Related
!! FUN APPLICATION, NOT INTENDED TO BE USED IN REAL WORLD AVIATION !!
For all those flight-sim enthusiasts out there: I have created this PFD (primary flight display) which shows the speed, the altitude the coards and the attitude similar to a real PFD display in an airplane cockpit.
Although the application has been created and tested on an HTC HD2, it might work on other devices as well. Internel or external GPS and the HTC G-Sensor are required. It is planned to add a serial communication channel to a PC server component which allows to use this little proggy with MS Flight Simulator displaying the flight parameters so you can get rid of the cockpit panel on your primary display.
Just keep in mind that this is the first version of the program, so there might be *some* improvement potential.
A screenshot of the program is also attached.
Pls let me know what you think about it.
Thanks.
As a pilot I cannot find any sense in this, sorry m8.
Useless for aviation.
Useful might be an HSI-App, an RAIM-availability-check-App, a NOTAM-APP and so on. An artificial horizon on a handheld device... No, not good...
Title correction...
It is just for fun, not intended for real world aviation.
I will change the title so that it becomes more clear.
As an airbus pilot I like it! Lots of scope for further development, keep up the good work !
when I read your discrip, I had something totaly dif pictured in my mind. guess Im falling back on my sail plane lesson days.
I had round face dial and a yaw bubble pictured.
While this is cool, I would prefer to see the old analog style with an altimeter that you can set your start altitude and a yaw bubble, rather than the HUD style. just IMO.
And yes this is just for fun.
Keep it up, I will follow your project with intrest.
cheers
Good Work mate
While some are critical in their opinion I'm pleased with your app.
Keep up the good work mate and for all others who are not so encouraging ...... what do you want for an app a full blown flight simulator, the guys making an effort to develop something and sharing it for a free with us so to help him out at least .... I mean at least encourage him and make some constructive criticisms.
It doesn't cost to be nice and help to motivate others...........
Customize it
Just as a hint for those of you, who would want to use other bitmaps. I have designed the application in a way which allows you to modify or extend it. If you look into the applications program folder, you will find an XML file describing the layout of the display. You may also change or add bitmaps. The bitmaps are all located in the
HTML:
resources
subfolder. If you change the size of the offset of bitmaps, you will have to adjust some values in the XML file as well.
As you can see there all items are drawn in 3d space.
However, I am working on a HSI as suggested before. I will post it later this week.
Known issue:
- After running the application 3 or 4 times, OpenGL is not working properly anymore (not able to create rendering context). This is an issue of OpenGL. You have to restart your device for OpenGL to work again. If anyone has a solution, please let me know.
leihen said:
It is planned to add a serial communication channel to a PC server component which allows to use this little proggy with MS Flight Simulator displaying the flight parameters so you can get rid of the cockpit panel on your primary display.
Thanks.
Click to expand...
Click to collapse
That would be awesome!
Cool app
Certainly would be better than no attitude information in the event of a vacuum and/or elec failure.Even if just to make flying the performance instruments a bit easier.
I would defintely like to see a HSI app possibly with an open (editable)waypoint database.Anyone interested!?
Keep up the good work!
I love it. Thanks so much...
I just wonder if settings could be made permanent until being changed again so as to start the application without my having to select REAL and AVERAGE everytime.
And as a proud HTC HD2 owner, the display could me made much
larger indeed...
Congratulations and thanks again.
This is a fine app with a lot of potential esp. for UL-pilots as an emergency backup if UL-airplane-pfd fails.
Would be nice if indicators are bigger and the layout equals to Flymap, Skymap or MGL-layout
There is enough place for a nice big yaw indicator. Maybe You can get some code from GPS-tracker V2.0 (MooNah) in this forum, it already has a full working PFD incorporated.
Good luck with further devs., when next version is online I will try it in my gyroplane.
I think it's fun!!!!!
next step: build a flightsim for wm where we could use this
You will be in real danger....if you do so.
troed said:
This is a fine app with a lot of potential esp. for UL-pilots as an emergency backup if UL-airplane-pfd fails.
Would be nice if indicators are bigger and the layout equals to Flymap, Skymap or MGL-layout
There is enough place for a nice big yaw indicator. Maybe You can get some code from GPS-tracker V2.0 (MooNah) in this forum, it already has a full working PFD incorporated.
