Hacking yourself a satellite - recovering BEESAT-1

In 2009, BEESAT-1 was launched into low earth orbit as the first 1U CubeSat of Technische Universität Berlin. In 2011, the satellite started returning invalid telemetry data. After a short amount of time spent diagnosing the issue, operators switched to the redundant on-board computer, which initially resolved the issue. However in 2013 the issue reoccurred on the second computer. With no other on-board computer available to switch to, operations largely ceased besides occasional checks every few years to see whether the satellite was still responding to commands at all. A recovery of BEESAT-1 back into an operational state was made particularly attractive considering that due to its higher orbit, it is currently estimated to remain in space for another 20 years or more, while many of the other spacecraft of the BEESAT series have since burned up in the atmosphere. Additionally, the spacecraft is equipped with a number of sensors and actuators which were not fully utilized during the primary mission and could be used in an extended mission. However, to fully utilize all the available hardware on the spacecraft, a software update is required. Unfortunately, the software update functionality was not completed at the time of launch and as a consequence is in a nonfunctional state. An alternate solution must be devised. Following an extensive effort that diagnosed the telemetry problem, developed a solution that would remedy both the telemetry problem and allow the upload of new flight software, and implemented this solution on the actual spacecraft in orbit, the satellite was finally recovered into an operational state with the ability to perform a software update in September 2024. This talk will cover the recovery process from beginning to end, including: - A crash course in spacecraft operations, including - a brief overview of the typical subsystems of satellites and BEESAT-1 in particular - the practicalities of operating a small satellite like BEESAT-1 in a sun-synchronous low earth orbit - Diagnosing the loss of telemetry without access to said telemetry - Engineering a solution to the diagnosed issue, including: - figuring out how to upload new software without a feature intended for that task - establishing a development and testing setup for flight software development years after the original setup was dismantled - developing a patch to enable returning the satellite to an operational state and establish the ability to upload new flight software, while under the constraints posed by the lack of a proper upload method and without compromising the safety of the spacecraft - Implementing this solution on the actual spacecraft in space - A brief look at the current state of the spacecraft and remaining future tasks Along the way, some of the fun and unexpected moments experienced while working with the 15 year old software and hardware will be shared. The talk is likely to be a mix of technical and non-technical. I hope to provide enough context so that you can follow without a background in space systems or computer security.