Implementing Kalman Filter Algorithm in Parallel Form: Denoising Sound Wave as a Case Study
Introduction: Signal filters were originally seen as circuits or systems with frequency selecting behaviors. The development of filtering techniques went on and more sophisticated filters were introduced, such as e.g. Chebychev and Butterworth filters, which gave means of shaping the frequency characteristics of the filter in a more systematic design procedure. During this stage, the filtering was mainly considered from this frequency. Mehtod: In (SIMD) model, a parallel computer consists of N identical processors, each of the N processors possesses its own local memory where it can store both programs and data, and all processors operate under the control of a single instruction stream issued by a central control unit. Equivalently, the N processors may be assumed to hold identical copies of a single program, each processor's copy being stored in its local memory. There are N data streams, one per processor. Result: It can be seen that the computation time decreases when we increase the number of cores from 2 to 12 cores shows that Kalman Filter can achieve nearly linear speed-up by increasing the number of cores, both results are illustrated consecutively. Conclusion: Parallel multicore implementation of the Kalman filter is studied. The implementation based on SIMD model which we splitting all signal points into large segments of data and applying equations on each segment simultaneously. Discussion: Through implementing the algorithm in parallel form on the noisy sound wave signals as a case study, it is found that the proposed parallel algorithm executes about twice as fast on double cores as the sequential form on a single core, it enhances the execution time to 44.86%. And is capable of achieving linear speedup in the number of cores used.