Melbourne Space Program
Re-architected 802.11 PHY for a novel satellite communication protocol
To deliver a system design within an appropriate time-frame, our team adapted the 802.11 WiFi physical layer for satellite communications. Many of the algorithms were nearly fit for purpose, but some key modifications included:
- Increasing the interleaver length (and with it, overall system latency) due to "bursty" channel error effects.
- Reducing the number of bits modulated with a subcarrier due to peak power limitations.
- Simplifying the FEC architecture for the viterbi decoder implementation.
- Reduced the number of subcarriers so that we could achieve a 5 MHz channel bandwidth.
Implemented an optimized PHY on an Analog Devices TS201S DSP
The most satisfying part of the whole program: building the satellite and fitting it onto a DSP chip under the constraints of an 8000 cycle clock budget, a ~1W average power budget, while achieving a data rate of 1.5 Mbps 600km away.
To contribute, I implemented optimized:
- Convolutional encoders and Viterbi decoders,
- Interleavers and de-interleavers,
- Symbol mappers and de-mappers,
- Symbol synchronisation, timing and frequency offset correction,
- Frequency equalisation, pilot subcarrier interpolation.
Designed a novel inflatable antenna and designed the system link budget
A key component of enabling a high data rate communication protocol is ensuring your satellite antenna can concentrate its energy well enough in a particular region of space (towards the relevant ground station).
Our program's most ambitious early objective was to proof of concept an inflatable antenna, made from aluminium-coated Mylar. This antenna was designed to inflate into a cassegrain topology. I built an antenna design framework in MATLAB, which designed the 3D antenna geometry and enabled us to shape the inflatable structure from the constituent 2D sheets of Mylar.
We hoped this design to provide up to 13dB of gain with respect to an isotropic radiator.
Mentored early engineering students so that they could contribute technically
Through my early involvement in the program, I quickly became a guide and mentor for first, second and third year students who wanted to get involved before their final years. This afforded them a great opportunity to learn about the types of engineering that interested them, and obtain a sense of practical relevance for the theory they obtained in their degrees. A few of these students then went on to lead the telecoms team and other teams after I left the project.
Worked with ITU, ACMA and legal to negotiate bandwidth allocation for the protocol
One aspect of the program I loved was just how multi-disciplinary each of us needed to be to progress the project in a timely manner. Allocating spectrum for our satellite was a legal challenge in itself, and required clear planning and declaration of intent with the ACMA and ITU. After careful consideration, the team made the decision to stick with an available amateur radio band of sufficient bandwidth at around 5.6/5.8 GHz.
Project managed and integrated design with other subsystems
Any large project requires a bit of healthy project management overhead. I worked with the other subsystem team leads to inform the project timeline, and communicate key subsystem constraints and requirements to the rest of the group.
