Doppler tracking is a widely adopted method of tracking spacecraft for high-profile missions, using networks of vast and expensive dedicated tracking equipment. Though, amateur astronomers are sometimes also capable of tracking these missions using comparatively simple and cheap radio hardware, typically when a spacecraft uses VHF, UHF, LoRa or S-band to communicate with groundstations. 

A similar effort is made using the DopTrack project of the Delft University of Technology, with the goal of being able to estimate and predict satellite orbits to similar accuracy as NORAD's TLEs. To this end, a concept was developed to use standardized commercial off-the-shelf components that could be turned into a cheap and reliable tracking network. Preliminary studies showed that this concept is a viable solution, but questions remained regarding the frequency stability of the hardware and the influence of an observation time bias on the estimated orbits. In addition, no design had been made yet that accounted for environmental constraints on the components and the protection thereof. 

The aim of this study was to first investigate the frequency stability of the GPS clock and software defined radio that were intended for the concept, in order verify that the desired estimation accuracy could be met. Then, this study investigated what the impact of a time bias in the observations is on the orbit estimation capabilities of the system. Subsequently, a prototype was developed and installed to record satellite passes in a realistic environment. This data was used to provide an estimation of the time bias in the real system, as well as compare its performance with the current DopTrack system.

It was found that the hardware recommended by the preliminary study satisfies the stability requirements of the project. Although it was also found that the system exhibits an observation time bias on the order of one second, the effects of this bias could be mitigated by estimating this bias during the orbit estimation. 

The data gathered by operating the prototype revealed two issues with the design. First, even in the Dutch weather, the components occasionally overheated, leading to reduction of the operational period of the system. Further investigation is needed to determine if different hardware components should be used, or if it can be mitigated by redesigning the box with cooling options.

Secondly, the signal-to-noise ratio of the recorded satellites does not meet the expectations set by earlier work, nor does it satisfy the established requirements. Investigations presented in this work show that the likely cause of this is an excess amount of noise picked up by the software defined radio when operated on the roof. Further investigation is needed to find if this can be mitigated by shielding of the equipment, or placing it at locations with less noise.