LAA CRC Errors for symbols 1-3 on 1st subframe of TXOP

[ctarver]

Hi,

In a previous post, I asked about the CRC not passing all the time for the LAA application example. https://forums.ni.com/t5/LabVIEW-Communications-System/LTE-Wi-Fi-Coexistence-Application-Example-Que...

The reply was that "Regarding the CRC errors: Those might happen as frequency estimation/tracking algorithm can not rely on the LTE frame structure anymore. Hence, it might happen that the CRC becomes wrong for the very first subframe of a TX opportunity. The higher the TXOP value is chosen, the lower the CRC error rate becomes." 

I have been trying to investigate this further. My application only uses 1 subframe TXOP so this CRC/frequency estimation issue is a severe loss of throughput.

When looking at the PDSCH constellation, symbols 1-3 are circular. Symbol 0 should be the PDCCH which looks fine. Symbol 4 onward doesn't have this problem. I imagine this is because symbol 0 has CRS, so the PDCCH can be ok. Then symbol 4 has CRS so that can be ok. I imagine that something about getting 2 consecutive CRS is being used, but I can't find it. 

My understanding is that the equalization is applied by interpolating channel estimation across the subcarriers and holding across the symbols. I don't understand why symbols 1,2, and 3 have an issue if it's just using the CRS from symbol 0. 

There is also some frequency estimation done at the autocorrelation and some done using the CRS. The CRS based frequency estimation is fed back to the sync block.  However, the "enable frequency tracking" control is off, so I don't think this frequency tracking actually has an effect. I can also use the "sync configuration from host" control to set a static cfo. However, even when configuring the sync to be static at 0 Hz  (while TX/RX on 1 USRP), I can't seem to overcome the circular symbols.

Any thoughts? I'm really looking for precise guidance to where in the sync blocks or channel estimation blocks this is happening. My scenario has no mobility, so I should be able to turn some of the tracking off. 

Thanks!

-ctarver

[James_Battleaxe]

Hi,

I'll have to escalate this issue to provide you with the information you're looking for. I'll keep you updated on my progress.

[ctarver]

Thanks, James! I appreciate the help.

[James_Battleaxe]

Hi ctarver,

We are a little short staffed in R&D this week due to travel and vacation schedules. Because of this I probably won't be able to get back to you until early next week. As before, I'll update you as soon as I get a response from R&D.

[James_Battleaxe]

Hi ctarver,

My colleagues have looked into this issue and they're putting together some information now. They will contact you directly to provide you with an explanation.