GPS: Explaining the Ubiquity of Time

Time was created by people when they wanted to put boundaries on what we now call a day, the length of time it takes the earth to make one rotation on its polar axis. Possibly it was the Egyptians that created a 24-hour day. The positions of the stars in the sky helped measure time as well, as the night was divided into 12 hours.

Well, time has come a long way since those many thousands of years ago. Now, in the 21st century, we have GPS with 27 flying clocks in orbit around the earth. explains some details of time in orbit like this: The time differences of GPS satellites due to relativity are measured on the scale of nanoseconds — one nanosecond being just one billionth of a second — however the difference is enough to be measured and enough to matter.

GPS satellites travel at approximately 8,700 mph (14,000 km/h) with respect to Earth. This means time runs 7,200 nanoseconds per day slower for a satellite relative to us on Earth as described by special relativity.

However, using general relativity it is possible to calculate that time goes faster for a GPS satellite by 45,900 nanoseconds per day, due to the satellite being 19,000km above the Earth (therefore in weaker gravity). This means overall time runs 38,700 (45,900-7,200) nanoseconds faster per day for a GPS satellite relative to us stationary on Earth.

GPS receivers in communication infrastructure
GPS receivers provide base stations with the precise time and frequency to synchronize their time so that when a cell phone moves between different base stations, we have a seamless hand-off of calls, streaming video with the quality we expect, accurate car navigation, and E911 service in the US.

GPS jamming and GPS spoofing (when a GPS signal appears to the receivers to be a valid signal but the frequency, position, and/or time-of-day content is altered) can and does occur. There are easy-to-get, low cost devices available in some electronics supply stores that can disrupt the GPS system and wreak havoc in a widely covered area.

One possible alternative solution to prevent this type of disruption might be a new, highly integrated packet-based primary reference source from Symmetricom called TimeProvider 1500. It provides a viable alternative to protect from GPS vulnerabilities.

Here's a bit more background information on Symmetricom's secure time products as they pertain to the smart grid (an intelligent power distribution grid):

The SyncServer SGC-1500 comes with a built-in IEEE 1588 v2 Telecom Profile input option. This enables the Smart Grid Clock to derive time from the communications wide area network (WAN), thus eliminating the need to have GPS at every substation and PMU. The rubidium atomic clock option offers holdover capability in the event of GPS disruption. These options result in a highly cost effective and resilient solution for power utilities.

Join us on Integration Nation on November 20 at 11:00 a.m. EST (16:00 GMT/UTC) for a chat session on clock integration ICs. Ian Dobson from IDT will be our resident expert in a text-based, lively, and educational discussion. IDT has a great integrated product called a Programmable Universal Frequency Translator that relates to this chat. See the article on EDN for a synopsis of this highly integrated, versatile device that reduces clock tree footprint.

Have you done any design work on ultra-high accuracy time sources such as these? What problems did you encounter?

Related posts:

30 comments on “GPS: Explaining the Ubiquity of Time

  1. eafpres
    November 12, 2013

    I used to work for NIST (originally I worked for NBS, then we changed our name to NIST when the head of the agency became a sub-cabinet position, but that is another story…) and worked in Building 2 of the campus in Boulder.  Over in Building 1 was the atomic clock, and all those award winning physicists who worked on time standards and such.  Becuase the radio transmitter for the time sync (WWV) was located in Fort Collins, CO, many people thought the atomic clock was there, not Boulder.

    In part thanks to all the good work of NIST, we have good clocks in the GPS satellites, and evolution of other methods such as the rubidium clock.  So many people take time for granted.  Now we can have atomic level timing precision on an IC, but most people don't spare a nanosecond of thought about all the commerce, security, and defense that is built on accurate time.

  2. Vishal Prajapati
    November 13, 2013

    Now we can have atomic level timing precision on an IC.


    That is interesting. Can you share some more thoughts on this? Is it possbile to put atomic clock in the IC form or a part of IC?

  3. Vishal Prajapati
    November 13, 2013

    This was nice insites on the GPS system priciple of working. I never thought the GPS satellites would be nenoseconds precise. Through they provide timing information but I thought they are derived from central servers of the GPS system.

  4. samicksha
    November 13, 2013

    You are on right door Vishal, GPS receiver works by measuring the relative time delay of signals from a minimum of four, but usually more GPS satellites, each of which has four onboard caesium or rubidium atomic clocks, Precision time references that use GPS are marketed for use in computer networks and cellular communications networks, and do maintain accuracy to within about 50ns.

  5. etnapowers
    November 13, 2013

    The satellite describes an elliptic curve around the Earth so the estimated distance of 19,000km above the Earth is a mean value, averaged on a day I suppose.