Good luck with further devs., when next version is online I will try it in my gyroplane.
Click to expand...
Click to collapse
As you will easily see, this PFD horizon responds to gravity, that is, to the accelerometers installed in the phone.
It has no gyros, no way to provide PITCH ATTITUDE, WHATSOEVER.
If you keep the phone perfectly vertical and quickly move forward (accelerate) it will SURELY show a pitch up...pretty much the same effects that a stand-by compass shows when changing speed (remember that?...accelerate-north, decelerate-south).
You will have the very same pitch information as the one given by a bottle full
af water suspended from a string, or a can of Coke : and it will KILL YOU as well as your passengers (if any), not to mention people on the ground.
The only usable parts of this FUN APP (as named by the author himself ) in aviation are:
1. True track (not magnetic heading).
2. Altitude (neither QNH, nor QFE related, but of course; unless your flying field happens to be at sea level and actual atmosphere matches the ISA atmosphere parameters exactly). Remember that QNH variations as well as heat will make the GPS readings very different to that of your altimeter, as height of the air column will vary a +/-4% every +/-10 degrees Celsius.
Example. Airplane flying at Flight Level 300. QNH in the area 1023. Terrain at 2000 feet. Atmosphere is ISA+15.
Therefore if we move the 1013.25 setting to 1023 there will be a reading increase from 30.000 feet to 30300 (calibrated altitude, QNH related).
That is 28.300 feet above ground (we start with this as the air column height for the calculation). Since ISA is +15 we add 15/10 times 4% and we have 28300*15/10*4/100= 1698. True altitude is 30300+1698=31988
So we are cruising at FL300, 30.000 feet pressure altitude, 30.300 feet calibrated altitude (QNH related), 31988 above sea level, 29988 feet above ground level.
GPS WILL SHOW 31988 FEET.
3. Ground speed ( neither IAS nor CAS nor EAS nor TAS), as provided by GPS system. Remember that your airplane aerodynamics are related to calibrated speeds....you will stall if too slow ore destroy the airframe if to quick !!!! But you can navigate and make good your track, of course.
Hope I have made myself understood....and that you read this at the earlier !!!
Hi!
It quite very good if the second part under we have the route. I am using the Oziexplore program, it can creates the route, and apear on screen. if this soft applied to that once, could be perfected!
Thanks
great app mate..just for fun...
Nice work thank you. On future can you make like pilot cocpit with all dash boards like in Flith simulator game if this possible. Thank you
leihen said:
!! FUN APPLICATION, NOT INTENDED TO BE USED IN REAL WORLD AVIATION !!
For all those flight-sim enthusiasts out there: I have created this PFD (primary flight display) which shows the speed, the altitude the coards and the attitude similar to a real PFD display in an airplane cockpit.
Although the application has been created and tested on an HTC HD2, it might work on other devices as well. Internel or external GPS and the HTC G-Sensor are required. It is planned to add a serial communication channel to a PC server component which allows to use this little proggy with MS Flight Simulator displaying the flight parameters so you can get rid of the cockpit panel on your primary display.
Just keep in mind that this is the first version of the program, so there might be *some* improvement potential.
A screenshot of the program is also attached.
Pls let me know what you think about it.
Thanks.
Click to expand...
Click to collapse
Indeed as an Ultralight-Gyroplane-Pilot I like this app and since the Horizon-tab in GPS-Tracker by MooNah doesn´t work right now this could be alternative as a a last resort-backup if glass-cockpit instruments (MGL-Avionics) fail ...... (better to have this than nothing at all and I also have Flymap PPC on my HD2 if main Moving map Nav. fails)
This app has great potential since it reads the infos from G-sensor and GPS.
For the future dev. if I might add my wishlist:
1. Add to screen: slide-indicator and compass, Altitude
2. slightly more refined graphics
As a start: BRAVO, hope You have the patience to go on developing this nice app ( like Dunc001, that made in 100s of hours of work a professional weather app - Duncans Weather Panel - see there - that is more reliable than many professional aviation weather briefings).
