3D GPS constellation · six orbital planes
Selected transmission / reception
The displayed signal journey is a slowed-down replay of an actual ~70–90 ms flight. It is not a two-way exchange.
Navigation solution: four pseudoranges
| Satellite | Geometric range | Pseudorange |
|---|
Physical effects are ALWAYS included in the synthetic received signals. The eccentricity, atmosphere, and Earth-rotation checkboxes control receiver corrections. The rate-compensation checkbox instead simulates turning OFF the satellite oscillator pre-offset; it changes the transmitted clock readings, not the laws of physics. A common rate error is absorbed into the estimated receiver clock bias, so position error need not grow proportionally. The four-satellite solver estimates receiver x, y, z and clock bias. Four satellites above the local horizon are chosen dynamically from the 24-satellite synthetic constellation. Geometry and dilution of precision vary with time.
Equations and corrections used
1. Signal emission and reception: moving satellite
Iteratively solve for the emission epoch t_e from the reception epoch t_r, evaluating the satellite at emission and receiver at reception. In this demonstration both positions are represented in an Earth-centered inertial (ECI) frame.
2. Pseudorange and four unknowns
Four independent satellites provide four equations for receiver position (x, y, z) and receiver clock bias δt_R. Satellite clock errors are corrected from broadcast parameters; the display includes illustrative atmospheric terms.
3. Satellite clock model and relativity
The synthetic pseudoranges always include the periodic eccentricity clock term; the Relativity checkbox controls whether the receiver corrects it. The rate-compensation switch models the nominal factory frequency pre-offset. With it OFF, the simulated satellite clocks accumulate approximately +38.6 μs/day relative to GPS coordinate time. To illustrate differential clock errors, this educational model adds small, explicitly synthetic satellite-dependent rate deviations (±0.12 μs/day). A common offset can be absorbed into the receiver clock bias at one epoch; the differential terms can affect position. This is NOT a prediction of actual GPS satellite drift without compensation. The periodic eccentricity correction is computed using a simplified eccentricity model. A typical GPS satellite's net rate relative to an Earth geoid reference is approximately +38.6 μs/day before the conventional frequency offset is applied. The receiver uses corrected satellite time, not unadjusted oscillator ticks. The elapsed-days control affects accumulated satellite clock drift only, not orbital positions.
4. Earth rotation (Sagnac) and propagation
For a moving Earth-fixed receiver, rotate endpoint coordinates to a common inertial frame at their respective event times. The model calculates geometric flight time in ECI; the Sagnac figure is the difference from a naive same-epoch ECEF distance.
5. Atmosphere, Shapiro delay, and instrumentation
Shapiro is calculated as an illustrative weak-field Earth correction. Ionospheric and tropospheric delays are adjustable illustrative range terms (not live atmospheric observations). Hardware bias and multipath are omitted from the solved example. A production receiver also handles antenna phase centers, Earth orientation, detailed ephemerides, signal-dependent biases and other effects.
6. Proper time origin versus clock display
Assigning τ = 0 is a conventional origin choice. It is not the same operation as setting a satellite clock's displayed GPS time. The navigation solution does not require all proper-time parameters to have a common origin.
Scope: this is a transparent educational model, not an operational GPS receiver. Its orbit is synthetic, its four-satellite selection and geometry are illustrative, and several propagation effects are simplified or parameterized. The actual GPS system is primarily one-way satellite-to-user broadcast; control-segment uplinks are a separate function.