125 kbps PHY Rx Sensitivity Not As Low As It Should Be

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

Dmitry — Fri, 20 Nov 2020 09:11:14 GMT

I would believe the online calculator from Bluetooth SIG that shows 7 dBm difference.

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

GaryS — Thu, 19 Nov 2020 22:21:05 GMT

OK, so are you happy with the sensitivity for LE125kbps being only 7 dB less than it is for LE1M? If not, what do you think it should be?

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

Dmitry — Thu, 19 Nov 2020 22:07:39 GMT

[quote userid="75815" url="~/f/nordic-q-a/68553/125-kbps-phy-rx-sensitivity-not-as-low-as-it-should-be/280994#280994"]By bit rate, I mean information rate - the rate at which uncoded bits are sent over the channel[/quote]

The radio don't agree with you :) it just works at 1M. We could consider 1/8x bitrate (and expect +9 dBm) in case if we transmit each symbol 8x longer, use 1/8x narrower band filter and 8x more accurate clock.

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

GaryS — Thu, 19 Nov 2020 21:50:45 GMT

By bit rate, I mean information rate - the rate at which uncoded bits are sent over the channel. It doesn't include redundancy for FEC or the bandwidth expansion from 125 kbps to 1 Mbps.

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

Dmitry — Thu, 19 Nov 2020 21:12:45 GMT

[quote userid="75815" url="~/f/nordic-q-a/68553/125-kbps-phy-rx-sensitivity-not-as-low-as-it-should-be/280980#280980"] The bit rate is fb = 125 kbps for LE125K and 1 Mbps for LE1M[/quote]

No, the bitrate is 1Mbps both for 1M and 125K. We have not a reducing of bitrate AND redundant encoding - we have a reducing of bitrate BECAUSE OF redundant encoding. The upper-layer encoding methods are quite different to consider them as simple lowering of bitrate. At physical layer, we have 5 dB between 1M and 125K, plus some gain given by convolutional encoding - I don't know the formula but it should include at least time on-air (crucial for CRC-protected 1M but almost irrelevant for coded stream).

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

GaryS — Thu, 19 Nov 2020 20:52:49 GMT

One more thing ... there may be a bug in the Matlab simulation code mapping EbN0 to SNR. Something doesn't look quite right in there. And 14 dB sounds like it could be too high. I will have a look and get back to you.

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

GaryS — Thu, 19 Nov 2020 19:42:51 GMT

Oops - I should have said "This shows that the sensitivity for LE125K should be 14 dB better than LE1M. ".

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

GaryS — Thu, 19 Nov 2020 19:41:30 GMT

Thanks for sending this simulation link Dmitry - this is great and it helps me make my point.

The plot below from the simulation shows that you get a 1e-3 BER at an Eb/N0 = 7 dB for LE1M, and 2 dB for LE125K. The Eb/N0 is the receiver's received energy per bit period (Eb) divided by the receiver's noise power spectral density. We can convert this to signal to noise ratio (SNR = signal power divided by noise power) by noting that SNR = (Eb/Tb) / (N0 * f0) = Eb/N0 * fb/f0, where Tb = 1/fb is the bit period in seconds, and f0 is the receiver's noise bandwidth in Hz. In dB, you get SNR (dB) = Eb/N0 (dB) + 10*log10(fb/f0). You can read about this relationship here. The bit rate is fb = 125 kbps for LE125K and 1 Mbps for LE1M. So the delta at BER = 1e-3 between the two schemes is 5 dB in Eb/N0, but it's 5 + 10*log(1M/125k) = 14 dB for SNR! This shows that the sensitivity for LE125K should be LE1M. The fact that it's not means the receiver is implemented sub-optimally.

Gary

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

Dmitry — Thu, 19 Nov 2020 16:37:43 GMT

Hi Gary,

I'm not an expert in signal processing, but I don't think you can make a direct assumption from bitrate factor. Coded PHY uses exactly the same physical channel bitrate as 1M PHY - 1 mbit/s, so you get the same BER/SNR curve after GFSK demodulation. First stage (500k) is a convolutional encoder that let us survive with higher level of error rates (for uncoded, any error in a single bit means packet is completely lost). I couldn't find a right method to compare sensitivity for these two cases. This simulation shows an 1dBm gain, but it considers only physical layer. Anyway it would be interesting to simulate a decoder with a noisy channel to see which BER of underlying channel will match, say, 10^-3 after decoding.
At second stage (125k), each symbol is encoded with 4 bits - here you can expect a direct increase of sensitivity up to 6 dBm as by lowering bitrate, also it helps to suppress interference from adjacent channels.

