GPS receivers are able to operate at signal level of -150 dbm or less, but mobile phones need far more than -115 dbm level signal. Why GPS receivers are so much more sensitive than GSM or WiFi receivers? See these receivers

  1. GNSS L86
  2. Wi-Fi FC10

There are a few factors at play.

Firstly as Olin indicated the signal bandwidth and data rates are different.

It is a fundamental rule of information theory that the maximum amount of data you can pass through a data channel depends upon the channel bandwidth and the signal to noise ratio. See the Shannon-Hartley theory for details.

This means that if you have a lower data rate or a wider channel you can cope with a higher signal to noise ratio.

The GPS navigation message sends data at 50 bits per second over a channel that is around 20MHz wide. WiFi 802.11n tries to get 72Mb/s or around 1 million times more data over a channel that is 20MHz wide.

This would indicate that GPS can cope with significantly lower signal levels than WiFi

But that only tells you what is theoretically possible, not how to achieve it.

A GPS signal is very predictable. Each satellite is broadcasting a pesudo random data sequence which repeats every millisecond, the data is then added on top of that every 20ms. The GPS reciever knows what this pattern is and looks for it in the background noise. It does this by making a guess at the timing of the signal and then moving this expected timing around until the correlation between the expected and recieved signals is maximised. By doing this the GPS reciever can effectivly find a signal that is weaker than the random background noise.

This has the added advantage that since the signal is random it looks like noise unless you are looking for it. This means that all the satellites can broadcast on the same frequency at the same time without interfering with each other.

Another advantage of this system is that the primary information of interest to a GPS reciever is the exact timing of the recieved signals. By maximising the correlation between the expected and actual signals you get a very robust and accurate way of determining the exact signal timing.

  • $\begingroup$ so you are saying that the difference in sensitivity of the two receivers is because of data rate and predictibility of the content but not the bandwidth because bandwidth of gps signal is same as that of wifi and more than gsm. is my understanding correct? $\endgroup$
    – akm
    Dec 9 '16 at 16:26
  • $\begingroup$ For the specific example of GPS and WiFi yes, the bandwidths are about the same, only the data rate is different. Although the bandwidth for GPS is a little fuzzy, you can pick the civilian signal up easily using a 10MHz bandwidth or less but the signal is technically wider. And then the new/military signals spread the energy across the band differently. However no matter how you work it out the band is always going to be between 1/2 and 2 times the width of WiFi. Given the factor of a million data rate difference that's close enough to consider about the same. $\endgroup$
    – Andrew
    Dec 9 '16 at 16:35
  • $\begingroup$ The bandwidth/datarate/SNR relationship is purely a theoretical rule on the maximum possible data rate for a given bandwidth and SNR. The repeatability doesn't change what you can theoretically achieve in any way, it is purely the method used to achieve that performance in a real world system. GPS achieves significantly different performance to other radio systems because it uses a significantly different method of operation. $\endgroup$
    – Andrew
    Dec 9 '16 at 16:39
  • $\begingroup$ Using Shannon–Hartley theorem, a receiver can read the signal even if the SNR is negative but BW is high enough. What is the intuition behind this? I mean if noise amplitude is more than the signal, how can the signal be read? $\endgroup$
    – akm
    Dec 9 '16 at 17:48
  • $\begingroup$ Noise is random and will average out so if you use a modulation system that allows averaging or correlation then it is possible to pick up a signal below the noise. That way the noise averages to zero but the signal doesn't. This generally means you need a repetitive and predictable signal. If the signal is repetitive and predictable then it's ability to transfer data is greatly diminished. Which all fits nicely with the mathematical theory. $\endgroup$
    – Andrew
    Dec 9 '16 at 19:56

Signal bandwidth, for one.

The data that a GPS receiver gets from the satellites is far less than a streaming phone conversation over the same time. This allows the RF bandwidth to be used to get higher signal to noise ratio.

  • 1
    $\begingroup$ Could you please elaborate a bit on how bandwidth affects the sensitivity of receivers? $\endgroup$
    – akm
    Dec 9 '16 at 14:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.