  6. etnapowers
    November 13, 2013

    This system is really interesting, it could enhance the overall effectiveness of satellite alarms by contrasting the use of the jammers

  7. etnapowers
    November 13, 2013

    @eafpres: were you involved in the qualification of this atomic clock grid?

    If yes , how much time did the qualification process take for completion?


  8. etnapowers
    November 13, 2013

    @Vishal: I think this will be possible when the IC will be realized on a substrate having the tickness of an atom for example on graphene material.

  9. eafpres
    November 13, 2013

    @Vishal–here are a couple of sources to read about the chip level atomic clocks:

    First, from our own Bill Schweber, on EE Times:

    Chip Scale Atomic Clock Approaches Performance of Modules

    And an article from Honeywell:

    Chip Scale Atomic Clock (CSAC)

  10. eafpres
    November 13, 2013

    Hi etnapowers–no, I was in another Division of NIST, so I was not directly involved.  However, it was great to work on the Boulder Campus of NIST, so many interesting projects and many famous researchers were there.  David Wineland won a Nobel Prize in 2012, Carl Wieman and Eric Cornell won in 2001, and John Hall won in 2005.  All either were at NIST or worked in collaboration between NIST Boulder and the University of Colorado.

  11. Davidled
    November 13, 2013

    ->Now we can have atomic level timing precision on an IC,

    As far as I know, IC generated its own Clock by generic OS with CPU Time and Clock function. I wonder how this time clock is precisely corrected comparing to atomic level timing.

  12. eafpres
    November 13, 2013

    Hi DaeJ–not sure if you are referring to the sync of the absolute time.  The IC atomic clocks allow very low drift of the system time.  However, you are right they do not help to synchronize.  There are ways to do this.  One obvious way is to connect a GPS receiver to the system, with an antenna that can lock enough satellites, then get the time from the GPS.  This would only be needed upon initialization or reset.

  13. Davidled
    November 13, 2013

    What if GPS satellites get a malfunction periodically or signal has interference, all communication of cellular would not work properly. I wonder that these kind system relaying on GPS satellites have any back-up plan.

  14. Netcrawl
    November 13, 2013

    its a scary scenario something that could make a huge impact, the biggest hit could be military and aerospace companies. This one has the potential to devastate US military's war effort capability because much of our own military's communication and capabilities rely more on satellites systems-GPS to do most of the most hard hitting strikes.   

  15. Netcrawl
    November 13, 2013

    In terms of interferences I think the single biggest threat would be the solar storm, this has the potential to take a huge devastating hit on most of our satellites and communication systems, anything that used electronics.Not just interference but also make our satellites useless or heaviliy damaged.

  16. Netcrawl
    November 13, 2013

    Those are system-based clock, they get their time through their CPU well its good enough but in term of precision or accuracy I'm stilll on atomic level timing.

  17. Netcrawl
    November 13, 2013

    Those are system-based clock, they get their time through their CPU well its good enough but in term of precision or accuracy I'm stilll on atomic level timing.

  18. etnapowers
    November 14, 2013

     How does it work the the rubidium atomic clock option? The clock is protected from failure? Where is it placed ?

  19. etnapowers
    November 14, 2013

    @Blaine: it was really great for you having the chance to work in an exciting place like NIST, I found this link on David  Wineland for his studies on quantum systems, getting the nobel price. 

  20. etnapowers
    November 14, 2013

    @Blaine I found on the web a biography of Carl Wieman:


    “I would be remiss if I failed to mention  the tremendous benefits that I have gained in my career by having  a wife who is a very talented and intelligent physicist (as well  as being a wonderful person of course). Shortly after we arrived  in Boulder, Sarah took a job at the NIST Boulder labs where she  has worked ever since. We worked together on the PV experiments”


    Blaine, the NIST experience has been really exciting for you, true?




  21. etnapowers
    November 14, 2013

    “Carl Wieman and Eric Cornell won in 2001”


    Here's the description of what they made:

    The Nobel Prize in Physics 2001 was awarded jointly to Eric A. Cornell, Wolfgang Ketterle and Carl E. Wieman “for the achievement of Bose-Einstein condensation in dilute gases of alkali atoms, and for early fundamental studies of the properties of the condensates” .

  22. eafpres
    November 14, 2013

    @etnapowers–In my first years it was exciting and challenging.  Unfortunately there was also a lot of reorganizing and budget struggles in the areas I was in.  Later I was in management and that had good and bad aspects.  Managing programs in the government is very different from the private sector.

    But yes, it was exciting and I'm glad to have the experiences in my early career.

  23. etnapowers
    November 15, 2013

    @Blaine: thank you for your reply, I know the issues that can arise in a private company when reorganizations happen. I'm interested in the differences between managing programs in the government and the private sector, could you describe any of these differences?

