High speed analog to digital conversion based on kernel shaping and digital equalization
Das Sharma, Debmalya
MetadataShow full item record
In this thesis a novel approach is introduced to increase the sampling rate in analog to digital conversion for a given technology. The ideal sampler is not achievable in reality as the input cannot be captured in zero time. Most existing research strives to achieve the narrowest possible sampling window or aperture window for a given technology. In this work we depart from this direction by discovering that the exact shape of the sampling waveform (kernel) plays an important role in the achievable sampling rate. The actual shape of the kernel and not just its width is shown to be of importance. The desired kernel is that which has a Fourier transform magnitude that is higher than a given threshold for the widest possible bandwidth. A system of multiple interleaved copies of the device is used to accommodate the higher sampling rate required for this wider band system. This is followed by a digital equalizing filter to restore the samples back to an ideal sampling scheme.