pret
Newbie level 1
- Joined
- Sep 6, 2013
- Messages
- 1
- Helped
- 0
- Reputation
- 0
- Reaction score
- 0
- Trophy points
- 1
- Activity points
- 17
Hello
I'm going to establish processing of electro-optical signals, actually a reception/detection of weak optical pulse (not LRF though the principles are close).
I have certain processing in my mind to try in Matlab simulation (actually tried the basic steps of one in Excel, looks not bad..), however I did not apply matching filter to the received signal, while all the references I figure in the net advise on applying matching filter to the received signal to boost an SNR before applying any other further processing (to further improve SNR to the acceptable levels to achieve a required detection probability and FAR).
The transmitted signal has regular rectangular shape as is received, however it undergoes ItoV transformation and analog amplification stages to have it conditioned for an A/D sampling, while any processing to be held in digital domain.
Now to the question:
All the references I was able to figure refer to the shape of matched filter in accordance with the transmitted pulse. However, I'm going to apply it in digital domain, after analog amplification and sampling, and obviously, the transmitted rectangular pulse will convert to a kind of triangularly-shaped pulse by the analog stages (since they act and LPF aside of amplification), so the actual pulse reaching the sampling stage has somewhat distorted triangular shape.
So, in digital domain, shall I use the matched filter to fit the shape of the expected pulse after sampling (i.e. kind of triangular) or still the matched filter must be a replica of the originally transmitted (afa rectangular) pulse ?
Thanks in advance, Alex
I'm going to establish processing of electro-optical signals, actually a reception/detection of weak optical pulse (not LRF though the principles are close).
I have certain processing in my mind to try in Matlab simulation (actually tried the basic steps of one in Excel, looks not bad..), however I did not apply matching filter to the received signal, while all the references I figure in the net advise on applying matching filter to the received signal to boost an SNR before applying any other further processing (to further improve SNR to the acceptable levels to achieve a required detection probability and FAR).
The transmitted signal has regular rectangular shape as is received, however it undergoes ItoV transformation and analog amplification stages to have it conditioned for an A/D sampling, while any processing to be held in digital domain.
Now to the question:
All the references I was able to figure refer to the shape of matched filter in accordance with the transmitted pulse. However, I'm going to apply it in digital domain, after analog amplification and sampling, and obviously, the transmitted rectangular pulse will convert to a kind of triangularly-shaped pulse by the analog stages (since they act and LPF aside of amplification), so the actual pulse reaching the sampling stage has somewhat distorted triangular shape.
So, in digital domain, shall I use the matched filter to fit the shape of the expected pulse after sampling (i.e. kind of triangular) or still the matched filter must be a replica of the originally transmitted (afa rectangular) pulse ?
Thanks in advance, Alex
