Since noise is random, you could not subtract noise from the recevide signal. However, you coudl use some digital filter, average to reduce the effects of the noise.
You can never remove noise. If you can, there is no need for the subject of communications.f
A naiive way is to jack up your signal power so that the noise becomes negligible.
You can however design a best receiver or detection scheme to minimize the noise effect (in MAP or ML sense). Or, as someone pointed out, you can use coding or add redundancy into your digital bits to combat the noise. The coding approach comes at a price of additional bandwidth.
In flat fading channels, increasing SNR will supress the effect of awgn(n(t)) but overall performance depends another parameter, receiver design. There are many strategies for flat fading channels. Channel estimation may be useful.
In flat fading channels, increasing SNR will supress the effect of awgn(n(t)) but overall performance depends another parameter, receiver design. There are many strategies for flat fading channels. Channel estimation may be useful.
Hi all,
the optimum receiver would have a matched filter at the begining. this filter (root raised cosine e.g.) maximizes de SNR in AWGN environment, thus minimizes noise n(t).
regards
Dani
In flat fading channels, increasing SNR will supress the effect of awgn(n(t)) but overall performance depends another parameter, receiver design. There are many strategies for flat fading channels. Channel estimation may be useful.