I bookmarked this thread and hope for further developments by U
Congratulations
like the Android solution ixGyro:
http://www.ixellence.com/index.php?option=com_content&view=article&id=227&Itemid=274&lang=de
JHimmelbauer said:
like the Android solution ixGyro:
http://www.ixellence.com/index.php?option=com_content&view=article&id=227&Itemid=274&lang=de
Click to expand...
Click to collapse
Jepp, EXACTLY THIS would be cool on the HD2 on WinMo (Android not running on mine) !!!!!!!!!!!!!!!!!!!!!!!!!
troed said:
Jepp, EXACTLY THIS would be cool on the HD2 on WinMo (Android not running on mine) !!!!!!!!!!!!!!!!!!!!!!!!!
Click to expand...
Click to collapse
a other solution for Windows Mobile is "Inclinometer77" -
visit http://forum.xda-developers.com/showthread.php?p=6677074
This App goes in the right direction, but unfortunately not to use without Gyro in a Plane, but a nice game for pilots ;-)
Happy Landings
Hello, I am looking into possibilities of Android devices (already existing or coming up in near future) to be used as a replacement for paper + ink signatures. Could this work ?
Here are my main concerns:
1.) Most of the devices nowadays work only with capacitive pens which typically don't have the feel of real pen + ink and also lack the precision which can all result in an altered shape of the signature when compared to paper + ink.
2.) Are there any devices now or in near future that will support pressure recognition ? (E.g. the harder you press the thicker the drawn line)
Since I've seen electronic signature pads in banks and they record actual pixels + the pressure data - the basic idea is if we could do the same on Android device, record pixels (without altering the shape of signature) + record pressure data - it would be as good as what banks use - so in theory it should be strong enough to be used for legal acts (as a full replacement of hand written signature) Or if that's not true then maybe someone with good security knowledge could enlighten me what data would need to be stored (with current state of touchscreen technology it would be good news if e.g. pressure data were not needed after all...).
So far the closest device most suited for this I could find seems to be HTC Flyer but I suspect it doesn't support pressure recognition - is that correct ? Then I also found some rumors about pressure recognition display technology "coming" but it's not obvious to me if it's matter of months or years.
the new Samsung Notes
can do what you want
That looks exciting ! Just to be sure - am I correctly assuming that at the moment there is no Android device on the market that supports pressure recognition / pressure sensitive pens ?
Production of this Android Smart Ring can start soon as soon as some Big Device manufacturer lays his eyes on this ( I mean when he/she gets to know about this project ). Hopefully Google should be attracted towards this
The Android Smart Ring is an inspired device from smart-watches but has an unique Design,Specifications and can do beyond possibilities.
Watch Video HERE
I am an College Student who got this project Idea. I just want to give this project a Go and see if I can make it a success. I wanna give people a whole new experience of Technology. It's a unique Idea which has never been done, if previously it's done then that must be not as good as this one. The Android Smart Ring features a curved 720p display, Bluetooth v4.0, NFC, High Clarity Speakers, Microphone, powered by 1GHz processor, 512MB RAM, couple of required sensors, has Transistors (acting as a battery which requires no charging) and this ring runs on Android KitKat (or) Android Wear (in near future).
Keep Sharing this Idea and let each and every men know about it. Help me making this Idea an reality Thanks for all your support in advance.
I don't think I understand how you intend to incorporate a 720 P display on a small ring along with a whole bunch of other hardware
I don't think we have hardware capabilities to make your idea feasible at this point
Sent from my HTC One using Tapatalk
thesparky007 said:
I don't think I understand how you intend to incorporate a 720 P display on a small ring along with a whole bunch of other hardware
I don't think we have hardware capabilities to make your idea feasible at this point
Sent from my HTC One using Tapatalk
Click to expand...
Click to collapse
What this dude says is true. It's only possible if you created a time travelling machine and travel to the future and retrieve the techology from there, now this is definitely a viable and possible option for you homie.
Okay.. Wait a min. So, you guys think that this isn't possible. Let me tell you one thing, you must have heard about Phonebloks, right ? Was it possible back then ? Hell no, now Google has taken the project and made it possible. So, nothing is impossible
I would think scale down what you are expecting out of it, maybe not 720P display. Start small, and work your way up. See if you can get a working prototype first even with a basic display that might be small, because 720P is quite a big ask for what I would imagine to be a very small device.
