Filtering method and system of parallel computing results

A method for filtering parallel computing results, comprising:

• receiving N-way parallel computing results from a digital circuit, wherein the digital circuit comprises application specific integrated circuit (ASIC) and field-programmable gate array (FPGA);
• dividing the received N-way parallel computing results into S fragments, where S=√{square root over (N)}, and both S and N are positive integers greater than 0;
• simultaneously initializing each fragment separately and generating input values of all the first valid positions fvp required for each fragment to perform filtering computation;
• simultaneously computing with input values of the first valid position fvp respectively corresponding to each fragment, to obtain an output result corresponding to the input value of each first valid position fvp;
• wherein the output result comprises the output result of the first valid position fvp and the output result of parallel computation;
• selecting one of a plurality of output results of the second fragment as the final output result according to the output result of the first valid position fvp of the first fragment, and selecting one of a plurality of output results of the third fragment as the final output result according to the output result of the first valid position fvp of the second fragment, to finally obtain the final output results of all the fragments, wherein the output result corresponding to the input value of the first valid position fvp of the first fragment is the final output result of the first fragment;
• combining the output results of parallel computation in the final output results of all the fragments to obtain filtered parallel computing results;
• and
• according to the output result of the first valid position fvp of the first fragment, the parallel computing results is filtered through selecting the output results of the second to the S-th fragments in sequence, to finally obtain correct parallel computing results, so that the original serial filtering computation is changed to parallel computation of S fragments.