View Single Post
Old Mar 25, 2008, 02:57 PM
Julez is offline
Find More Posts by Julez
Proud to eat Kraut ;-)
Julez's Avatar
Germany
Joined Dec 2003
5,609 Posts
Quote:
I have seen NO test of Futaba that shows what it actually does in low voltage situations.
Here you are:
http://www.rcgroups.com/forums/showt...88#post7443588

My opinion:
http://www.rcgroups.com/forums/showt...66#post7753366

My opinion concerning FHSS vs. DSSS

I think that the FHSS/DSSS hybrid, which Fasst uses, has a advantage over DSSS systems.
These are my reasons:

1) Reliability in presence of heavy interference:

Current DSSS stays on one or 2 channels. This means, that these channels must remain free of heavy interference, or the system will not operate any more.
FHSS uses all channels. This means, if all channels but one will be saturated with interference, the FHSS system will still get its information through. It is inevitable that it will make use of a free channel, as long as there is one.
The latency will increase, that is true, by the amount of
t=n/f (t=latency, n=channel number, f= hopping frequency)
For a Fasst system, only using one channel, this would be:
t= 36/125Hz= 0.288s

2) Amount of time the Reciever takes to lock on a channel:

Current DSSS system make use of 1 or 2 channels. This means, that the reciever must find the channels first, before operation is possible.
Thus, a DSSS reciever must check the entire band until it finds its transmitter. This usually takes some time, as discussed in this thread.
This process is much simpler with a FHSS system. No scanning is required. The reciever can simply listen on a channel, until the transmitter hops by. Then it recieves the control signal and the information about the hopping pattern, allowing it to sync up with the TX hopping pattern instantly.
This process takes a maximum time of
t=(n/f)+b (t=latency, n=channel number, f= hopping frequency, b=RX boot time)
For a Fasst system, this would be:
t= (36/125Hz)+???= ~0.9s
I do not know the boot time, but it must be somewhere around 0.6s, given that the total time when powering on the reciever is about 0.9s maximum.
The scanning of a DSSS system always* requires more time than a hopping sequence. Thus, the maximum sync time of a FHSS system will always* be shorter than the maximum sync time of a DSSS system**.


To sum this up: A FHSS system requires the least amount of free bandwidth to get the information through.
A FHSS system requires a very small amount of time to sync up after a power loss or similar.

These considerations make the Fasst FHSS/DSSS hybrid superior to me in terms of difference between FHSS and DSSS systems.

Personal considerations:
What I specifically like about the Fasst implementation:
1) The antennas can be neatly routed out of a carbon fuselage, as only the ends are sensitive. No need to care where one places the reciever, if there are wires or carbon next to it, etc.
It is easy to make antennas with customized length for Fasst RXes:
http://christian-hanke.blogspot.com/...a-antenna.html
1a) The lack of sattelites make the installing very easy.
2) Exceptional resistance against low voltage spikes. This means, that a Fasst system will also work with a high resistance battery, and won`t crash a plane, even if one cell is defective in a pack. This is no excuse not to care about the power supply, but if happens, there is a better chance the plane survives.

Cheers,

Julez

* all FHSS systems I know hop faster than DSSS systems scan.
If this is not correct, I am eager to learn.
** assuming the channel(s) the DSSS transmitter chooses is (are) placed randomly in the band.
Julez is offline Find More Posts by Julez
Reply With Quote