(Not) Using Bluetooth to Measure distance.

An often asked question in here, using RF stuff to measure distances, do localisation etc.

The UK Government have been working on a contact traceing application, but its been delayed quite a bit, this comment was on the BBC website.

Bluetooth to estimate distance.jpg

So problems even 'estimating' if another device is within 2 metres.

Bluetooth to estimate distance.jpg

So, they have taken this long to work out how good an idea that is? No wonder there are 40 000 dead there. Perhaps the British government should have placed a couple of highly-paid boffins on the Arduino Forum as wet-behind-the-ears newbies.

So I carried out a simple test.

I had written RSSI checkers for 2.4Ghz LoRa radio devices, which is the same frequency as Bluetooth.

With a transmiiter and receiver 2m apart and being held as if two people where meeting together (i.e. not talking but staring at thier mobiles) the average signal strength was -68.95dBm, averaged over 100 transmit and receive iterations each about 500mS apart.

By turning around so that the person holding the reciever had it the same 2m apart as before, but now their body blocked the view of the other device, the average signal level dropped to -78.28dBm.

So if two people who were holding phones (with Bluetooth devices) were facing each other 2m apart just turn around, then without moving they are now 10 times further apart, so 20m apart.

So despite what the many sceptics say, you can measure distance with Bluetooth, as long as the error of 1000% is acceptable.

Test carried out in the middle of a small garden maybe 8m square.

Nick_Pyner:
No wonder there are 40 000 dead there.

I have no particular faith in the UK Government but I object strongly to that grossly over simplified statement.

The only way to prevent the vulnerable from dying is to keep them absolutely isolated from the virus and I can't think of any practical way to achieve that. Unlike sterile medical equipment you cannot keep a human in a plastic bag. Things will only get back to normal (like in summer 2019) when everyone has been exposed to the virus. Any other outcome is just a pipe dream.

The real problem is that the UK Government gave people the impression it was omnipotent and could protect everyone. The courageous thing would have been to admit that they were essentially powerless in the same way that they would be if a tsunami occurred.

Of course they could have taken the advice of experts over the past few years and been better prepared with enough hospital beds and equipment so that the lockdown and its associated economic catastrophe could have been avoided. £10bn spent in 2018 or 2018 would be small beer compared to the cost of the lockdown.

...R

Robin2:
The only way to prevent the vulnerable from dying is to keep them absolutely isolated from the virus and I can't think of any practical way to achieve that.

Some proper and prompt effort in that direction, like other countries did, would never have gone astray. The real problem is that the UK government has demonstrated a bumbling incompetence, and they have 40 000 dead to prove just how incompetent they are, and the fact they are still fartarsing around with a bluetooth app is further proof - hence my comment.

I imagine there may be tens of millions around the world who have lost family, friends, and lovers due to those clowns. Grossly over-simplified indeed.

Nick_Pyner:
Some proper and prompt effort in that direction, like other countries did, would never have gone astray. The real problem is that the UK government has demonstrated a bumbling incompetence, and they have 40 000 dead to prove just how incompetent they are,

If you can provide some details about how deaths might be minimised over the long term I would be very interested to know. I don't believe it is possible, however unpleasant reality may be for surviving friends and relatives.

Of course Britain, being an island, could have imposed total isolation immediately that the virus was reported in China. But that would have meant that no person, not even a UK national, could enter Britain. How could one ensure that the virus could not enter the country via imported food or goods? And when might that state of isolation end?

...R

Returning the the subject of the Original Post, people on this Forum have for a long time been advising others that distance measurements using RSSI are completely unreliable. It's rather a pity that none of the experts thought of consulting us.

In any case I certainly don't want some "big brother" system that keeps track of my movements. I'm quite happy to manage my own health risk and I consider that another step towards a police state would be a much too high price for any reduction in my risk of dying - even if it was to protect me from a serial killer.

...R

Robin2:
It's rather a pity that none of the experts thought of consulting us.

Indeed, I am just stunned at the ineptitude of so called 'experts' that expected to be able to make it work reliably.

The simple test I carried out, described in post#2 took me maybe 30 minutes to demonstrate that even under very controlled conditions, the 'distance' variations can be huge. Then when you try to factor in variations of device orientation, RSSI accuracy of individual devices, differences in location such as open fields, gardens, in shops, in homes or on public transport, you start to laugh.

If the purpose behind all this was say to sell advertising when you maybe walked past a particular shop, then it could be acceptable.

But the idea was that based on these so called measurements, you might get a call asking you to stay indoors for 2 weeks and potentially not be paid by your employer.

Robin2:
If you can provide some details about how deaths might be minimised over the long term I would be very interested to know. I don't believe it is possible, however unpleasant reality may be for surviving friends and relatives.

