Comparative Effects of Gaussian, Coloured and State-Dependent Noise on Fractional-Order Chen and Lorenz Chaotic Systems: A Quantitative Analysis Across the Memory-Induced Chaos Boundary

Noise and memory both shape chaotic dynamics, yet the three common noise classes, Gaussian white, temporally correlated and state-dependent, have not been compared on fractional-order chaotic systems under matched intensity conventions. This study compares them on the Caputo fractional-order Lorenz and Chen systems with a stochastic Adams–Bashforth–Moulton scheme in which the noise enters through the memory kernel. The scheme was verified against the exact Mittag-Leffler solution (observed order 1.8–2.0), the exact variance of the stochastic Volterra integral (within 1.3%) and the linear-response theory of the equilibrium fluctuations. The deterministic analysis shows that the classical Lorenz parameters are chaotic only for q > q = 0.994, whereas the Chen system is chaotic down to q = 0.824, so the fractional order was used relative to this chaos boundary. A full factorial design (two systems, six orders, five noise models, eight intensities, 24 realisations) was analysed with dispersion, attractor deformation, spectral entropy, switching rate and the largest exponent computed along the same noise realisation. Below the boundary, white noise of modest intensity sustains an attractor with a positive exponent (λ₁ = 0.44 for Lorenz at q = 0.99, η = 1). Above it, strong white noise removes sensitivity to initial conditions in the Chen system (λ₁ falls from about 2.6 to zero at η = 8, 12 and 16 for q = 0.85, 0.90 and 0.95), although dispersion and switching increase. Coloured noise is weaker than white noise at equal spectral intensity and comparable at equal variance, and state dependence raises the exponent in 43 of 44 significant comparisons and limits survival. Per-step additive noise has no continuum limit, and Box–Muller noise with random arguments is Gaussian noise. These results give a quantitative basis for modelling stochastic fractional-order chaotic systems.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-18
DOI
https://doi.org/10.21203/rs.3.rs-11495978/v1
Primary Topic
Chaos control and synchronization
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Comparative Effects of Gaussian, Coloured and State-Dependent Noise on Fractional-Order Chen and Lorenz Chaotic Systems: A Quantitative Analysis Across the Memory-Induced Chaos Boundary

Samuel Adedapo Olawoore, Tolulope Ayodeji Ojuola, Iyanuoluwa Olaniyi Ajayi, Samuel Oyindamola ADEOSUN
2 links
Research Square
Chaos control and synchronization
preprint

Comparative Effects of Gaussian, Coloured and State-Dependent Noise on Fractional-Order Chen and Lorenz Chaotic Systems: A Quantitative Analysis Across the Memory-Induced Chaos Boundary

Samuel Adedapo Olawoore, Tolulope Ayodeji Ojuola, Iyanuoluwa Olaniyi Ajayi, Samuel Oyindamola ADEOSUN
preprint en

Abstract

Noise and memory both shape chaotic dynamics, yet the three common noise classes, Gaussian white, temporally correlated and state-dependent, have not been compared on fractional-order chaotic systems under matched intensity conventions. This study compares them on the Caputo fractional-order Lorenz and Chen systems with a stochastic Adams–Bashforth–Moulton scheme in which the noise enters through the memory kernel. The scheme was verified against the exact Mittag-Leffler solution (observed order 1.8–2.0), the exact variance of the stochastic Volterra integral (within 1.3%) and the linear-response theory of the equilibrium fluctuations. The deterministic analysis shows that the classical Lorenz parameters are chaotic only for q > q = 0.994, whereas the Chen system is chaotic down to q = 0.824, so the fractional order was used relative to this chaos boundary. A full factorial design (two systems, six orders, five noise models, eight intensities, 24 realisations) was analysed with dispersion, attractor deformation, spectral entropy, switching rate and the largest exponent computed along the same noise realisation. Below the boundary, white noise of modest intensity sustains an attractor with a positive exponent (λ₁ = 0.44 for Lorenz at q = 0.99, η = 1). Above it, strong white noise removes sensitivity to initial conditions in the Chen system (λ₁ falls from about 2.6 to zero at η = 8, 12 and 16 for q = 0.85, 0.90 and 0.95), although dispersion and switching increase. Coloured noise is weaker than white noise at equal spectral intensity and comparable at equal variance, and state dependence raises the exponent in 43 of 44 significant comparisons and limits survival. Per-step additive noise has no continuum limit, and Box–Muller noise with random arguments is Gaussian noise. These results give a quantitative basis for modelling stochastic fractional-order chaotic systems.

Research Square
Ekiti State University (NG), Nigerian Defence Academy (NG), Bamidele Olumilua University of Education, Science and Technology (NG)
Chaos control and synchronization
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon