Hi,
for Jackson - I see your message in my mail, but not here on E2E, so guess I'll have to start a new thread to reply...
Yes, both research project AND definite product(s) in mind! There is a very good technique in existence for what I'm trying to achieve, but it's not really practical, being either lab-based or very large (and expensive!) equipment. I think there's a modification which could bring this into the realms of much easier to do - and potentially with hand-held equipment ... that's where the 1843 or similar comes in. With a bit of luck, this could be required in quite large numbers - not consumer - but well worth trying for. I wouldn't manufacture, I'd license the IP - but if demonstration equipment uses particular chips sets and shows that they're a good solution, then it's highly likely that they'd be used for production equipment.
I need, initially, a research tool for the RF front end-to-DSP, first to gather data and then to add on-chip processing, working towards highly integrated, minimum parts count design.
My first requirement (working now) is to produce an 1843 design to take much longer time-records than your examples - first attempt will be simply to fill L3 memory with ADC samples (DFE in continuous mode). This needs a mix of what's in the memCapture & OOBdemo code. The OOB is pretty complex, but has everything I need round about, as well as the data capturing, but I need to change the EDMA areas ... looking at both the 1843 and the 1642 versions, as 1642 doesn't have the HWA, thus the "out end" of the EMDA from the ADC buffer should be available to modify i.e. not pointing into the HWA. I may be doing this a long way round, bit its also good training in how the 1843 & TI code work, so I can (hopefully!) write compatible code.
When I get this working, I then need to add input from one of the GPIO lines, to synchronise the 1843 with my other processing (FPGA based), as timing is key. Then it's on to post-processing and algorithm design ... to find out if my idea works or not. Getting close now! And I'm sure there will be more questions to come as I did deeper.
many thanks
Alan Milne
