6

I was reading this article about NASA's PhoneSat Flight Demonstrations and came across this sentence.

"To achieve this, NASA's PhoneSat design makes extensive use of commercial-off-the-shelf components, including an unmodified, consumer-grade smartphone. Out of the box smartphones already offer a wealth of capabilities needed for satellite systems, including fast processors, versatile operating systems, multiple miniature sensors, high-resolution cameras, GPS receivers, and several radios."

My Question: Is this just enthusiastic writing, or can "Out of the box smartphones ... offer ... capabilities needed for satellite systems, including ... GPS receivers..."? Is it known if they (or anyone) has tried it? Possibly with firmware modification? (see below)

Background: (phones in sats)

NASA's PhoneSats use ordinary smart phones as the main control system for the nanosatellite. University of Surrey's STRaND-1 uses a smart phone to control some functions, but with a more standard computer to share in the satellite's operation. It was the first 'smart phone satellite' in space. From here:

"During phase two, the STRaND-1 team plan to switch the satellite’s in-orbit operations to the smartphone, thereby testing the capabilities of a number of standard smartphone components for a space environment. The satellite is operated from the Surrey Space Centre’s ground station at the University of Surrey."

Based on all of the information and comments in this question and answer pair, I'm guessing that the GPS chips inside off-the-shelf smart phones would never provide data when in orbit, either because they are moving too fast, or the altitude is too high, or both. (I think that some manufacturers apply the 'OR', others the 'AND' of the two conditions) The block is usually applied in Firmware, so could conceivably be circumvented by clever hacking programming of the phone. This comment mentions this YouTube video from Adafruit's founder Limor "Ladyada" Fried (read more) where the firmware has been modified to extend the altitude to at least 50km as an example.

The article says further on:

"NASA's PhoneSat 1.0 satellite has a basic mission goal–to stay alive in space for a short period of time, sending back digital imagery of Earth and space via its camera, while also sending back information about the satellite's health."

and then:

"PhoneSat 2.0 also will supplement the capabilities of PhoneSat 1.0 by adding a two-way S-band radio to allow engineers to command the satellite from Earth, solar panels to enable longer-duration missions, and a GPS receiver. In addition, PhoneSat 2.0 will add magnetorquer coils – electro-magnets that interact with Earth's magnetic field – and reaction wheels to actively control the satellite's orientation in space. "

Which suggests that the "GPS and several radios" of the smartphones may not have been used. I'm curious if it's ever been attempted and then written about.

Community
  • 1
uhoh
  • 148,791
  • 53
  • 476
  • 1,473

2 Answers2

7

They could not have used it without some serious modifications. Most GPS systems fall under what is known as the "CoCom", which prohibits GPS from working over 60,000 feet or 1200 miles per hour. Thus it could not have functioned without some significant improvements in the firmware.

PearsonArtPhoto
  • 121,132
  • 22
  • 347
  • 614
  • 1
    I think CoCom was a voluntary (and widely followed) guideline for export of GPS devices, not a law about operating them, which expired in 1994, correct me if I'm wrong. It still exists in spirit, and in two other evolved forms known as the Wassenaar Arrangement and MTCR - see this comment below the answer I've already linked to. I think they are also both voluntary agreements on export. – uhoh May 03 '16 at 12:13
  • I also think a firmware update to a phone on a desk is not too hard for NASA to pull off. They seem to be able to do it (or similar) to rovers on other planets just fine! (...or GPS module within said phone on said desk) If you can help me track down the actual regulatory issues related to changing the firmware in one's own module, then putting it in one's own cubesat, then please let me know. It seems fine to purchase them for $3,000 from a cubesat supplier, is it not OK to DIY the same thing for two orders of magnitude less? – uhoh May 03 '16 at 12:21
  • 2
    If the firmware exists, it would certainly be legal for NASA to make the update to fly a satellite. The question becomes does it even exist? And the answer might very well be no. – PearsonArtPhoto May 03 '16 at 12:24
  • Got it. Based on my reading, I think in this case it might be as simple as commenting out a few lines or just changing the set-points. The device must in fact correctly measure altitude and speed (through multiple fixes) before it can even begin to apply the test to the values of altitude and speed to see if it is not supposed to give them to us. So the 'crime' has already been committed, and all we have to do is 'foil the cover-up' loosely speaking. – uhoh May 03 '16 at 12:34
  • 1
    That might hold true for the altitude, but the speed very well could be another matter entirely. Computing a fix at orbital speed might well require some serious upgrades in the algorithm, from one second to the next the distance will vary considerably. – PearsonArtPhoto May 03 '16 at 12:36
  • So we need to defer to people who really know how modern GPS chips work, but I think they have a large number of separate channels that each apply the correlation to 'their own' satellite, referenced to the chips internal timebase. Just because it's already linked, the AdaFruit GPS I mentioned has 66 channels and can monitor 22 satellites simultaneously. it's not like they measure one satellite at a time and then 'Oops! we moved, Darn! Start again.' – uhoh May 03 '16 at 12:42
  • I've started looking into it now. Comments welcome! – uhoh Jun 06 '16 at 13:14
  • 1
    @uhoh I'm almost positive that while what you said is true, most Phone receivers depend on knowing at least approximately where one is to work. They have a limited range of correlations that they can reference. A cold start GPS solution is something rather difficult to achieve. – PearsonArtPhoto Jun 06 '16 at 14:26
  • Wow, that's pretty hard to actually test in an definitive way. I guess removing the SIM card, turning off the WiFi and the power, and then taking a vacation to another part of the planet might do, but you'd also have to make it forget the time as well?. Hmm... I understand that a fast cold start for an impatient consumer can be difficult, but the process itself is straightforward, no? A cubesat in a good orbit has plenty of time on its hands, or am I missing something? – uhoh Jun 06 '16 at 14:34
  • Take a look at http://stackoverflow.com/a/33707/544198 – PearsonArtPhoto Jun 06 '16 at 15:18
  • That answer is 8 years old - that's four Moore's law doublings ago. I just found out chips can have 256 correlators or more, so what used to be "rather difficult" might be not-so-much now. But I'm sure you are right - since phones are still working hard to conserve power, they may use every possible shortcut to getting a fix quickly and efficiently. – uhoh Jun 06 '16 at 17:54
  • 1
    Not to mention cost. An AGPS receiver costs cents, a full GPS receiver costs much more... – PearsonArtPhoto Jun 06 '16 at 18:14
  • 3
    @uhoh yes, what you said is correct, but despite having lots of correlators, a device will probably only search the range of dopplers that it considers plausible both to save power and to avoid false decodes that could degrade the position. If the chip doesn't think that it's possible to be going several km/s, then it won't achieve a lock under those conditions. – hobbs Jul 30 '18 at 20:31
4

No, they didn't. (personal knowledge)

pericynthion
  • 10,086
  • 1
  • 43
  • 59
  • I'm not sure if your answer is considered binary, or boolean, but it's certainly to-the-point. Thanks! Do you mean NASA PhoneSats only, or does it include Surrey's STRaND-1 also? Wisely or not, I've included information from both, but it seems I'm just asking about the former. So I'll be a happy camper with either one. – uhoh May 03 '16 at 06:47
  • 1
    Both,actually :) – pericynthion May 03 '16 at 16:14