -88 to -89 dBc” for 2 tones at -10 dBm each
- NI claims that they “can go ahead and calculate the TOI of both instruments” and states that “the third order intercept of the Agilent PXA is +19 or 20 dBm”
- Both of these statements are INCORRECT:
o The Agilent PXA has a measured TOI of >+36 dBm for this wide tone spacing measurement. The Agilent PXA can achieve this specification because the PXA has built in attenuation that can eliminate internal mixer contribution.
- NI shows a measurement of -95 dBc for their PXIe-5665. The video shows -85 dBc. It is not clear to Agilent if NI is confusing a dBc measurement with dBm or if the NI TOI measurement algorithm for calculating TOI is faulty.
- Note that the Agilent PXA has a built in 1-button routine that automates and optimizes TOI measurement to minimize TOI distortion.



Figure 3: PXA One button TOI measurement sets optimal attenuation setting for TOI measurement of +36.7 dBm

4. Harmonics Measurement

In the “Head to Head” YouTube posted video, NI makes the following comments about the Agilent PXA Signal Analyzer.
- NI states that “5665 is almost 1.7 times faster than the Agilent PXA….. for harmonics testing”
- Realistically, harmonics testing is done measuring odd harmonics, even harmonics, or all harmonics.
- In the video, NI selects the arbitrary measurement of the 2nd, 5th, 10th and 14th harmonics for the harmonics measurement comparison. NI suggests that the measurement speed of the PXIe-5665 is 404 ms and the PXA is 720 ms.
- This statement is misleading and requires CLARIFICATION:
• The Agilent PXA has a list sweep mode that was not used when NI made the measurement, which can perform harmonic measurements faster than what was suggested. The Agilent PXA in list sweep mode measures the 2nd, 5th, 10th and 14th harmonics in 233 ms.
• In addition, the Agilent PXA can measure the first 12 harmonics is less than 350 ms.
• The Agilent PXA can perform multiple modulation accuracy measurements like EDGE EVM, Φ and freq error, and measure up to the 3rd harmonic in < 127 ms.
- There is also a fast 1-button harmonics distortion routine automated in the PXA to facilitate the harmonics measurement.



Figure 4: Measurement of first 10 harmonics.

5. Price

In the YouTube posted videos, “Head to Head: NI PXIe-5665 Versus Traditional Boxed Instruments”, and NI Week presentation “NI 14 GHz Vector Signal Analyzer” NI makes the following comments about product pricing comparison to the Agilent PXA Signal Analyzer.
- NI states that “the NI 5665 is a fraction of the cost of Agilent PXA”
- This statement calls for CLARIFICATION:
• The price of a similarly configured 13.6 GHz PXA is $68,210 which is within $6,000 of the equivalent NI PXIe-5665 14 GHz vector signal analyzer configured with a PXI controller and LabView.
• The PXA is a member of the X-series signal analyzers which offer a range of price/performance choices, including the CXA (prices from $12,700), EXA (prices from $17,000), the MXA (prices from $30,000) to the highest performance PXA (prices from $52,000). All the X series have traceable specified RF performance and have the industry’s widest range of measurements applications and demodulation capabilities.

Additional useful reference materials:

1. 8 Hints for better Spectrum Analysis, http://cp.literature.agilent.com/litweb/pdf/5965-7009E.pdf
2. Agilent Spectrum Analyzer basics, http://cp.literature.agilent.com/litweb/pdf/5952-0292.pdf
3. Optimizing Dynamic Range for Distortion Measurements, http://cp.literature.agilent.com/litweb/pdf/5980-3079EN.pdf
4. Optimizing Spectrum Analyzer Amplitude Accuracy, http://cp.literature.agilent.com/litweb/pdf/5968-3659E.pdf
5. X-Series Signal Analyzer Spectrum Analyzer Mode User’s and Programmer’s Reference, http://cp.literature.agilent.com/litweb/pdf/N9060-90027.pdf