VST and LTE Toolkit

[zenthoef]

I am considering using the PXIe-5645R and the NI LTE Measurement Suite to make LTE cellular measurements on devices during Manufacturing Testing. 

Our LTE cell modules have a test mode that allows them to generate and receive modulated LTE signals with no data in the signals. The modulated signal bandwidths the module can generate and accept are the typical ones: 1.4, 3, 5, 10, 15, and 20 MHz. I'd like to do this testing with either a NI PXIe-5645R or PXIe-5644R and the NI LTE Measurement Suite. 

I'm concerned that the 5645/44R may not have enough instantaneous bandwidth to generate the LTE signals needed for this testing. It seems like the two VSTs I am interested in should have enough bandwidth at 80MHz when I need to generate 20MHz at most, but I'd like to know if anyone has done anything similar before I purchase any of this equipment/software. Has anyone done anything like this? What were the results?

Maybe it will have enough to generate the modulated signal without data, but maybe I'll run into issues if I try to generate signals with data. Any thoughts on this?

A tangential question: what does the 5645 have that the 5644 doesn't? When I look through the specifications I am hard pressed to notice any major differences. Why is the 5645 nearly 10k more?

[Wild_man]

Hi Matt,

This is a fairly broad question and some of the answers will depend on the application needs, but I will do my best to help out. I'm going to make a few assumptions that my answers will depend upon, and I'll try to make that clear everywhere. The first assumption is that your DUT will be a larger assembly that contains a finished LTE module that has been tested to be operational by the manufacturer and that you will be needing to implement higher level functional tests (i.e. will this connect to an LTE network properly). Is that the case? NI LTE Measurement Suite is well suited for physical layer testing of the LTE interface, but is not designed with high-level functional or signaling test in mind.

In case that description is unclear, LTE Measurement Suite is targeted to testing layer 1 of the 7 layer OSI model.

The 80Mhz bandwidth will be sufficient for generating and/or receiving an LTE signal with a 20Mhz channel width. The channel width does not change in regard to the presence or absence of data. I suspect your concern is instead with the data throughput of the module instead of the analog bandwidth. The data throughput is dependent on several system factors, namely the underlying PCIe architecture of the module, chassis, and PXIe controller. PCIe theoretical throughput can vary from 250MB/s for x1 Gen1 PCIe all the way up to 7.88GB/s for x8 Gen3 within NI PXIe systems, so some attention is needed in selecting all components to maximize streaming throughput.

To answer the tangential question, the PXIe-5645 contains a baseband IQ modulator in case direct hardware Inputs and Outputs are needed for IQ data. The PXIe-5644 is more general purpose, cheaper, and takes up one fewer slot in a system, so if baseband IQ is not needed there is no need to go with the PXIe-5645.

Thanks,

[zenthoef]

Hi Alex - 

That is very helpful!

Your assumption that our DUT is a larger assembly that containers a finished LTE module, which has been tested by the LTE module manufacturer is correct. We are making sure that module still works once it has been soldered down and that the path from the module to the antenna is good. What you've said is reassuring because I do not want to do signalling tests in production test, especially since our modules support a non-signalling test mode. 

If we need to do non-signalling testing (like for development testing, not production testing) that involves data, like PER test of some kind, my main limitation won't be the 5644, but rather the PCIe and PXIe architecture, correct?

That makes sense about the difference between the 5644 and 5645, thanks!

[Wild_man]

Hi Matt,

The reasoning for why you're testing makes sense, and I'm glad to hear you don't intend to do signaling tests for production.

I suspect in most cases you would still be able to send plenty of data for non-signaling (PER, etc.) tests over most any PXIe chassis and controller that you would end up using, I just wanted to make sure you were aware of the limitations and where they came from. If you have a specific target in mind for data throughput (now or in the future) you can work with our support team to make sure the specific hardware is capable of that throughput.

Thanks,

[zenthoef]

Excellent. Thank you Alex!

[jkerrigan12]

Hi Alex,

Your assumption that our DUT will be a larger assembly that contains a finished LTE module that has been tested to be operational by the manufacturer and that you will be needing to implement higher level functional tests (i.e. will this connect to an LTE network properly) is exactly true of my testing needs.

Can you recommend any NI hardware (preferably a PXIe module) and software toolkits that could achieve what you described? Most of what I have found so far has been for lower level testing that we are not required to do. 

Thanks

[Dafrizzy]

Hey there,

Most of our RF software suites are designed for lower-level testing. That said, I would definitely recommend reaching out to either our Applications Engineering or Sales Engineering department to discuss your application in more detail to see if we could meet your application needs with our VST and RF software offerings. If you scroll down to the bottom of the page, there is a link to Open a new service request. This will connect you with one of our Applications Engineers who can work with you 1x1 to help determine if NI products are the right fit for you.

Best,

Daniel

[jkerrigan12]

Sounds Good. Thank you!