Xref: utzoo sci.crypt:2373 sci.electronics:8127 Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!wuarchive!gem.mps.ohio-state.edu!apple!mips!mark From: mark@mips.COM (Mark G. Johnson) Newsgroups: sci.crypt,sci.electronics Subject: Noise generator (hardware) Message-ID: <29071@obiwan.mips.COM> Date: 7 Oct 89 14:46:35 GMT References: <281@hhb.UUCP> <131@arinc.UUCP> Reply-To: mark@mips.COM (Mark G. Johnson) Organization: MIPS Computer Systems, Sunnyvale, CA Lines: 35 In article <131@arinc.UUCP> tsp@arinc.UUCP (Tony S. Patti,4-100C,,,) writes: >... I would also be interested in compiling a list of random >(noise) sources even if no schematic was available. > >I am aware of only ONE article which details how to build a noise-based RNG: > > "Build a Noise-Based Random Number Generator" by Terry Mayhugh which > appeared in the May 1981 BYTE Magazine (pages 452-456). > A back issue of Audio Amateur magazine contains an article that may be of interest. 1985 #1, p. 17: "A Digital Noise Generator". Try (603) 924-6371; that's Old Colony Sound Labs, who carries back issues. The circuit uses four MM5837 chips. Each 5837 contains a CMOS 17-stage linear-feedback shift register running asynchronously (from its own internal R-C relaxation oscillator clock). The outputs of the four 5837 chips are analog-summed, to smooth out and schmear out the behavior of the individual 5837's. Of course, an eager experimenter could easily do a couple of stunts to further randomize and noise-ify the output noise: 1) Use 10 5837's instead of 4. Run them off different supply voltages and/or put a resistor directly on top of (some) 5837's to change their temperature. Different V+ voltages and temeperatures affects the clock periods of the individual oscillators, further uncorrelating their outputs. 2) At the analog summing point, add in the noise from a diode or two (using the Byte ckt or others) -- -- Mark Johnson MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086 (408) 991-0208 mark@mips.com {or ...!decwrl!mips!mark}