Analyse it from every possible angle and lens you can think of. Here are some examples to start you off: Technical ability of construction, marketing, corporate usage, end user.
Here are some examples:
Technical:
Start trying to see if you can get electronic components small enough to even complete something like this first. If you can, what's the limit available today? Maybe you can't get a speaker small enough, so you work around not having that. Work out where your limits are.
Then let's say with marketing: Who are you trying to sell it to? How can you convince them to buy it? Does it provide any perceived value (not necessarily actual value), or will people see it as an expensive paperweight?
Corporate usage: Could you display advertisements on it without being too intrusive from general functions? Could it be used in a corporate setting, add any value to business running?
End user: Will it be comfortable and easy to use throughout every phase including charging, or would it overheat and cause discomfort? Will it last? Can it be adjusted to suit the needs of people (e.g. different finger sizes).
Give this a thought, and work out if you can provide solutions to every aspect, and identify potential problems BEFORE you even hit the market. You wouldn't want to go through extreme difficulty to process and manufacture it, and in the end it flops badly because of overpriced/poor quality, or maybe simply nobody wants one.
thedeejay said:
I would think scale down what you are expecting out of it, maybe not 720P display. Start small, and work your way up. See if you can get a working prototype first even with a basic display that might be small, because 720P is quite a big ask for what I would imagine to be a very small device.
Analyse it from every possible angle and lens you can think of. Here are some examples to start you off: Technical ability of construction, marketing, corporate usage, end user.
Here are some examples:
Technical:
Start trying to see if you can get electronic components small enough to even complete something like this first. If you can, what's the limit available today? Maybe you can't get a speaker small enough, so you work around not having that. Work out where your limits are.
Then let's say with marketing: Who are you trying to sell it to? How can you convince them to buy it? Does it provide any perceived value (not necessarily actual value), or will people see it as an expensive paperweight?
Corporate usage: Could you display advertisements on it without being too intrusive from general functions? Could it be used in a corporate setting, add any value to business running?
End user: Will it be comfortable and easy to use throughout every phase including charging, or would it overheat and cause discomfort? Will it last? Can it be adjusted to suit the needs of people (e.g. different finger sizes).
Give this a thought, and work out if you can provide solutions to every aspect, and identify potential problems BEFORE you even hit the market. You wouldn't want to go through extreme difficulty to process and manufacture it, and in the end it flops badly because of overpriced/poor quality, or maybe simply nobody wants one.
Click to expand...
Click to collapse
Well, I haven't just though about the production and all because I ain't ready to do that, I just provided a concept (or) an Idea and asked people to share it if they liked it and then maybe in the future a company might notice this Idea and bring it to reality. I never thought of bringing this thing to production by myself (coz I ain't a rich guy). Speaking of the speaker, it can be replaced with vibration alert which will be more effective (it can be annoying too) for people to never miss an message alert (or) call alert or whatever notifications you get.
When talking about marketing and corporate usage, well there's always a way to target everyone around you for it. However, it just requires an simple yet effective Idea to present in front of the public and it won't be expensive (deduction of specs to low-end ones would help).
Of course, it will be comfortable and easy to use. It's just same as your usual ring replaced by this Tech things from which you can see what's going on your device without a need to take a look at it by taking out it from your pocket. It's just a device which shows you alerts about your notifications and allows you to send messages directly to your colleague by using microphone (voice messaging service).
I Hope you are satisfied. Feel free to ask more questions, I am happy to assist you and NOTE :- It's just and Idea which I haven't thought to bring in production by myself.
Beyond B78 Fashion Smartphone can't read some character Blackberry Messenger
How do I cope with Blackberry Messenger and unicode autotext chinese letters sometimes appear on smartphones B78 beyond fashion? I 've been dressing droidsansfallback.ttf from 9MB to 22MB in size but nothing matched, whereas the Galaxy smartphone chat droidsansfallback.ttf 13.7 MB in size can read all Unicode characters.
