Real-time identification and control of influential pandemic regions using graph signal variation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912584860434432 |
|---|---|
| author | Darapu, Sudeepini Ghosh, Subrata Ghosh, Dibakar Hens, Chittaranjan Nannuru, Santosh |
| author_facet | Darapu, Sudeepini Ghosh, Subrata Ghosh, Dibakar Hens, Chittaranjan Nannuru, Santosh |
| contents | The global spread of pandemics is facilitated by the mobility of populations, transforming localized infections into widespread phenomena. To contain it, timely identification of influential regions that accelerate this process is necessary. In this work, we model infection as a temporally evolving graph signal and propose graph signal variation-based metrics to capture spatio-temporal changes. Both graph domain and time domain locality are modeled. Based on this metric, we propose an online algorithm to identify influential regions. Simulations demonstrate that the proposed method effectively identifies geographical regions with a higher capacity to spread the infection. Isolating these regions leads to a significant reduction in cumulative infection. Simulations, along with analyses of hybrid H1N1 data and real-world Indian COVID-19 data, underscore the utility of proposed metric in enhancing our understanding and control of infection spread |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_10281 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Real-time identification and control of influential pandemic regions using graph signal variation Darapu, Sudeepini Ghosh, Subrata Ghosh, Dibakar Hens, Chittaranjan Nannuru, Santosh Signal Processing The global spread of pandemics is facilitated by the mobility of populations, transforming localized infections into widespread phenomena. To contain it, timely identification of influential regions that accelerate this process is necessary. In this work, we model infection as a temporally evolving graph signal and propose graph signal variation-based metrics to capture spatio-temporal changes. Both graph domain and time domain locality are modeled. Based on this metric, we propose an online algorithm to identify influential regions. Simulations demonstrate that the proposed method effectively identifies geographical regions with a higher capacity to spread the infection. Isolating these regions leads to a significant reduction in cumulative infection. Simulations, along with analyses of hybrid H1N1 data and real-world Indian COVID-19 data, underscore the utility of proposed metric in enhancing our understanding and control of infection spread |
| title | Real-time identification and control of influential pandemic regions using graph signal variation |
| topic | Signal Processing |
| url | https://arxiv.org/abs/2509.10281 |