CC1200: What is the relationship between RX filter bandwidth and IF frequency?

I don't have a good figure for this so I'll try to describe it by words:

When receiving the signal it's mixed down around the IF frequency meaning that the center of the signal is on the IF frequency. You have to ensure that this signal is not filtered by the RX BW filter. In the case with high signal bandwidth (high datarate) the IF has to be set to zero to fit the full signal bandwidth within the the RX BW. 

Hi TER,

Many thanks for your response. Do you mean this RX filter is a low pass filter?

For 2GFSK with symbol rate 50k, deviation 25kHz and channel spacing 200kHz, we pass the adjacent channel rejection with (IF frequency, RX filter bandwidth) = (416kHz, 333kHz) but failed with (555kHz, 416kHz). So I'm wondering what is the relationship between RX filter bandwidth and IF frequency.

Best Regards,

M.L.

First of all: It would save time for both of us if you wrote the background for the question in the first post. 

- With the given datarate/ deviation: What is the background for the RX BW values you have measured with? 

- Which values do you measure for ACP for the two cases? 

I believe I described the filters slightly wrong. The RX BW filter is centered around the IF frequency but should not cross 0 Hz since then it will be filtered by the DC filter.   

Hi TER,

Thanks for your kindly response. We did the test in the outside LAB and we only knew the result pass or fail.

I think I were caring what is the relationship between RX filter bandwidth and IF frequency not the test. For example, if (IF frequency, RX filter bandwidth) = (555kHz, 416kHz), then the pass band of the RX filter is from (555-208)kHz to (555+208)kHz? Or how to calculate the pass band of the RX filter?

And for Zero-IF case, how to decide the RX filter bandwidth?

Best Regards,

M.L.

Yes, simplified what you write will be true: "For example, if (IF frequency, RX filter bandwidth) = (555kHz, 416kHz), then the pass band of the RX filter is from (555-208)kHz to (555+208)kHz?" In reality the filter is not a brick wall and it's not fully symmetrical. 

For the zero IF case the bandwidth will be the same, just around 0 Hz. But for 50 kbps it's no upside of using zero IF. 

- With the given datarate/ deviation: What is the background for the RX BW values you have measured with?