algorithm - Generating Gaussian Random Numbers without a Uniform Random Number Generator -

i know many uniform random number generators(rngs) based on algorithms, physical systems , on. eventually, these lead uniformly distributed random numbers. it's interesting , important know whether there gaussian rngs, i.e. algorithm or else creates gaussian random numbers. precisely want don't want use transformations such box–muller or marsaglia polar method gaussian uniform rngs. interested if there paper, algorithm or idea create gaussian random numbers without of use uniform rngs. it's pretend don't know there exist uniform random number generators.

as noted in answers/comments, virtue of clt sum of iid random number made reasonable looking gaussian. if incoming stream uniform, bates distribution. ami tavory answer pretty amounts using bates in disguise. @ closely related irwin-hall distribution, , @ n=12 or higher lot gaussian.

there 1 method used in practice , not rely on transformation of u(0,1) - wallace method (wallace, c. s. 1996. "fast pseudorandom generators normal , exponential variates." acm transactions on mathematical software.), or gaussian pool method. advice read description here , see if fits purpose