Crowd sensing aware disaster framework design with IoT technologies


Küçük K. , Bayilmis C., Sonmez A. F. , Kacar S.

JOURNAL OF AMBIENT INTELLIGENCE AND HUMANIZED COMPUTING, cilt.11, ss.1709-1725, 2020 (SCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 11
  • Basım Tarihi: 2020
  • Doi Numarası: 10.1007/s12652-019-01384-1
  • Dergi Adı: JOURNAL OF AMBIENT INTELLIGENCE AND HUMANIZED COMPUTING
  • Sayfa Sayıları: ss.1709-1725

Özet

When a disaster occurs, a huge amount of inconsistent victim or damage information data is received by many different sources. Disaster management systems achieve the completion of a significantly vital task, which is to reduce the number of victims or amount of damage caused by a disaster, with real-time information monitoring infrastructure. A fundamental role of these systems that could help rescue teams is to make a quick and accurate decision about the region that will be affected by the disaster and the possible effects of the tragedy. Employing IoT solutions in these systems provides the possibility of rapidly and precisely orienting rescue teams to be dispatched to the disaster area and also quickly receive specific information about the effects of the disaster. To achieve this purpose, we present a post-disaster framework using the IoT communication technologies for disaster management based on the proposed crowd sensing clustering algorithm in this paper. The proposed framework provides information about the damage status of buildings with crowd density data along with efficient real-time data collection, data aggregation, and the process of monitoring dissemination stages. This framework realizes clustering of resident density by using the cellular networks and Wi-Fi connections and calculating the damage status of buildings through the designed and specifically implemented IoT unit data. Furthermore, it employs a fuzzy logic-based decision support system to manage the resources. The proposed framework, on real base stations and access points dataset, has shown significant results for identifying crowd densities with the highlighting status of buildings in the disaster area.