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

GaryS — Thu, 19 Nov 2020 13:45:23 GMT

I'm working on a healthcare IoT application, where there are lots of tags in a hospital running around that are beaconing at 125 kbps. In the long run, the hospital networking equipment will have BT receivers built-in (along with WiFi, typically). But before that happens, hospitals will need to use an overlay network of gateways to pick up the BT transmissions. I say overlay because most hospitals already have WiFi APs now, but they're not yet equipped with BT of any sort. Asking a hospital to purchase and install a second overlay network of BT gateways is OK if there are only 2-4 gateways per floor. But a 6 dB drop in sensitivity could easily quadruple the number of gateways, making our overall business case less compelling.

Your technical description of why you're giving up the 5 dB still isn't clear. If you use the SAME EXACT detection scheme for LR as you do for 1 mbps, you would gain 9 dB in sensitivity - this is because the data rate is 8 x slower, and 10*log10(8) = 9 dB. So why don't you just do that? Then you'll get another 3 dB from the coding gain. Your discriminator and selectivity vs. sensitivity tradeoff makes no sense to me. Discriminator-based detection is known to be 3 dB worse than coherent non-discriminator based detection, and could explain a 3 dB loss in performance somewhere, but it doesn't answer my question above about using the same exact detection scheme. Both the discriminator and sensitivity vs. selectivity comments sound like they came from someone who doesn't really understand how the Rx demod works throwing out buzzwords to try and appease an upset customer. This is a really important issue for me and any other customer that wants long range to really be long range. And therefore it's an important issue for Nordic. So I think it's worth taking the time to answer this correctly. Should probably even have a white paper on this.

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

run_ar — Thu, 19 Nov 2020 10:39:52 GMT

I am sorry you are disappointed with the range improvement. May I ask what type of application you are working on?

Comparing 1mbit to Long Range we are "gaining" 3 dB for 1M compared to the 12 db improvement for LR stated by the BT sig. This is because we where already using a discriminator type receiver for BTLE1M and are therefor missing out on that improvement. The rest of the reduction is unfortunately a trade off between selectivity and sensitivity, which is something we will try to improve in future products. But our expectation is that BTLE125kbits would be up to 9 dB better than 1mbits.

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

GaryS — Wed, 18 Nov 2020 15:54:59 GMT

Sorry - one more comment. I'm not trying to be a jerk about this - sorry if I'm coming across that way. I was just really surprised and disappointed to see that what I thought was going to be a 12 dB range increase turned out to only be a 7 dB increase. I'm working on a very range sensitive IoT application.

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

GaryS — Wed, 18 Nov 2020 15:17:08 GMT

I'm personally an expert when it comes to digital communications and signal processing, so I'm not buying your answer - sorry. For example, if you can do -96 dBm at 1 Mbps, it would be trivial to lower the sensitivity by 9 dB just decreasing the data rate to 125 kbps - without any coding. It doesn't matter what extra steps you took to achieve -96 dBm at 1 mbps. Can you please maybe ask an expert to help you come up with a better explanation for this? There has to be some detailed reason for the shortfall that would make sense to someone like me who's experienced with this kind of stuff, e.g., your the preamble can't be reliably detected below -103 dBm even though the data can be reliably decoded down to -108 dBm (if this happened it could be a protocol design issue - a bad one). Another example explanation would be you're using a phase coherent demod or an equalizer of some sort for 1 Mbps, but haven't had the time yet to develop it for 125 kbps. I'm trying to figure out two things here: (1) what's the specific reason for the 5 dB shortfall, and (2) can it be fixed in the future?

RE: 125 kbps PHY Rx Sensitivity Not As Low As It Should Be

run_ar — Wed, 18 Nov 2020 14:54:01 GMT

Hi,

Unfortunately, BTLR125kb does not categorically add 12 dB of sensitivity. For 1 mbit we where already applying some methods to improve the range, so we do not get the theoretically maximum improvement compared to our 1mbit performance. Regardless you can find more details on BTLR here: https://devzone.nordicsemi.com/nordic/nordic-blog/b/blog/posts/testing-long-range-coded-phy-with-nordic-solution-it-simply-works-922075585