Published June 16, 2026 | Version v1

DCCP-TRNG: A Physical True Random Number Generator Based on JohnsonNyquist Thermal Noise with SHA-256 Conditioning and Faraday Shielding

Description

DCCP-TRNG is a physical True Random Number Generator (TRNG) based on the Johnson–Nyquist thermal noise of a resistor, acquired via a 24-bit ADS1256 ADC and conditioned with SHA-256. The system is validated through three independent criteria: physical scaling V_rms ∝ √R with R²=0.978 over four resistance values, 94% EMI reduction via Faraday cage at R=1MΩ (residual ratio 5.2× above Johnson–Nyquist), and NIST SP800-22 15/15 with TestU01 BirthdaySpacings p=0.61 on physical ADS1256 data. The system operates on commodity hardware (ESP32, ADS1256, resistor, steel pot). The hardware limit preventing a QRNG claim is honestly documented: the ADS1256 internal noise floor at ~14 μV imposes a minimum ratio of 5.2× JN at R=1MΩ. The concrete path to QRNG is identified as R>10MΩ (expected ratio ~1.6×). Part of the DCCP corpus — DOI: 10.5281/zenodo.19610345.


DCCP-TRNG è un generatore fisico di numeri casuali (TRNG) basato sul rumore termico Johnson–Nyquist di una resistenza, acquisito tramite ADC ADS1256 a 24 bit e condizionato con SHA-256. Il sistema è validato attraverso tre criteri indipendenti: scaling fisico V_rms ∝ √R con R²=0.978 su quattro valori di resistenza, riduzione EMI del 94% tramite gabbia di Faraday a R=1MΩ (rapporto residuo 5.2× sopra Johnson–Nyquist), e NIST SP800-22 15/15 con TestU01 BirthdaySpacings p=0.61 su dati fisici ADS1256. Il sistema opera su hardware commodity (ESP32, ADS1256, resistenza, pentola in acciaio). Il limite hardware che impedisce il claim QRNG è documentato onestamente: il noise floor interno dell'ADS1256 a ~14 μV impone un rapporto minimo di 5.2× JN con R=1MΩ. Il percorso concreto verso QRNG è identificato in R>10MΩ (rapporto atteso ~1.6×). Parte del corpus DCCP — DOI: 10.5281/zenodo.19610345.

Files

dccp_trng_paper_v1_ENG.pdf

Files (731.8 kB)

Name Size
md5:95e19f9d75f07c7ed72e78e98e37ac0c
300.2 kB Preview Download
md5:3591bb7b53e4e788dfa1796c6a29f468
302.3 kB Preview Download
md5:8435ce3faaf12d575c3f1b0a92b58956
781 Bytes Preview Download
md5:2a646752fc636f04e2450b871f067385
3.5 kB Preview Download
md5:c1db3e63952354c99668ad1e629d1065
125.0 kB Download

Additional details

Related works

Is derived from
Preprint: 10.5281/zenodo.19610345 (DOI)