The character I mean for example:
♪ ♫ * ¨ * ❤ * ¨ * ● ๋ • тєηgкソ υ 4 ѕнαяιηg ● ๋ • * ¨ * ❤ * ¨ * ♫ ♪ ˙ · 0 • ● ♥ Ƹ̵̡Ӝ̵̨̄Ʒ ♥ ● • • ● ♥ Ƹ̵̡Ӝ̵̨̄Ʒ ♥ ● • 0 · ˙ ♥ ♠ ѕє ℓ αмαт мα ℓ αм ѕαнαвαт ♥ ♠ ヾ( ¯ ▽ )ゞ
( ( ( ' ^ Ω ^ ` ) ) ) J ( 'ー` )し( ● '艸` ) ☆ L ( ' ▽ ` L ) ♪ ( o_ _ )ノ 彡 ☆ ♪ ( * ¯ ω ¯ ) v ( .つ ∀ ≦ . )
(゚c_ ,゚` . )フ( ' , `ノ) (艸^ ^ ) ( ≧ 艸 ≦ * ) ƪ ( ~ ε ~ ) ʃ ƪ ( ▿ ) ʃ ƪ ( ~ ε ~ " ) ʃ ▹ ( ⌣ ) ◃ ƪ ( ˘ ε ˘ " ) ʃ ( o · _ · )ノ" ( ⌣ _ ⌣ .. )
*ƪ ( ^ ε ^ ) ʃ ╭ ╮ ( ¯ , ¯ " ) ╭ ╮ ( ⌒ ˛ ⌒ )
usually if there are characters who do not read the text box will appear, but in my case the emergence of China. Please Help..
i‘m yedos
that's alittle bit like having an idea of something that's implanted in your eye
ive got an idea, its called the eye phone
what you get is a new retina
with the hardware inplanted into your brain
not sure of the hardware yet,,but the concept is there
also you can have a fax machine incorporated to
this fax comes out your bumbum
Just the part about a small 720p curved display, it can be done for sure but at the cost thats can scale into production? I don't think so! if it was Samsung would already make this kind of device
sent from Carina Nebula with my Nexus 5 inter dimensional cruiser...
I think you need to do some searching and find out for yourself that current technology is not there yet.
Google some of the components you're saying will be in the devices, then look at the smallest ones available. It's just not going to work.
Transistors? They won't be large enough to drive the hardware, if you can find them that will fit in a ring.
720p screen? Why? It's on a ring. Anything at that resolution is going to be UNBELIEVABLY SMALL.
High Clarity speakers? Aside from being extremely vague, you show that you have no idea how this technology works. Get some bose earbuds. Their quality is great. Then again, look how big they are. If they could make them smaller, they would. Hell, even hearing aids aren't that small, and their quality is acceptable at best.
1ghz cpu? Find me a 1ghz ARM cpu at this size. They're not available yet. The manufacturing process just isn't available.
512mb ram? Again. The chips aren't going to fit.
Maybe you can get some of these things into a ring, but definitely not all of them. The technology just isn't there yet. I don't mean to discourage you from your goal, but you really just need to be smart about it. Patent your idea, and keep your eyes and ears on the tech world. Be realistic. When the technology is available, you can bring your product to market.
Hey guys,
Why is there no app that uses your front camera and projects a keyboard on the surface of the table? You look at your phones screen and type away on the desk. This would be a pretty cool app. Do you know if there is something like this around?
Thanks
rethan2 said:
Hey guys,
Why is there no app that uses your front camera and projects a keyboard on the surface of the table? You look at your phones screen and type away on the desk. This would be a pretty cool app. Do you know if there is something like this around?
Thanks
Click to expand...
Click to collapse
It's a little difficult to hold your phone and type at the same time.. you'd need a stand which would kind of get in your way.
There's tech out there with eye tracking technology where you could use Holo glasses to type with your eyes although traditional typing is still phenomenally faster. And mouth tracking technology w/ face cam that converts how you move your mouth/tongue to text might be a better tech idea since it could work in any noise level environment and be faster than as fast as most typers can type.
Some day we'll have signal detectors on our heads intercepting directions we think.
But aside from the ergonomic issues, your idea is fun.
Depth sensors
rethan2 said:
Why is there no app that uses your front camera and projects a keyboard on the surface of the table
Click to expand...
Click to collapse
Kinect, Touch+, Leap Motion, Nimble Sense, Project Tango, SoftKinetic, Intel RealSense, etc. are all various combinations of hardware and software to detect depth and gestures.
