Packet loss due to interference in ble.

RE: Packet loss due to interference in ble.

Kenneth — Tue, 26 Jul 2022 13:20:23 GMT

If you look at the peripheral BLE examples you can find that most of them use gatt_init(), which will upon connection exchange max MTU sizes, you can get the result of the max supported MTU size by looking at the callback handler: gatt_evt_handler() and the events NRF_BLE_GATT_EVT_ATT_MTU_UPDATED (max attribute size) and NRF_BLE_GATT_EVT_DATA_LENGTH_UPDATED (max on air size, a long attribute may be split across several of these). Example of implementation is shown in this reply:
RE: BLE5 - Make payload size 200 bytes - the right way

Typically you want the data sent to be as close as possible to the max data length (NRF_BLE_GATT_EVT_DATA_LENGTH_UPDATED) for most efficient transfers.

In a perfect world with no packet loss having a slow connection interval can give the highest throughput, but considering we don't live in a perfect world I would claim you should have a fast connection interval, this will reduce the max peak throughput, but allow stable throughput.

Best regards,
Kenneth

RE: Packet loss due to interference in ble.

Nabeel Ar — Tue, 26 Jul 2022 13:12:09 GMT

Hi,

From this post I came to know to enable DLE I have to increase the NRF_SDH_BLE_GAP_DATA_LENGTH value, but My NRF_SDH_BLE_GAP_DATA_LENGTH value is already in 251. Does the DLE is enabled already?

I saw the following paragraph in this blog. This suggests keeping the connection interval less in the time of interference. But to achieve higher throughput it suggests having a higher connection interval. What to follow??

"A higher connection interval will give higher throughput if this option is enabled. This is different compared to older Nordic BLE devices and stacks, where you wanted the connection interval to be as short as possible for maximum throughput. One drawback here is that if there is a lot of interference and packets start getting lost. If a packet is lost, the device needs to wait for the next connection event before sending more packets (after resending the lost packet). The wait time can be anywhere from close to zero to the length of the connection interval. In other words, a long connection interval will improve the throughput when the link quality is good, but could reduce the throughput when there is a lot of interference or when the distance between the devices is large."

If this NRF_SDH_BLE_GAP_DATA_LENGTH value doesn't determine the DLE, Then how can I enable it?

RE: Packet loss due to interference in ble.

Kenneth — Mon, 25 Jul 2022 11:11:02 GMT

There is a nice blogpost here that describe how to improve throughput:
https://novelbits.io/bluetooth-5-speed-maximum-throughput/

Have in mind that S112 does not support Data Length Extension (DLE), which means the maximum packet that can be sent on-air is only 23bytes, this will decrease the throughput a lot. I recommend to consider the S113 that support up to 244bytes of data in each packet.

Kenneth

RE: Packet loss due to interference in ble.

Nabeel Ar — Mon, 25 Jul 2022 10:13:34 GMT

Hi,

[quote userid="2111" url="~/f/nordic-q-a/90205/packet-loss-due-to-interference-in-ble/378504"]What throughput do you have here?[/quote]

I assume NRF_SDH_BLE_GAP_EVENT_LENGTH is the factor that determines throughput, It is set to 6.

[quote userid="2111" url="~/f/nordic-q-a/90205/packet-loss-due-to-interference-in-ble/378504"]What latency requirements?[/quote]

The data is sent continuously and every piece of data is crucial. So the Latency is set to 0.

RE: Packet loss due to interference in ble.

Kenneth — Mon, 25 Jul 2022 09:50:38 GMT

I don't really see that will help no. What throughput do you have here? What latency requirements?

Kenneth

RE: Packet loss due to interference in ble.

Nabeel Ar — Fri, 22 Jul 2022 14:06:14 GMT

Can I change the application from BLE to RF proprietary? Will it help to avoid interference?

RE: Packet loss due to interference in ble.

Kenneth — Fri, 22 Jul 2022 12:53:27 GMT

I would say no, but you can double check with the module manufacturer that you have placed the module according to their recommendations.

Kenneth

RE: Packet loss due to interference in ble.

Nabeel Ar — Fri, 22 Jul 2022 12:19:10 GMT

Hi,

[quote userid="2111" url="~/f/nordic-q-a/90205/packet-loss-due-to-interference-in-ble/378085"]So you need to improve the link budget by optimizing the design (e.g. make sure the antenna is properly tuned to 50ohm in the final product casing [/quote]

I'm already using 3rd party module with an antenna designed with the configuration mentioned above. Can anything be done on the firmware side to overcome the issue?

RE: Packet loss due to interference in ble.

Kenneth — Thu, 21 Jul 2022 12:11:15 GMT

[quote user="Nabeell"]Does Channel Hopping has to do anything to make it efficient? Will QoS or Adaptive channel hopping help me to overcome the problem??[/quote]

No, this is simply a matter of link budget and signal level of the interference. So you need to improve the link budget by optimizing the design (e.g. make sure the antenna is properly tuned to 50ohm in the final product casing and tune the output power to 4dBm, adding a PA on the front-end will help (e.g. n21540)). Other than that it's very little one can do).

Kenneth

RE: Packet loss due to interference in ble.

Nabeel Ar — Thu, 21 Jul 2022 06:02:24 GMT

Hi Kenneth,

I'm already using maximum tx power (4 dBm) and the tolerance is 500ppm. The connection interval is 7.5ms (min) and 15 ms (max), which I think is a shorter connection interval. Though I'm getting data loss due to interference.

Does Channel Hopping has to do anything to make it efficient? Will QoS or Adaptive channel hopping help me to overcome the problem??

Thanks in advance.

RE: Packet loss due to interference in ble.

Kenneth — Wed, 20 Jul 2022 11:55:07 GMT

Hello,

Make sure to use maximum output power, and make sure you have got help to tune the antenna for optimal performance, a poorly tuned antenna will reduce the effective output power radiated. If you are using the internal LFCLK (internal RF 32kHz), then make sure you have set the tolerance to 500ppm and that the LFCLK callibrate the RC oscillator. This should ensure that you have a good design with good output power and that the peripheral tolerate timing drift between the two peers.

You can also consider to use shorter connection interval, this will allow more time to send data.

Best regards,
Kenneth