First two-way contact via geostationary satellite bounce

In an exciting first, we managed to successfully complete a two-way contact by bouncing a signal off a geostationary satellite. The satellite in question is stationed at an altitude of over 35,000 kilometers, and together with Dan, HB9Q, in Switzerland, we alternated between transmitting towards the satellite and receiving reflections. This first ever amateur QSO over geo satellite the remarkable capabilities of modern amateur radio technology when paired with large, high-gain dishes.

The Dwingeloo telescope, operating as PI9RD (previously PI9CAM)  has long been a key player in Earth-Moon-Earth (EME) communication. At an average distance of 380,000 kilometers, the Moon provides a substantial and well-established target. Though it is much farther than geostationary artificial satellites, it is also significantly larger, making it an easier target to reflect signals off. Bouncing signals off artificial geostationary satellites presents even greater challenges than EME and represents an entirely new and ambitious frontier for amateur radio. The signal levels arriving at the satellite are extremely small, so there is no effect on normal operations of the satellite.

Our journey towards this milestone began in 2015 with a QSO with Andreas, DJ5AR, utilizing a satellite in a polar orbit at 650 kilometers altitude. Encouraged by this initial success, we set our sights on the more challenging goal of achieving reflections from a geostationary satellite.

The successful contact over a geostationary satellite was made possible using the recently introduced digital mode Q65, part of the WSJT-X software suite. Q65 is specifically designed for extremely weak signal communication, making it particularly well-suited to overcoming the significant path losses inherent in satellite reflections.

The signal levels of the QSO were -21dB/-32dB. WSJT-X manages to decode this by applying averaging. After initial success, we directly tried another satellite, which gave even stronger reflections: -15dB/-24dB. Two other satellites we tried did not work, highlighting that some luck is involved – possibly involving the orientation of the solar panels towards us.

The following audio was taken in Dwingeloo and contains the second reflection. It is even audible, though barely. The signal is prepended with two tones to guide the ear.

Update 2 February 2025: in the received signals also reflections were visible at +125Hz and +171Hz. Further analysis reveals that these Doppler shifts match exactly with the known orbital elements of two other geostationary satellites.