| GPS Q&A
By Wendy Corcoran Q.What effects, if any, does the Earth's electrically-charged ionosphere have on GPS signals? Does this influence my GPS readings, and, are there times when the interference is more pronounced?
R.J. Seattle, WA A.Most users of GPS technology concern themselves with the accuracy obtainable from the system. When the final solution doesn't agree with ground truth, trying to understand why can become a lesson in physics. Considering the GPS signal travels over 20,000 kilometers (km) from satellite to receiver, there is much that can go wrong without even considering user error.
A significant source of error as the GPS signal travels is encountered when passing through a layer of the atmosphere called the ionosphere. The ionosphere has been defined by Rishbeth & Garriott (1969) as "the part of the Earth's upper atmosphere where ions and electrons are present in quantities sufficient to affect the propagation of radio waves." GPS signals are radio waves at a high enough frequency as to pass through this layer of the atmosphere but changes in the ionosphere can still ultimately effect how they arrive to the GPS receiver.
Recent users of the GPS technology may not know about the ionospheric error. From previous experience with the ionosphere, field practices and software algorithms were established to compensate for ionospheric effects without the user being aware. For example, over short distances (<30 km), the ionospheric effect to a GPS signal is considered to be the same at both the base and remote, essentially canceling when we perform real-time differential or post processing of the data. Over longer distances (>50km), the field practice is to use dual frequency receivers. This is because the software can compare the effects of the ionosphere on the L1 to the effects on the L2 signal and for all intents and purposes, compute an ionospheric free solution.
Over the next few years, the user will not have the luxury of ignoring what effect the ionosphere has on GPS signals because of what is called the "Solar Cycle." This cycle is in an upward climb to what is termed the 'solar maximum,' which will increase the ionosphere error significantly and will bring attention back to the vulnerability of the GPS. Users will need to understand why they may not be achieving the required accuracies and learn how to avoid the ionospheric pitfalls. The Ionosphere
The ionosphere is a layer of charged electrons approximately 80 - 1000 km above the Earth's surface. The ionospheric effect on radio waves is proportional to the "total electron content" or TEC. The TEC is the total number of electrons along the path between the satellite and GPS receiver. The TEC varies according to solar and geomagnetic conditions time of day, geographic location and season. Area of highest ionospheric activity are the equatorial anomalies centered around a 10¡ swath at +/- 15¡ N/S of the magnetic equator, the auroral regions (65-75¡ geomagnetic latitude) and the polar caps (75-90¡ geomagnetic latitude). At times of high ionospheric activity, these areas can increase to include portions of the mid-latitudes which are typically not as effected by the ionosphere.
The ionospheric effects can typically be accounted for but this will become more difficult as we approach that part of the solar cycle called the solar maximum. A solar cycle occurs every 11 years and has an average time to maximum of 4.3 years and an average time from maximum to minimum of 6.6 years. We are currently approaching the 23rd solar maximum by the year 2000. The solar maximum is caused by an increase in sunspot activity accompanied by eruptions in the suns surface called solar flares. These flares creates ionospheric storms with high velocity solar winds, which are pushed out into planetary space and change the electron density of the ionosphere. Magnetic storms also occur in conjunction with the ionospheric storms and have a similar effect. This means that the ionospheric effects will become a major concern in the application of GPS as we approach the year 2000. Unfortunately, these effects on new generations of GPS receivers are somewhat unpredictable until we actually go through the experience. Many have tried modeling the effects or estimating the impact with respect to their GPS receiver, but the only true test will be experiencing a severe ionosphere storm. The GPS Signal
The GPS signal is comprised of two frequencies - L1 and L2. The L1 carrier is modulated with the C/A code and both L1 and L2 carriers are modulated with the P-code. These complex signals accommodate several quite distinct types of observations. The P and C/A codes give what is termed a pseudorange measurement. The pseudorange is computed by taking the travel time of the GPS signal (transmit - receive) and multiplying by the speed of light. It is called a pseudorange or 'false' range because the distance computed includes all error sources and is, therefore, not a true range.
With the L1 and L2 carrier measurements, the GPS receiver measures the phase of the incoming signal. Knowing the number of cycles between the satellite and receiver multiplied by the wavelength (i.e.: Ll = 20 cm), the distance can be computed. Unfortunately it is not that straight forward. The GPS receiver can start counting the number of cycles of the signal it has encountered since power up and can measure the fractional part of a cycle at the time of measurement, but the GPS receiver has no knowledge of how many cycles have past prior to power up. To compute the total number of cycles between the receiver and satellite, an unknown value called the 'cycle ambiguity' has to be resolved. Effects of the Ionosphere on GPS
The ionosphere is a dispersive medium that has been characterized as stable to erratic and difficult to model. The ionosphere effects both the GPS code and the carrier phase measurements in an equal and opposite way. Ionospheric scintillation caused by irregularities in the electron density can disturb the amplitude and phase of a GPS signal as it travels to the GPS receiver. The phase scintillation causes the carrier phase to advance (time advance) and the code to be delayed (group delay). This causes an apparent change in the distance between the satellite and receiver that effects GPS measurement accuracy. If the phase advance is severe enough, the GPS receiver could lose lock to one or more satellites. The amplitude scintillation causes another effect called amplitude fading which randomly scatters the GPS signal energy causing a reduced signal to noise when tracking the satellites. This too can cause loss of lock or increase measurement noise.
Not only does the user have to be concerned about the quality of the measurements taken but also the signal availability. Ionospheric effects increase with a decrease in frequency so the L2 signal (1227.60 MHz) is more susceptible to outages during high ionospheric activity than the L1 signal (1575.42 MHz). If the L2 data is not tracked, the ability to compute the ionospheric correction is gone. If satellites are being dropped and reacquired, this can also effect the DOP and, therefore, instantaneous position being computed for navigation purposes. Conclusion
So what can be done? Unfortunately the best we can do is wait and see. There are methods available today to compensate for the ionosphere but how effective they will be by the year 2000 is unknown. For single frequency receivers, the GPS satellites transmit a prediction model for the ionosphere that could compensate for 50% of the ionospheric error but in times of high ionospheric activity, this may not suffice. There are other models that have been investigated but are currently not used in commercial software because their performance hinges on decreased solar activity or the region in which the data is collected. The practice of shortening the L1 baseline lengths to take advantage of the fact that the base and remote signals traveled a similar path and therefore have common error from ionosphere, is the best bet.
For dual frequency receivers, comparing the L1 to L2 and computing an ionospheric free combination has been the method used. This only holds true if the GPS receiver can track through the ionospheric interference and maintain lock on the L2 signal. This will also effect real-time kinematic surveys since it uses a combination of L1 with L2 for its solutions. The best solution to hope for at these times is an L1 float solution (+/- 10 cm). The ability of the GPS receivers tracking loops will be the determining factor. Back |