  24. Ian.Dobson
    November 15, 2013

    @Netcrawl – I attended the annual WSTS conference hosted by NIST in 2011 and we heard from Dr Marc Weiss of NIST, who was one of the physicists who helped develop the GPS system about what he considered the biggest threat to the GPS system.  What he described was a spectrum allocation issue where the FCC was considering granting licenses for a new cellular base-station band less than 30MHz away from the civilian GPS frequencies.  This license would allow a nation-wide network of base-stations pumping out 10kW signals in close physical and frequency proximity to GPS signals that are usually measured in microwatts at the receiver.  By physical proximnity I mean that the GPS receiver used by many cellular base-stations may have its receiver antenna only a few feet away from the transmitter for this new RF band.

  25. eafpres
    November 15, 2013

    @etnapowers–in the US Government, the Human Resources rules are very complex.  There is a complex performance management system which, while designed to reward for performance, does not effectively ensure that.  In the case of poor performance, it can be extremely difficult for a manager to address it.  For instance, you have to notify the employee of the specific problem areas, then you must develop a PIP (performance improvement plan) and give the employee an opportunity to “improve”, and if that fails you then can consider some further actions but it is very difficult even then.

    Hiring is likewise complex.  In many cases you have to form selection committees which review candidates and make recomendations, there are lots of rules about having to “compete” positions–i.e. if you find a candidate you like, you have to show that you considered others in the process, including internal candidates.

    Termination can be almost impossible.  This leads to the impression that many in government are not caring about their jobs and other sterotypes.  In reality, there are many very dedicated persons but also some “dead wood” that is hard to remove.

    During budget cutbacks or shifts in funding between parts of an organization, this starts a process called RIF–reduction in force.  There are mountains of rules about this.  One very interesting aspect is that frequently, higher graded employees initially on the list to eliminate their jobs can “bump” into lower graded positions that were not targeted to remove.  These higher graded employees can then take over the lower job but keep their pay level.  

    A lot of the above is in place in many large corporations as well.  Much of it comes from the practice of avoiding legal actions regarding employees at all cost.  In the US, it is relatively easy to challenge an employer in court for discrimination or other illegal practices if you are terminated.  In some states, like where I am in Colorado, it is “at will employment”.  Thus in Colorado, employers can terminate for no reason.  As long as they tell the employee nothing other than “we no longer need you” they can do as they please.  But in many cases that isn't the case and the policies and procedures are based on avoiding the chance of an employee taking legal action if terminated.  This creates some crazy outcomes.  

  26. eafpres
    November 15, 2013

    Hi Ian–I assume he was talking about the huge mess with LightSquared.  Fortunately, the FCC came to their senses after a huge uprising in the GPS community.  LightSquared has since filed for bankruptcy.

  27. Ian.Dobson
    November 15, 2013

    There's another interesting angle to having cellular base-stations in close time synchronization to one another: data bandwidth.  The LTE protocol that the most recent handsets use can use 64QAM encoding in the absence of much interference to transfer more data bits per timeslot.  However if interference rises, the communication will have to drop back to a lower-density encoding scheme, thus reducing data bandwidth.

    However the biggest source of interference for one base-station is the signals coming out of other nearby base-station transmitters.  But what if the base-stations were able to tell each other over a wired link what they were about to transmit?  Then the receiver could cancel out that part of the signal and get a better 'view' of the rest of the signal, including the real data.  While there are clearly some latency and bandwidth issues that need to be dealt with over those wired links, it is also pretty obvious that those base-stations better have pretty accurate time synchronization to be apply the signal cancellation at exactly the right time. 

    As mentioned by another poster, GPS signals are considered to be accurate to 50ns of time.  For this application, we will need something much tighter than that though and that will mean a wired network scheme such as IEEE 1588.

    This is something that is being worked-on by the ITU for the LTE-Advance standard.

  28. Ian.Dobson
    November 15, 2013

    @eafpres – yes, that's the one.  I hadn't tracked it and am glad to hear it was resolved sensibly.

  29. etnapowers
    November 15, 2013

    @Blaine: many thanks for your explanation. It's strange to see many analogies between USA and Italy governments. As concern the private companies the difference is higher , there are some rules that regulate the termination of an employee and the employers aren't free to terminate whenever they want to.

  30. Vishal Prajapati
    November 18, 2013

    @eafpres, I never knew this small module exists. It is really nice tiny portable module that can be embedded inside almost anything, even in the portable systems also. Thanks for the pointers. Both are nice articles.


    I like to interact on this site for this reason only. I get good knowledge by experts like you and others.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.