Bar code decoding using moving averages to break the (N.K.) code barrier for UPC, EAN, code 128 and others

Abstract

The new generic bar code decoder system described, with its multistage error correcting, is far more powerful than dedicated reference decode algorithms for popular linear and two dimensional bar codes. For example: the 1-7 and 2-8 EAN and UPC character substitution problems are 100% solved (because EAN and UPC become 100% self-checking with the new decoder); the ink spread tolerance improves, especially for EAN, UPC, Code 128 and Code 16K; far greater levels of random systematic edge noise are tolerated and decoded; even many spike edge errors greater than 0.50X module can be corrected; and, generous levels of acceleration are tolerated. These are all the ways bar code scanning data gets distorted. Using this new generic decoder system, the ultimate linear binary coded binary bar code with error correcting, BCB, also described, performs dramatically better than any other, despite being more compact. Computer simulation testing of the generic bar code decoder has already been conducted on BCB and UPC timing count scan data records that were distorted with various levels of ink spread, noise, spikes and acceleration using new bar code testbench software, all of which is also described including test results, to show the advances that have been made in bar code science.

Dependencies are inferred from the claim text.