There was also a MIT and Microsoft article about the ability to use infrared LED lights and machine learning techniques to turn a simple single-lens camera into a 3-D one.
technologyreview/com/news/529986/turning-a-regular-smartphone-camera-into-a-3-d-one/
However, at this point, there seems to be a shift to depth cameras, such as the Intel RealSense that is already integrated into laptops.
Time-of-flight (TOF) cameras for point cloud data have more potential,.
It would need to be built into your phone to be cost-effective, and even then, look how pricey Project Tango is right now.
Intel RealSense is in the HP Sprout (has a projector), but it can’t do surface touch yet.
3divi has a "turn a surface into touch surface" prototype Youtube video (youtube/com/watch?v=upGTLrSUa5c ) that uses Kinect, and a Pico projector.
Touch+ was the biggest help last year, as it was only $75 for surface touch, but the people behind it can’t even get the drivers out yet.
Check out Wired’s article on the UI of Magic Leap:
>Magic Leap UI: Totems: cameras on headset could track any piece of material that’s been defined as “mouse.”
>Shows keyboard made from soft rubber that would deform and provide haptic feedback while HMD overlaid images.
>AR system may render virtual computer keyboard on surface of rectangular aluminum
wired/com/2015/01/magic-leaps-vision-for-virtual-reality/
It's like the Optimus Maximum keyboard of 2008, where every keyboard key has an OLED display so that you can remap any key to do whatever you want.
It's like that, except it doesn't cost $2000.
Eye-tracking
TryHardBlueonMac said:
There's tech out there with eye tracking technology where you could use Holo glasses to type with your eyes although traditional typing is still phenomenally faster.
Click to expand...
Click to collapse
An Android SDK for eye-tracking was revealed at CES.
dailydot/com/technology/eye-tribe-eye-tracking-android-sdk-ces-2015/
I think that modifying the camera for eye-tracking is cheap, and most likely much cheaper than changing the camera to a depth sensor like in Project Tango.
With surface touch, you still might be looking at the surface some of the time, like in the dual screen HP Sprout.
With eye-tracking, you could have an eye-tracking “select-what-am-looking-at” button.
Look, touch an easy-to-reach “tap-where-I’m-looking” button, look, and then touch the same button again.
You don’t have to keep changing your hand and finger positions between each tap.
For typing:
>Microsoft patents eye-tracking keyboard software
>The idea’s just like swipe-based keyboard software, but instead of tracking the motion of your fingertip, the system tracks eye movement.
pocketnow/com/2014/12/24/eye-tracking-keyboard
Automated lip reading
TryHardBlueonMac said:
And mouth tracking technology w/ face cam that converts how you move your mouth/tongue to text might be a better tech idea since it could work in any noise level environment and be faster than as fast as most typers can type.
Click to expand...
Click to collapse
It’s a tough problem, but people are working on it:
>“The first problem for automated lip reading is face and lip recognition.
This has improved in leaps and bounds in recent years.
A more difficult challenge is in recognizing, extracting and categorizing the geometric features of the lips during speech.
This is done by measuring the height and width of the lips as well as other features such as the shape of the ellipse bounding the lips, the amount of teeth on view and the redness of the image, which determines the amount of tongue that is visible.
Hassanat’s own visual speech recognition system is remarkably good.
His experiments achieve an average success rate of 76 percent, albeit in carefully controlled conditions.
The success rate is even higher for women because of the absence of beards and mustaches.
All this suggests that there is significant potential for visual speech recognition systems in the future, particularly as an aid to other forms of speech recognition.”
– Ahmad Hassanat at Mu’tah University in Jordan
technologyreview/com/view/530641/the-challenges-and-threats-of-automated-lip-reading/
TryHardBlueonMac said:
Some day we'll have signal detectors on our heads
Click to expand...
Click to collapse
EEG
I’ve heard about the Emotiv EEG.
Just a few basic commands that are paired up with other inputs, like eye-tracking, would be useful.
E.g. of some adequate, basic commands:
* 1) open menu of on-screen shortcuts
* 2) select-what-I’m-looking-at (using eye tracker)
* 3) no/back/cancel
What Emotiv offers:
> “In order to provide consistency and a simple range of possible actions, each user profile will contain space for training data for up to 15 different commands, which are internally labelled COMMAND1 to COMMAND15.