And that is the nub of the "flattening the curve" process. :grinning:

The concept is to reduce the deathrate somewhat by making available sufficient ICU beds and resources so that the best possible treatment is available where someone is severely affected by the infection, supposing that treatment will actually reduce - by some amount - the risk of dying. This will not affect the risk at all for those elderly in Residential Aged Care Facilities and Nursing Homes who become ill but are not considered appropriate for ICU care.

The side issue is the proposition that a usable vaccine will at some time become available which will then permit the remainder who did not become affected, to avoid the worst effects of infection, given that vaccines never totally prevent infection but generally reduce the severity of same. :cold_sweat:

srnet:
But the idea was that based on these so called measurements, you might get a call asking you to stay indoors for 2 weeks and potentially not be paid by your employer.

This is probably nonsense, although, if you are referring to the UK government. it just might be true. Down here in God's Own, we indeed have actually had a phone app in service for a couple of months or so. There is no excuse for dithering with it, you use it for what it can do, rather than try making it do what it can't. When you get past all the silly stories, it is just an additional contact tracer. If your number comes up, you get advised that you may have been exposed, you should get tested, and where and how to that. That's all.

On the technical side, it includes what I understand to be a vital criterion that the above posts forget - duration of contact, which I think is 15 minutes. This gives it time to sort out varying signal strengths to see what is of interest and what's not. Nobody claims it is the greatest app but, while BoJo and his mates in Westminster twiddle their thumbs, it has been in service since April, about six million have it, and it has given some results. Indeed, one clear reason why the results have not been that great is that it is just an aid, and other initiatives are also in use. As I understand it, the main purpose of the phone app is tracing who have it but don't know it.

Paul__B:
And that is the nub of the "flattening the curve" process.

The concept is to reduce the deathrate somewhat

Indeed, and if you don't do anything, you land in deep shit, and the economy gets wrecked anyway. Even with our scandals, blame shifting, elitist skullduggery, and mind-numbing incompetence, we have managed to keep the deaths down to about 100 in a population of 25 million. This would suggest that, with a population of 66 million, the UK might reasonably expect 300 dead, and the other 41 700 take quite a bit of explaining.

Robin2:
Of course Britain, being an island, could have imposed total isolation immediately that the virus was reported in China. But that would have meant that no person, not even a UK national, could enter Britain.

Rubbish. When you see what's going on in Wuhan, you let your own citizens in, quarantine them in all those now empty hotels, and shut the gate. How hard can that be?

Hi Nick, I too live in Gods Own, and do not have the app. Why? My phone is on android 5 or something, one version too old to be supported (really?). Am I worried? Not at all, I've always been sceptical of the app, its need for 15 minutes contact to record an exposure is crazy when a casual contact of seconds in say a supermarket can result in an infection. Plus, without having any technical understanding of bluetooth I've been dubious of measuring bluetooth signal strength adequately to use it for distance measurement. The tests by you in post #2 demonstrate that doubt is well founded.

The success of Australia in having minimal deaths is due to the fast response to close our borders and the social distancing which I have been told by our GP has also reduced the flu rate to about 20% of its norm for this time of year.

I do worry though that we are developing no herd immunity and how we will be able to open the gates again without disaster. Lucky for me i do not have to solve that problem.

I got here on an idle whim, having no technical knowledge of bluetooth, so this is my one and only post. Bye

mikep32:
android 5 or something, one version too old to be supported (really?).

Probably not BLE. It seems that, if your phone isn't BLE, you cannot even download it. No explanation given either.
I don't think Android 5 is the problem. Although my Android 4.2.2 doesn't have BLE, I'm pretty sure it was around at the time, and you just needed a more upmarket phone. As I said, it's not the greatest app, but anything is better than nothing. At least it's limitations are recognised, and it has played a part in our success in flattening the curve while they are dying like flies in the Dark Land of Pom.....

mikep32:
Hi Nick, I too live in Gods Own, and do not have the app. Why? My phone is on android 5 or something, one version too old to be supported (really?).

Same on my wife's (Aldi) Phone. She has suggested I buy her a new one, but I missed the recent Really Good Deal at Aldi. Oh well ...

British Government have just announced that they are dropping plans to produce their own app in favour of using the Apple\Google one.

I wonder what plans Apple and Google have for the data they are not collecting ?

srnet:
British Government have just announced that they are dropping plans to produce their own app

I wonder how much they have already spent on it (wasted) ? No doubt some smart sales guy persuaded some clueless politician that his company could do a better job. Hopefully Private Eye will be told.

It's not the first time money has been wasted on UK health IT projects - when will they ever learn?

...R

Quotes from one of the UK Governments ministers just now;

At the Downing Street briefing, Health Secretary Matt Hancock suggested the original plan might have worked had it not been for Apple's restrictions on third-party apps' use of Bluetooth.

"Apple software prevents iPhones being used effectively for contact tracing unless you're using Apple's own technology," he said.

"Our app won't work because Apple won't change that system... and their app can't measure distance well enough to a standard that we are satisfied with."

So Bluetooth cannot measure distance to a high enough standard, what a shocker.