It goes without saying that I'm not about to launch into a detailed description of the binary sequence, and nor will I attempt to explain the maths. Where a predictable outcome is needed, the shift register is loaded with a 'seed' value (which may be derived from your password or the current date and time). Pseudo-random number generators are common in computing, where they are used for anything from determining the outcome of a game to generating secure passwords. By feeding selected shift register outputs back to the input via one or more 'exclusive OR' (XOR) gates, if the right connections are used the result is the somewhat mysterious 'maximum length sequence'. The algorithms are rooted in complex mathematics, common enough for cryptographers, crypto-analysts and others who love polynomials and similar high level maths functions, but somewhat mysterious to more common folk (including many electronics engineers). It's actually not random at all, and the sequence will always repeat after a period determined by the length of the shift register. This type of noise source uses a (digital) shift register, with feedback that causes the output sequence to change in an apparently random manner. Most test sets that provide a noise stimulus use what's called a 'maximum length sequence' (MLS), also known as a 'pseudo-random binary sequence' (PRBS), or linear feedback shift register (LFSR). Generating good quality noise has always been a problem.
0 kommentar(er)
