bloc97

 BIRTHDAY SPACINGS TEST, M= 512 N=2**24 LAMBDA=  2.0000
           data.bin        using bits  1 to 24 p-value=  .241977
           data.bin        using bits  2 to 25 p-value=  .994238
           data.bin        using bits  3 to 26 p-value=  .190980
           data.bin        using bits  4 to 27 p-value=  .088161
           data.bin        using bits  5 to 28 p-value=  .449030
           data.bin        using bits  6 to 29 p-value=  .584528
           data.bin        using bits  7 to 30 p-value=  .098073
           data.bin        using bits  8 to 31 p-value=  .549704
           data.bin        using bits  9 to 32 p-value=  .932893

bloc97

bloc97

Two Fast Methods of Generating True Random Numbers on the Arduino

Arduino true random number generator

B. Peng

December 2017

Abstract

The AVR series microcontrollers are a collection of cheap and versatile chips that are used in many applications ranging from hobbist projects to commercial infrastructure. One major problem for some hobbists is the lack of secure random number generation on the Arduino platform. The included pseudo-random number generator (PRNG) is very easy to defeat and is useless for any crypto-related uses. One recommendation from the Arduino Reference Manual is to use atmospheric noise from the chip's analog sensor pins as seed data^[6].
Unfortunately this method is extremely weak and should not be used to emulate a true random number generator (TRNG). Existing methods such as using the internal timer drift or using a dedicated generator are either too slow, requires extensive external hardware or modifications to the microcontroller's internal mech

bloc97

#include <MPU6050.h>

uint8_t randomByte(uint8_t analogPin) {		
  
  int firstValue = analogRead(analogPin);

  const uint8_t minTemporalEntropyScale = 8; //Resolution in microseconds of the onboard micros() timer
  
  int bitsCaptured = 0;