logo

Ikeda Map 📂Dynamics

Ikeda Map

Definition

A map that defines a dynamical system in the complex plane C\mathbb{C} is known as the Ikeda map. zμzexp(i[a+bz2+1])+c z \mapsto \mu z \exp \left( i \left[ a + {\frac{ b }{ |z|^{2} + 1 }} \right] \right) + c When a complex number is set as z=x+iyz = x + iy, the Ikeda map is generally expressed in terms of the real number x,yx, y and parameter tt, as a=0.4a = 0.4, b=6b = -6, c=1c = 1 as follows. x1+μ(xcostysint)yμ(xsint+ycost)t=0.461+x2+y2 \begin{align*} x \mapsto& 1 + \mu \left( x \cos t - y \sin t \right) \\ y \mapsto& \mu \left( x \sin t + y \cos t \right) \\ t = & 0.4 - {\frac{ 6 }{ 1 + x^{2} + y^{2} }} \end{align*}

Description

The Ikeda map was proposed by the Japanese physicist Kensuke Ikeda in a 1979 paper, simplifying the phenomenon of light passing through an optical resonator1. Typically, the bifurcation parameter is μ\mu, and it becomes chaotic when μ0.6\mu \ge 0.6.

Ikeda.png

From a research perspective, understanding such a system can add value to a paper. Chaotic systems among 22-dimensional maps can be found abundantly, including the famous Hénon map, but the Ikeda map is mathematically more complex2.

Code

Here is the Julia code that calculates the trajectory of the Ikeda map.

using CSV, DataFrames, Plots

μ = .9
function ikeda(x,y)
    φ = .4 - 6/(1 + x^2 + y^2)
    cosφ = cos(φ)
    sinφ = sin(φ)
    return [1 + μ*(x*cosφ - y*sinφ), μ*(x*sinφ + y*cosφ)]
end

trj_ = [rand(2)]
for t in 1:10000
    push!(trj_, ikeda(trj_[end]...))
end
data = DataFrame(x = first.(trj_), y = last.(trj_))
scatter(data.x, data.y, legend = :none, size = [500, 500], msw = 0, ms = 1, xlabel = "x", ylabel = "y"); png("Ikeda.png")
CSV.write("ikeda.csv", data)

  1. Ikeda, K. (1979). Multiple-valued stationary state and its instability of the transmitted light by a ring cavity system. Optics communications, 30(2), 257-261. https://doi.org/10.1016/0030-4018(79)90090-7 ↩︎

  2. Moradi, M., Panahi, S., Bollt, E. M., & Lai, Y. C. (2024). Data-driven model discovery with Kolmogorov-Arnold networks. arXiv preprint arXiv:2409.15167. https://doi.org/10.48550/arXiv.2409.15167 ↩︎