With an eye tracker, you probably don’t need that many brain commands if you’re just going to be using the mind for a “Yes,-select-what-I'm-looking-at” or “back” for on-screen, virtual buttons.
Thought functions merely as the switch.
Hi,
Does anybody know how to interpret the data coming from the Watch 2's PPG sensor (not heart rate, but the raw PPG values themselves)? For reference, PPG data is usually 1-2 traces (for red and/or green wavelengths), centered at 0 with a range of +/-1000, depending on the presentation (sometimes capillary blood volume, sometimes blood pressure).
The raw PPG data coming from the Watch 2, however, has a crazy format: each reading is a length-16 array of floats; the first, second and fourth values (indices 0, 1, and 3) are either numeric or NaNs; the third value (index 2) is always numeric; the remaining 12 values are always 0.0. To make things weirder, the ranges of the numeric values are very small, and centered around ~10^-40.
That said, the third value (index 2) does provide something that is recognizable as a PPG trace when plotted. But we can't figure out what the other values represent, why some of them are often NaN and others are always 0, and why the value range is so small and centered around 10^-40 (roughly).
If anybody has more experience with the raw PPG readings from this device, any thoughts would be appreciated. I reached out to Texas Instruments (makers of the AFE4405 PPG board used in the Watch 2) and they said they had no idea -- the crazy data is a Huawei thing, not TI.
Many thanks,
Erik
Any progress?
I'm working on it and I found a similar situation. I wonder how do you know the PPG sensor integrated is AFE4405. I have browsed the instruction page for AFE4405 on the official TI website, but I found no valuable information. Do you have any progress these days? Thanks a lot!!
erisinger said:
Hi,
Does anybody know how to interpret the data coming from the Watch 2's PPG sensor (not heart rate, but the raw PPG values themselves)? For reference, PPG data is usually 1-2 traces (for red and/or green wavelengths), centered at 0 with a range of +/-1000, depending on the presentation (sometimes capillary blood volume, sometimes blood pressure).
The raw PPG data coming from the Watch 2, however, has a crazy format: each reading is a length-16 array of floats; the first, second and fourth values (indices 0, 1, and 3) are either numeric or NaNs; the third value (index 2) is always numeric; the remaining 12 values are always 0.0. To make things weirder, the ranges of the numeric values are very small, and centered around ~10^-40.
That said, the third value (index 2) does provide something that is recognizable as a PPG trace when plotted. But we can't figure out what the other values represent, why some of them are often NaN and others are always 0, and why the value range is so small and centered around 10^-40 (roughly).
If anybody has more experience with the raw PPG readings from this device, any thoughts would be appreciated. I reached out to Texas Instruments (makers of the AFE4405 PPG board used in the Watch 2) and they said they had no idea -- the crazy data is a Huawei thing, not TI.
Many thanks,
Erik
Click to expand...
Click to collapse
Same situation as above
Hello! I'm facing the same situation too. I was looking into the Float array of this sensor and I've got same values. So we should figure it out a way to represent this data in a PPG chart...
Any ideas?
I will let you know if I find something useful.
Kind regards,
Angel
erisinger said:
Hi,
Does anybody know how to interpret the data coming from the Watch 2's PPG sensor (not heart rate, but the raw PPG values themselves)? For reference, PPG data is usually 1-2 traces (for red and/or green wavelengths), centered at 0 with a range of +/-1000, depending on the presentation (sometimes capillary blood volume, sometimes blood pressure).
The raw PPG data coming from the Watch 2, however, has a crazy format: each reading is a length-16 array of floats; the first, second and fourth values (indices 0, 1, and 3) are either numeric or NaNs; the third value (index 2) is always numeric; the remaining 12 values are always 0.0. To make things weirder, the ranges of the numeric values are very small, and centered around ~10^-40.
That said, the third value (index 2) does provide something that is recognizable as a PPG trace when plotted. But we can't figure out what the other values represent, why some of them are often NaN and others are always 0, and why the value range is so small and centered around 10^-40 (roughly).
If anybody has more experience with the raw PPG readings from this device, any thoughts would be appreciated. I reached out to Texas Instruments (makers of the AFE4405 PPG board used in the Watch 2) and they said they had no idea -- the crazy data is a Huawei thing, not TI.
