Skip to content

Synchrony Analysis

Synchrony analysis evaluates dependence between two or more entropy streams.

Implemented in openentropy_core::synchrony.

What it is: A cross-stream coupling analysis for shared information and event synchronization.

Use it for: Detecting whether two or more streams are influenced by common causes.

Input shape: Two streams for pair methods; three or more for global event detection.

Use this when

You have 2+ streams and want coupling/synchronization evidence.
You suspect shared external events affecting multiple sources at once.
Cross-correlation alone is insufficient and you need MI/coherence/event overlap.

Methods

mutual_information: shared-information estimate with normalized MI
phase_coherence: binary sign-coherence proxy (not Hilbert phase extraction)
cross_sync: lagged cross-correlation and best lag
global_event_detection: simultaneous outlier events across multiple streams
synchrony_analysis: one-call pairwise orchestrator

CLI

Requires 2+ sources:

openentropy analyze clock_jitter sleep_jitter --synchrony

Python SDK

from openentropy import synchrony_analysis, mutual_information, global_event_detection

pair = synchrony_analysis(data_a, data_b)
mi = mutual_information(data_a, data_b)
global_events = global_event_detection([data_a, data_b, data_c])

Rust SDK

use openentropy_core::{synchrony_analysis, mutual_information, global_event_detection};

let pair = synchrony_analysis(&data_a, &data_b);
let mi = mutual_information(&data_a, &data_b);
let events = global_event_detection(&[&data_a, &data_b, &data_c]);