Low-Earth-orbit (LEO) constellations are a possible source of positioning signals outside GNSS. Iridium NEXT broadcasts structured L-band bursts that can be received with low-cost software-defined radios. Published Iridium positioning uses Doppler or time of arrival from a single antenna, and the one study with a spatial observable uses a linear array that measures azimuth alone. This thesis builds a passive Direction-of-Arrival (DoA) system for Iridium NEXT ring-alert bursts, based on a five-channel KrakenSDR and a five-element circular patch array. The reception chain was validated first on the bench and then outdoors. The spatial stage uses the unmodulated preamble tone of each burst and two-dimensional MUSIC, Capon and Bartlett estimators. On one outdoor session, with the phase calibration fitted on the same data, the median error against orbital predictions from TLE sets is about 3.7◦ in azimuth and 3.3◦ in elevation with multi-burst averaging. The same bursts are then used to position the receiver without GNSS. Angles alone give only a coarse position, so the fix relies on Doppler. The satellite positions and the time are decoded from the Iridium frames themselves, so the positioning solver needs no orbital catalogue at run time, no network connection and no disciplined host clock. On the same session the receiver locates itself to a few hundred metres. With ephemeris arcs fitted over the whole session, and therefore not from a cold start, the array also shortens the time needed to reach the kilometre level compared with a single antenna. The chain runs offline; a live receiver is left to future work.
Low-Earth-orbit (LEO) constellations are a possible source of positioning signals outside GNSS. Iridium NEXT broadcasts structured L-band bursts that can be received with low-cost software-defined radios. Published Iridium positioning uses Doppler or time of arrival from a single antenna, and the one study with a spatial observable uses a linear array that measures azimuth alone. This thesis builds a passive Direction-of-Arrival (DoA) system for Iridium NEXT ring-alert bursts, based on a five-channel KrakenSDR and a five-element circular patch array. The reception chain was validated first on the bench and then outdoors. The spatial stage uses the unmodulated preamble tone of each burst and two-dimensional MUSIC, Capon and Bartlett estimators. On one outdoor session, with the phase calibration fitted on the same data, the median error against orbital predictions from TLE sets is about 3.7◦ in azimuth and 3.3◦ in elevation with multi-burst averaging. The same bursts are then used to position the receiver without GNSS. Angles alone give only a coarse position, so the fix relies on Doppler. The satellite positions and the time are decoded from the Iridium frames themselves, so the positioning solver needs no orbital catalogue at run time, no network connection and no disciplined host clock. On the same session the receiver locates itself to a few hundred metres. With ephemeris arcs fitted over the whole session, and therefore not from a cold start, the array also shortens the time needed to reach the kilometre level compared with a single antenna. The chain runs offline; a live receiver is left to future work.
BEYOND GNSS: DIRECTION FINDING AND SELF-POSITIONING WITH IRIDIUM NEXT SIGNALS OF OPPORTUNITY
FLORIO, VINCENZO PIO
2025/2026
Abstract
Low-Earth-orbit (LEO) constellations are a possible source of positioning signals outside GNSS. Iridium NEXT broadcasts structured L-band bursts that can be received with low-cost software-defined radios. Published Iridium positioning uses Doppler or time of arrival from a single antenna, and the one study with a spatial observable uses a linear array that measures azimuth alone. This thesis builds a passive Direction-of-Arrival (DoA) system for Iridium NEXT ring-alert bursts, based on a five-channel KrakenSDR and a five-element circular patch array. The reception chain was validated first on the bench and then outdoors. The spatial stage uses the unmodulated preamble tone of each burst and two-dimensional MUSIC, Capon and Bartlett estimators. On one outdoor session, with the phase calibration fitted on the same data, the median error against orbital predictions from TLE sets is about 3.7◦ in azimuth and 3.3◦ in elevation with multi-burst averaging. The same bursts are then used to position the receiver without GNSS. Angles alone give only a coarse position, so the fix relies on Doppler. The satellite positions and the time are decoded from the Iridium frames themselves, so the positioning solver needs no orbital catalogue at run time, no network connection and no disciplined host clock. On the same session the receiver locates itself to a few hundred metres. With ephemeris arcs fitted over the whole session, and therefore not from a cold start, the array also shortens the time needed to reach the kilometre level compared with a single antenna. The chain runs offline; a live receiver is left to future work.I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14253/110798