About the radio frequency parameters and distance of WL33.

[WWang.15]

Dear ST experts.

We are a module manufacturer. We have designed a module based on WL33. We have been using ST's dk board for verification. However, when we tested the distance, the original dk board was about 200 meters. With a 20dbm and -12X sensitivity, we have high expectations for this chip. Do you have ST distance test reports and parameters?
2. If our deviation is <2.5Khz, does it mean that we need to replace the TCXO?

B.R. Weili Wang.

[Saverio GRUTTA]

Hello

In order to understand better the cause of this behaviour I need to have more details on the RF parameters  used (TX power, data rate, frequency deviation and channel filter).

Consider that the max power our kit can deliver is 16 dBm. To reach 20 dBm is necessary a BOM change.

Are you using one kit as transmitter and a second kit as receiver.

Regards

Saverio Grutta

[WWang.15]

Hi Saverio.

We use STM32WL33 original dk board for both TX and RX.
I just asked our engineer, we set a power supply voltage in the GUI,
Then the tx power is measured by spectrum and is about 18-19dbm@CW

RF parameter as below.

background noise as below.

the distance about 160M,

Sent 100 packets and received about 10 packets.

could you give me some suggestions?

Thx.

[Saverio GRUTTA]

Hello Weili,

We started to analyze your problem. First, we checked the sensitivity in conducted mode with the same radio configuration to verify the receiver's performance. The value we measured is -106 dBm, which is in line with our expectations. Then, we tested our kit in radiated mode using the same antenna we provide with the board and measured a noise floor close to what you measured, in the range between -110 to -105 dBm. Considering an SNR of around 8 dB, we expect to start losing packets with an RSSI close to -100 dBm. Indeed, the plot below shows that we start to lose packets with power lower than -97 dBm. This is coherent to me. I would expect to see the same behavior in your side, but you started to lose packets at -84 dBm, which is far from both the noise floor and sensitivity.

Noise floor

Packet received at -97 dBm

We can consider a link budget of around 115 dB. Using the two-ray model as the mathematical model, the expected range of communication should be around 900 meters.

Regarding the transmitter, the correct settings are TX mode = TX HP and SMPS = 1.5V.

Could you repeat the test at a different frequency to verify if the problem is caused by an interferer ?

Best regards

Saverio Grutta

[WWang.15]

Hi Saverio.

I appreciate your reply. Please wait for some time while we test it internally.
In addition, I would like to ask you a few questions.
Q1: Does ST have any reference designs using TXCO crystal oscillators?
Q2: Regarding SNR, does any combination of RF parameters in WL33 need to be at 8db for smooth transmission?
Q3: In the TX POWER part, if I want to set it to 20dbm on the DK board, I know that I need to adjust the SMPS, so my understanding is wrong, right? If my understanding is wrong, how can I adjust it on the WL33 DK board? Output 20 dBm tx power?
Q4: Is your noise floor measured in a shielding room or in the open field?

[Saverio GRUTTA]

Hello Weili,

A1: We have a reference design which uses a TCXO for 169 MHz frequency, 27 dBm with external FEM. Here the link: https://www.st.com/en/evaluation-tools/stdes-wl3c4eew.html

A2: The SNR of 8 dB is to give you an indication of the signal level required at the receiver for good reception. In other words, to achieve good performance, a rule of thumb is to ensure that the signal level at the receiver side is 8 dB higher than the measured noise floor.

A3: The evaluation board is designed and optimized for 16 dBm. If your target is to reach 20 dBm, please consider the following reference design: STDES-WL3C4SHH. You will need to change a few components to achieve the same performance as our reference design. After these modifications you can reach the 20 dBm using an SMPS voltage of 2.4V.

Q4: The measurement of the noise floor was done in an open field.

Best regards

Saverio

[WWang.15]

Hi Saverio.

may I ask the last question?

If we use TCXO, do we need to do additional settings in the STM32WL3 GUI?

[Saverio GRUTTA]

Hello Weili,

If you don't want to control the turn ON and OFF of the TCXO by GPIO, no modifications or additional settings are needed.

Regards

Saverio

[Saverio GRUTTA]

Hello Weili,

To improve communication distance in a very crowded environment, one suggestion is to use DSSS (Direct Sequence Spread Spectrum).

Using Direct Sequence Spread Spectrum (DSSS) in a communication system can indeed improve the link budget. This improvement occurs because DSSS spreads the signal over a wider bandwidth, which helps combat noise and interference, ultimately enhancing the signal-to-noise ratio. By spreading the signal, DSSS allows for a more robust and reliable communication link, especially in environments with high levels of interference.

One key advantage of DSSS is that it enables achieving the sensitivity of a low data rate modulation scheme without reducing the channel bandwidth. This means that even with a wider bandwidth, DSSS can maintain a high level of sensitivity similar to that of a low data rate modulation scheme. This characteristic is particularly beneficial in scenarios where maintaining a higher data rate is essential while still ensuring reliable communication over longer distances.

In summary, the use of DSSS can increase the link budget by improving signal quality, enhancing resistance to interference, and enabling better sensitivity without compromising channel bandwidth.

Regards

Saverio

[WWang.15]

Hi Saverio.

I appreciate your help,
Could you please provide me with a GUI of RF parameters and a test combination with the lowest rate and longest distance?