Multimessenger Astronomy App (MMA_app)

This application is a unified framework for real-time and distributed analysis of astrophysical events using multimessenger astronomy — combining signals from gravitational waves (GW), radio waves, and other messengers to better understand cosmic phenomena. This application enables end-to-end multimessenger inference of astrophysical events by integrating data from gravitational waves (GWs) and electromagnetic (radio) observations. It supports low-latency event detection, AI-driven data parsing, and joint parameter estimation to improve cosmological constraints (e.g., Hubble constant $H_0$).

The app integrates three core modules:

---

Modules Overview

This meta-repository aggregates the following three core components:

1. MMA_GravitationalWave

• Purpose: Detects BNS events using strain data from LIGO detectors in a federated fashion — raw data stays at each site.
• Key Features:
  • Distributed inference using deep learning across observatory nodes
  • Kafka-based messaging for event publication (e.g., PotentialMerger events)
  • Real-time classification of GW events
• Output: Publishes PotentialMerger events to the Octopus event fabric for downstream EM follow-up.

Repo: MMA_GravitationalWave

2. MMA_RadioWave

• Purpose: Handles radio follow-up of GW candidates via GCN alerts.
• Key Features:
  • Listens to GCN Kafka stream for LVK alerts and partner circulars
  • AI parser for extracting observation data (flux, time) from radio circulars
  • Federated MCMC fitting of radio light curves while preserving data locality
• Output: Posterior samples and fitted light curve parameters for downstream analysis.

Repo: MMA_GravitationalWave

3. MMA_MultimessengerAnalysis

• Purpose: Performs joint inference by overlapping posterior samples from GW and radio observations.
• Key Features:
  • Combines $d_L$ and $\theta_{\rm obs}$ posteriors from GW and EM data
  • Produces KDE-based corner plots and credible intervals
  • Improves cosmological parameter estimation (e.g., Hubble constant $H_0$)
• Output: Joint plots, metrics, and harmonized posteriors.

Repo: MMA_GravitationalWave

---

🚀 End-to-End Workflow

1. GW Detection:
   MMA_GravitationalWave detects a merger and publishes a PotentialMerger event.

2. GCN Response:
   MMA_RadioWave listens for GCN alerts, matches to GW events, and parses radio data.

3. Radio Modeling:
   Distributed MCMC is run on radio observations to produce posterior samples.

4. Joint Inference:
   MMA_MultimessengerAnalysis combines GW and radio posteriors to constrain cosmological parameters.

---

Setup & Requirements

Each module contains its own setup instructions and dependencies.