AMICo: Peak matching similarity of signals

In kleinbub/rIP: Synchronization of physiological and behavioral time-series of two interacting agents.

Peak matching similarity of signals

Description

Peak matching similarity of signals

Usage

AMICo(
  experiment,
  signal,
  lagSec = 6,
  weightMalus = 50,
  match_threshold = 0.1,
  minSizeSec = 4,
  maxPeakDuration = "hrt",
  interval_sec = 10,
  algorithm = c("v2RC0.4"),
  outputName = "AMICo",
  sgol_p = 2,
  sgol_n = 25,
  correctionRangeSeconds,
  minPeakAmplitude
)

Arguments

experiment
signal	the name of a DyadSignal contained in experiment
lagSec
weightMalus	weightmalus è la percentuale di malus per il lag più estremo. Es, con weightMalus= 20 e r = 1 al massimo lag, la trasformazione diventa r' = 0.8
match_threshold	a value between 0 and 1, specifying the similarity threshold to assign a peak-peak match.
minSizeSec	the smallest allowed window size. very small windows may artificially inflate synchrony.
maxPeakDuration
interval_sec	in AMICO2, how long should unmatched sequences be?
algorithm	character. The version of AMICo algorithm.
...	Further arguments passed to peakFinder. Specifically, sgol_p, sgol_n, correctionRangeSeconds and minPeakAmplitude must be set according to the signal type.

Details

Good values for skin conductance are correctionRangeSeconds = 0.5, minPeakAmplitude = 0.05.

The difference in algorithms can be found in the changelog. Basically between 1.0 and 1.1 there is a change in peakFinder to avoid detecting micro-peaks in generally flat areas.

The difference with 2.0 is much more substantial. Instead of calculating similarities twice, in v2 the same approach is used to choose the best peak-to-peak associations and to report the final values. Also exclusively valley-peak-valley or valley-peak-halfrecovery sequences are used for the calculation. Unmatched sequences are evenly split at interval_sec intervals.

XBEST guide: row: the peak number of s1 wich was matched col: the peak number of s2 similarity: some similarity computation lag: the delta between the sample of p1 and p2 a1, p1, b1: sample position of onset, peak, and end of a feature a2, p2, b2: the same for s2 a, b: the average start and end between s1 and s2 ta, tb: the time corresponding to a and b

kleinbub/rIP documentation built on Dec. 21, 2024, 9:15 a.m.