STUDY OF DIGITAL FILTER SHARPENING Dissertation

STUDY OF DIGITAL FILTER SHARPENING - Dissertation Example This dissertation tells that in the engineer’s daily work on dealing with Digital Signal Processing (DSP) for clients, he is often faced with several challenges that hinder the successful implementation of his task. The challenges come in diverse forms but one of such common challenges in the engineer dealing with digital signal processing has to do with system specification problems. These system specification failures often cause symmetric FIR filters in the entire software being used in the work not to function or work as it should. In a situation like this, most experts suggest a reduction in passband ripple or the increase in stopband attenuation. The fact of the situation however remains that the engineer is not always privileged to have access to any filter design software at his disposal to enable the reduction of passband ripple or the increment of the stopband attenuation. The latter scenario always happens and most engineers are always confused and limited in ideas. It is for typical situations like this that the researcher has put this research work together to serve as a catalyst for coming out with workable interventions that can be fallen upon in hard times such as the one described in the scenario. Though the researcher admits that the problem may be confronting and highly frustrating, it comes with a number of solutions, one of which is the practice of making do with the existing filter and sharpening it. This therefore brings to the floor, the issue of digital filter sharpening in handling system specification failure cases in basic engineering. ... It is against this background that the ASKAP Project (2009) mentions some expected specifications for the DSP system specifications. Among the specifications is the fact that the DSP system should be in a position to handle computed load of not less than 1 peta (1015) operations for each given second. Simultaneously, there should be a continuous flow of data at a rate of 70 Tera (1012) bits per second. In the event of all these, the practice of filtering is not eliminated. In fact, one of the major operations of the DSP system has always remained the â€Å"digitizing, filtering and beamforming of the data from the individual phased array feeds† (ASKAP Project, 2009). Anything less of the specified rate of performance calls for a reinforcement, which in the case of this research work is going to be the sharpening of available filters. Statement of the Problem The problem at hand has to do with the persistent failure in system requirements and specification for working on digita l signal processing (DSP) at a time when an engineer may not have access to filter design software to reduce passband ripple or increase stopband attenuation. It has been hypothesized through preliminary literature research that there could exist a number of interventions to handle the situation. Among the interventions however, the problem will be tackled from a perspective where the original symmetric FIR filters that was at the disposal of the engineer would be fallen on by sharpening it to get the systems back to functionality. Filter sharpening has been used for a number of complex digital signal processing; especially digital filter sharpening. Ultimately, digital