We're running some open air tests on SiLabs EFR32BG13 demo board/dev kits. Using BGTool, we're just advertising on 37/8/9 and recording it with a sympathetic receiver in the correct PHY. The results received were concerning on a few fronts. Primarily, 2 items.
1: The coded PHY seems to have an issue with packet loss when advertising. This has been repeated here in our lab in a multitude of setups/environments.
2. The long range - low power setup i had intended to create instead gave me similar RSSI but particularly poor packet rate across the board. This is counter to my expectations. This will be recreated again when it stops raining.
Transmitter and Receiver: SILabs demo boards, X axis horizontal. Elevation of system, rooftop ~25ft. Sensors on plastic elevations ~4ft tall.
Receiver using BGTools, generic scanning mode with PHY set to match transmitter
5 minute scans.
+10dB
Coded PHY 300ft range
RSSI -79.9
Packet reception rate 40.1%
1M PHY at 300ft range
RSSI -80.8
Packet reception rate 79.4%
+0dB
Coded PHY 300ft range
RSSI -81.7
Packet reception rate 34%
1M PHY at 300ft range
RSSI -80.8
Packet reception rate 39%
*****edited*****
Retested at ½ range
+0dB
Coded PHY 150ft range
RSSI -71.3dB
Packet reception rate 57.9%
1M PHY at 150ft range
RSSI -75.5
Packet reception rate 94%
Bluetooth Low Energy
Discussion Forums
Answered
Answered
Hi,
I am not very clear about your problem. But you can try to test the range or RSSI in a RF clean area. And more you can use the Range test functions in the Simplicity Studio for compare.
Regards,
Jun
0
I have the same problem using BGM13P22 radio module.
Using default 1M PHY advertising or a CODED PHY advertising doesn't change the advertising range.
Is it the expected behaviour ?
Regards,
Marco.
0
Sorry i don't know this well, what coded phy you are using, 125k (S-8) or 500k (S-2).
We're using 125k. EFR32BG13 can't do 500k S-2 to my knowledge.
For clarity, I believe we're doing something wrong. There's no chance that SiLabs would offer a "long range" phy that loses 7/10 packets in an open air test.
Jun,
A RF clean area that is more than 300ft long is a luxury I do not possess. Running the test on our roof with elevated platforms next to a 1M phy test that shows high packet count and stable RSSI is my way of getting as close as I am able. I hope your not asserting that the long range PHY works, but only in a chamber. That would be relatively useless for real world applications.
Can you elaborate on the range test function you described?
0
Delu,
Still hoping to hear about the range test function you mentioned.
0
Is this thing on? Hello?
Repeated test data at 100ft and 10ft.
Coded at 10ft -13 RSSI 60% packets
Coded at 100ft -49 RSSI 60% packets
1M at 10ft -10 RSSI 95% packets
1M at 100ft -49.5 RSSI 96% packets
0
Hi,
You are being supported through our support portal as well. We can continue the discussion there and post the conclusions here for the benefit of the community.
Regards,
Tiago
0
This item has been resolved.
Per SiLabs, the BT5.0 coded PHY uses secondary advertisements. These advertisements contain a pointer to another channel where the actual data can be found. As a result, a ~60% advertisement packet rate is not considered to be a malfunction. This is incompatible with beacon style applications that require a high packet capture rate. When implemented in low level physical RF, the benefits of a coded PHY are clear but they are lost when implemented as a beacon.
Performance of Long Range Coded PHY (125k)
We're running some open air tests on SiLabs EFR32BG13 demo board/dev kits. Using BGTool, we're just advertising on 37/8/9 and recording it with a sympathetic receiver in the correct PHY. The results received were concerning on a few fronts. Primarily, 2 items.
1: The coded PHY seems to have an issue with packet loss when advertising. This has been repeated here in our lab in a multitude of setups/environments.
2. The long range - low power setup i had intended to create instead gave me similar RSSI but particularly poor packet rate across the board. This is counter to my expectations. This will be recreated again when it stops raining.
Transmitter and Receiver: SILabs demo boards, X axis horizontal. Elevation of system, rooftop ~25ft. Sensors on plastic elevations ~4ft tall.
Receiver using BGTools, generic scanning mode with PHY set to match transmitter
5 minute scans.
+10dB
Coded PHY 300ft range
RSSI -79.9
Packet reception rate 40.1%
1M PHY at 300ft range
RSSI -80.8
Packet reception rate 79.4%
+0dB
Coded PHY 300ft range
RSSI -81.7
Packet reception rate 34%
1M PHY at 300ft range
RSSI -80.8
Packet reception rate 39%
*****edited*****
Retested at ½ range
+0dB
Coded PHY 150ft range
RSSI -71.3dB
Packet reception rate 57.9%
1M PHY at 150ft range
RSSI -75.5
Packet reception rate 94%
Hi,
I am not very clear about your problem. But you can try to test the range or RSSI in a RF clean area. And more you can use the Range test functions in the Simplicity Studio for compare.
Regards,
Jun
I have the same problem using BGM13P22 radio module.
Using default 1M PHY advertising or a CODED PHY advertising doesn't change the advertising range.
Is it the expected behaviour ?
Regards,
Marco.
Sorry i don't know this well, what coded phy you are using, 125k (S-8) or 500k (S-2).
https://www.silabs.com/community/wireless/bluetooth/knowledge-base.entry.html/2017/06/15/le_coded_phy_on_efr3-WMeL
Delu,
We're using 125k. EFR32BG13 can't do 500k S-2 to my knowledge.
For clarity, I believe we're doing something wrong. There's no chance that SiLabs would offer a "long range" phy that loses 7/10 packets in an open air test.
Jun,
A RF clean area that is more than 300ft long is a luxury I do not possess. Running the test on our roof with elevated platforms next to a 1M phy test that shows high packet count and stable RSSI is my way of getting as close as I am able. I hope your not asserting that the long range PHY works, but only in a chamber. That would be relatively useless for real world applications.
Can you elaborate on the range test function you described?
Delu,
Still hoping to hear about the range test function you mentioned.
Is this thing on? Hello?
Repeated test data at 100ft and 10ft.
Coded at 10ft -13 RSSI 60% packets
Coded at 100ft -49 RSSI 60% packets
1M at 10ft -10 RSSI 95% packets
1M at 100ft -49.5 RSSI 96% packets
Hi,
You are being supported through our support portal as well. We can continue the discussion there and post the conclusions here for the benefit of the community.
Regards,
Tiago
This item has been resolved.
Per SiLabs, the BT5.0 coded PHY uses secondary advertisements. These advertisements contain a pointer to another channel where the actual data can be found. As a result, a ~60% advertisement packet rate is not considered to be a malfunction. This is incompatible with beacon style applications that require a high packet capture rate. When implemented in low level physical RF, the benefits of a coded PHY are clear but they are lost when implemented as a beacon.