Many thanks,
Erik
Click to expand...
Click to collapse
Conclusion
angelrc96 said:
Hello! I'm facing the same situation too. I was looking into the Float array of this sensor and I've got same values. So we should figure it out a way to represent this data in a PPG chart...
Any ideas?
I will let you know if I find something useful.
Kind regards,
Angel
Click to expand...
Click to collapse
We eventually gave up on the device for various reasons, but we did get in touch with a research team at Oxford who had published some work using the HW2's PPG sensor. They reached the same conclusion that we had: only the third trace is useful, the rest is basically noise. They had a little bit more information about what the traces represented (one turns out to be infrared, which is used to detect when the sensor is being worn), but nothing that fundamentally cleared up the odd data format.
In short, if you need PPG from the device, index 2 in each float array will give useful data. Consider the rest noise.
If you come up with anything better I'd be interested to know.
Best,
Erik
Sensor
mauriceluo69 said:
I'm working on it and I found a similar situation. I wonder how do you know the PPG sensor integrated is AFE4405. I have browsed the instruction page for AFE4405 on the official TI website, but I found no valuable information. Do you have any progress these days? Thanks a lot!!
Click to expand...
Click to collapse
If you Google around there's a way to force a dump of various specs on the watch. If you do that, you get a reference to the AFE4405. See my reply below, but in short the third trace is meaningful, the rest can be considered noise.
Best,
Erik
erisinger said:
We eventually gave up on the device for various reasons, but we did get in touch with a research team at Oxford who had published some work using the HW2's PPG sensor. They reached the same conclusion that we had: only the third trace is useful, the rest is basically noise. They had a little bit more information about what the traces represented (one turns out to be infrared, which is used to detect when the sensor is being worn), but nothing that fundamentally cleared up the odd data format.
In short, if you need PPG from the device, index 2 in each float array will give useful data. Consider the rest noise.
If you come up with anything better I'd be interested to know.
Best,
Erik
Click to expand...
Click to collapse
Perfect. Thank you so much for your answer. Sincerely, Angel
erisinger said:
We eventually gave up on the device for various reasons, but we did get in touch with a research team at Oxford who had published some work using the HW2's PPG sensor. They reached the same conclusion that we had: only the third trace is useful, the rest is basically noise. They had a little bit more information about what the traces represented (one turns out to be infrared, which is used to detect when the sensor is being worn), but nothing that fundamentally cleared up the odd data format.
In short, if you need PPG from the device, index 2 in each float array will give useful data. Consider the rest noise.
If you come up with anything better I'd be interested to know.
Best,
Erik
Click to expand...
Click to collapse
Hi Erik,
Thanks for providing the information. It is really helpful. We also want to use some smartwatch to get the raw PPG data. As you said, you gave up the Huawei Watch 2. Would you please give me some information about what device you choose now?
Thanks very much,
Ruixuan
TTworld said:
Hi Erik,
Thanks for providing the information. It is really helpful. We also want to use some smartwatch to get the raw PPG data. As you said, you gave up the Huawei Watch 2. Would you please give me some information about what device you choose now?
Thanks very much,
Ruixuan
Click to expand...
Click to collapse
Hi Ruixuan,
We ended up going with the MotionSense HRV (https://md2k.org/software/how-to/supported-sensors.html) built by Ohio State University. I'm not sure what the availability is like outside of academic research circles, but you could reach out to MD2K for more information.
Best,
Erik
erisinger said:
Hi Ruixuan,
We ended up going with the MotionSense HRV built by Ohio State University. I'm not sure what the availability is like outside of academic research circles, but you could reach out to MD2K for more information.
Best,
Erik
Click to expand...
Click to collapse
Hi Erik
Thanks very much. We are also in research. We may dig further into the Android Watch a bit, since it has great flexibility.
Regards,
Ruixuan
Help please!
Hey! I am also working on a Huawei Smart watch, and facing the same concerns as you. My values are in the range 10^-40s, and none of the event indices seem to give values that, when plotted resemble a PPG. Here is what I am trying, is this the index 2 that you have mentioned? Any guidance welcome!
if (event.sensor.getType() == 65537 &&
event.values.length > 0){
Float instant_HR = event.values[2]);