Petrosian fractal dimension

Petrosian fractal dimension \(D = \\log_{10}(N) / (\\log_{10}(N) + \\log_{10}(N / (N + 0.4 N_\delta)))\), where \(N_\delta\) counts sign changes of the first difference. A fast complexity proxy for EEG/ECG.

Usage

pfd(x)

Arguments

x

Numeric vector.

Value

List with value (D), name, and extra (n_delta, n).

Details

Reference: Petrosian, A. (1995) "Kolmogorov complexity of finite sequences and recognition of different preictal EEG patterns", Proc. 8th IEEE Symp. Comput.-Based Med. Syst., pp. 212–217.

Examples

# \donttest{
set.seed(1)
x <- cumsum(rnorm(1000))
res <- pfd(x)
res$value
#> [1] 1.027872
# }