[Namaste-dev] Re: Weekly Report June 23 - July 4, FH-CDMA

Paul Williamson kb5mu at amsat.org
Sun Jul 6 22:20:46 PDT 2008


At 8:58 AM -0400 7/6/08, Rick Hambly \(W2GPS\) wrote:
>Does your scheme accommodate the issues that would be associated with an
>Eagle hosted ACP such as Doppler, phase discontinuities that could occur
>with antenna array de-spinning, etc?

These are good questions.

I think Doppler shift and delay timing issues will be manageable with an orbit model built into the ground station for pre-correction of the uplinks. However, I have not done any calculations to demonstrate this. I don't know how fancy the model will have to be, or how often it will have to be re-calibrated to give sufficient performance. These issues will have to be studied for any uplink modulation scheme.

I can't speak knowledgeably about the consequences of phase discontinuities from the de-spun antenna. My guess would be that the baseline single-frequency PSK modulation would be more sensitive to phase jumps than the alternative FH-CDMA scheme would be, but someone who knows more about demodulation will have to study that issue.

>It is also possible that we
>will launch both Eagle and rideshare payloads eventually. We should consider
>both orbits while pursuing the best design.

I agree completely. We also need to stay alert to the possibility that the two orbits may well require somewhat different solutions.

With care in the design of the air interface protocol, we should be able to accommodate both (and a whole family of other solutions) dynamically by just varying parameters. I envision that a stream of "overhead" data coming down from each satellite will contain various parameters that control ground station behavior dynamically. This can be used to tailor the uplink behaviors for different orbits. It also provides important flexibility for dealing with design oopses, or equipment degradation, or overloading, or interference, or emergency access priorities, or any number of other possibilities. Such a scheme is going to look too complicated at first glance, but will pay dividends over the life cycle of the design.

>I recommend that the ground station be reprogrammable to the maximum extent
>possible, both inside and outside.

Your point is a good one, but in practice it's always a tradeoff. The "maximum extent possible" would not be practical. We'll need to make design decisions that result in a groundstation that's both flexible and affordable, to the extent that's possible.

>No matter how carefully we plan our
>design it is unlikely we will get it right the first time. The more
>flexibility we build into the system now the more likely we will achieve a
>successful system after launch.

I agree that we won't get it right the first time. I also agree that there are some aspects of the system that will be difficult or impossible to test properly before launch. I do hold out a hope that we will create and execute a proper test plan that will reduce this set of unknowns to a minimum, well before launch, and before finalization of the groundstation hardware design for initial production.

73  -Paul
kb5mu at amsat.org


More information about the Namaste-dev mailing list