Finding a Phone From Orbit Using Doppler Shift and Delay Locating Handsets From Space

This application covers a satellite communication system that works out where a ground device, such as a mobile phone, is located by combining three things: which beam of a spaceborne phased array hears it, how much the signal frequency has shifted because of the satellite’s motion, and how long the signal takes to make a round trip. It was filed by AST & Science, LLC and names Raymond J. Sedwick and Abel Avellan as inventors. It is an Australian divisional of application 2021219662.

The Problem

A satellite that talks directly to phones has to know roughly where they are, and ideally more precisely than “somewhere inside my footprint”. The specification builds on an earlier AST patent, US 9,973,266, which describes assembling many small satellites into a large array that communicates with end users on Earth. That earlier system already localises radio frequency devices to the extent that it can hear them, but a phased array’s field of view on the ground can be about 2700 km across, and each individual beam covers a cell of tens of kilometres. Knowing only that a device is somewhere inside a cell is a coarse answer.

The specification also points out a geometric trap. Lines of constant Doppler shift and lines of constant signal delay are parallel when you look along the direction the satellite is flying, so in that direction they tell you nothing new. A usable method has to recognise where the two measurements complement each other and where they do not.

What This Invention Does

Claim 1 is a communication system with two parts. The first is a phased antenna array that receives a signal from a wireless device through a set of beams. The beams define the array’s field of view, which is divided into cells, with each beam tied to one cell. The second is a processing device that detects the signal and works out a coarse geographic location, namely the cell the device is in.

The claim then adds the refinement. Each beam has a beam frequency, and the processor finds a fine geographic location along an iso-Doppler line within the cell, using the shift in beam frequency caused by the Doppler effect. It also uses the round-trip time of flight of a signal between the device and the array. The claim finishes by saying the iso-Doppler contour is orthogonal to the iso-flight time contour in the cross-track direction, meaning at right angles to the satellite’s direction of travel. Where two orthogonal sets of contours cross, you get a position in two dimensions instead of just a line.

The title and the claims agree here. Two further independent claims offer variations. Claim 11 has the processor measure signal delay relative to what would be expected at the centre of the cell, defining the location by contours of constant relative delay. Claim 14 measures the frequency shift relative to what would be expected at the cell centre, which reduces the uncertainty of the signal location to contours of constant frequency shift.

The specification supplies worked geometry. For a satellite at 700 km altitude, the field of view is about 2700 km across, assuming a minimum elevation angle of 20 degrees, with beam cells about 45 km in diameter (claim 7 says 40 to 50 km, at 400 to 900 MHz). Across the whole field of view the Doppler shift runs from about -16000 Hz to +16000 Hz, but within the central cell it only spans about -600 to +600 Hz, and at the edge of the field of view about -300 to +300 Hz. On timing, a round-trip time known to about 1 microsecond fixes distance to within 150 m, and signal delay near the edge of the field of view is up to 5 milliseconds beyond that at the point directly below the satellite.

The cell width follows from wavelength, range and array diameter. At a wavelength of 30 cm, a range of 700 km and a 20 m array, the cell is just under 13 km wide at nadir. Near the edge of the field of view the cell stretches to roughly 26 km by 78 km. The figures also show that in the central cell the Doppler and delay measurements leave a two-fold ambiguity, while in a cell at the edge of the field of view they resolve the position with no ambiguity.

Key Features

• Cell-level coarse location. Because each beam points at a known cell, simply detecting a signal in a beam places the device within that cell, even if the device has not set up a connection. The claims cover both a device that is disconnected from the beams and one that is in communication with them.
• Iso-Doppler fine location. Using the shift in beam frequency due to Doppler, the system narrows the position to a line inside the cell. The satellite’s own motion is the main source of that shift for a stationary or slow device.
• Round-trip time of flight. The delay of the signal between device and array adds a second, circular, set of contours centred on the point beneath the satellite.
• Orthogonal contours. Claim 1 relies on the Doppler and flight-time contours crossing at right angles in the cross-track direction so that the intersection pins down a location.
• Measurement relative to the cell centre. Claims 11 and 14 take delay or frequency shift relative to what a device at the cell centre would produce, then map contours of constant relative delay or constant frequency shift.
• Works with other cues. The specification notes that Doppler, delay and received power can be used together or separately, and that power relative to the beam centre can help break the ambiguity where Doppler and delay contours line up.

Who Is Behind It

The applicant is AST & Science, LLC, the operating company behind AST SpaceMobile. Its website at ast-science.com describes a satellite network intended to deliver cellular broadband directly to ordinary smartphones through its BlueBird satellites, and says it has partnered with more than 60 mobile operators. The page does not state a head office location. Abel Avellan, named here as an inventor, is also the named inventor on the earlier US 9,973,266 patent that this specification builds on, and that patent lists AST & Science LLC as assignee.

On priority, the specification states that application 2021219662 is the Australian national phase of PCT/US2021/017223, which claims the benefit of US Provisional Application No. 62/975,027, filed 11 February 2020. This application is a divisional of 2021219662, so the priority country is the United States.

Why It Matters

A direct-to-phone satellite network is in effect a very large cell tower in orbit, and like a cell tower it benefits from knowing where its users are. Position helps with beam steering and handover, and the specification points to the same family of techniques used by the Global Positioning System, which depends on knowing the satellite’s own position precisely. This document says satellite positions would be established by other methods, such as triangulating the satellite’s transmissions with ground stations, and that this is outside its scope.

What makes the approach attractive is that it needs no new hardware on the phone. The phone only has to emit a signal the array can hear, and the satellite does the measuring. That is also the limitation: the quality of the result depends on whether the phone’s transmit frequency is known, whether the signal is cooperative enough to time the round trip, and where in the satellite’s field of view the device sits. The specification says all this openly, and gives no measured accuracy from a real test. It gives geometry and the theoretical case, and a probabilistic accuracy statement left with unspecified variables.

Related Concepts

• AST SpaceMobile – the satellite-to-phone broadband company operating through the applicant.
• Phased array – the antenna type that forms the many beams and ground cells the method relies on.
• Doppler effect – the frequency shift due to relative motion that gives the iso-Doppler lines.
• Time of flight – the signal travel time that gives the iso-delay contours.
• Multilateration – the general idea of fixing a position from where several measured contours intersect.

---

AU 2026202239 was published in the Australian Official Journal of Patents on 9 April 2026 and is open for public inspection. Patent applications represent inventions that are sought to be protected and do not necessarily reflect commercially available products.