Cosine Modulated Filter Banks Systems in the Presence of Multipath Transmission
Abstract (Summary)In some systems it is desirable to transmit the signal over several channels. This type of transmission is usually used in environments where one or more channels can be out of order during some time, that makes the channel unreachable for some periods of time. By transmitting the signal over several channels we are providing the system signal diversity. Another form of Multichannel communication is Multicarrier communications, where the frequency band is divided into subchannels, and the signal is transmitted on each of the the subchannels.In this Thesis the main concepts of the parallel filter banks based on cosine-modulated filter banks are presented. Using uniform channel separation, an efficient implementation consisting of an FFT and subfilters and critical sampling of each channel, and FIR filters.There are several problems of signal design for complex channels that have randomly time-variant impulse responses. The time-variant impulse responses of these channels are a consequence of the constantly changing physical characteristic of the media. First we will focus on an ideal CMFB, without any distortion in the channel, then we will see that the recovering of the transmitted signal is perfect.A more realistic analysis of the environment in OFDM needs to be done. For this, the main problems are exposed in this type of channels, then we will give the simulation results for an specific channel with multipath transmission first without noise and after introducing AWGN noise. After detected the distortions introduced to the received signal we are going to discuss the different ways to combat the Intersymbol Interference and the Intercarrier Interference. The channel distortion results in an Intersymbol Interference and an Inter carrier Interference, which if left uncompensated, may cause errors, as we will see. The solution to the ISI problem is called equalizer and consists on designing a receiver that will be capable of cancel or reduce the ISI in the received signal.
Source Type:Master's Thesis
Date of Publication:01/01/2007