DRJ Fall 2019

Conference & Exhibit

Attend The #1 BC/DR Event!

Summer Journal

Volume 32, Issue 2

Full Contents Now Available!

Tuesday, 20 January 2015 06:00

Information Collaboration and Technologies Used for Disaster Management

Written by  Annie Singla & Pankaj Kumar

An early warning system uses wireless network sensors that can provide a few seconds to tens of seconds warning prior to any disaster. The warning messages can be used to reduce damage, costs, and casualties in an earthquake. The network sensors are distributed over an area.

Input from hundreds or thousands of sensors feed into databases that can be queried in any number of ways to show real-time information on a large or small scale. The sensors communicate with neighboring nodes to eventually develop a “global traffic picture” which upon queried by human operators or automatic controllers generate control signals. In general, sensor networks pose considerable technical problems in data processing, communication, and sensor management. Important technical issues include the degree of information sharing between nodes and how nodes fuse the information from other nodes. There is a tradeoff between performance and robustness. The data is distributed across nodes, and geographically dispersed nodes are connected by unreliable links. These features render the database view more challenging, particularly for seismic indications given the low-latency, real-time, and high-reliability requirements of the battlefield. This article describes about how information is collected during disasters using distributed systems.

Disaster is termed out to be a catastrophic activity which creates havoc leading to exorbitant losses to life and property. The vulnerability of disasters varies from region to region, a large part of the country is exposed to natural hazards, which often turn into disasters causing significant disruption of socio-economic life of communities leading to loss of life and property. For each and every possible disaster, too many databases and software have been developed and designed, and millions of dollars have been expended and so many sensor networks and technologies have been installed according to the geographic locations and disaster involved.

  1. Scenarios of Disaster and Wireless Sensor Networks Model

In disastrous situations, the communication structure may not work well and in that case, sensor networks can be a very good solution. We assume that wireless sensor networks (WSNs) are launched in the region to detect hazards and that agents do not have access to the base station. This is because the network infrastructure in the region has been destroyed. Therefore, the rescue teams can get any node launched in the region to adapt it to a laptop and use it to inject queries in the WSN. This node will be used to start processing a query and to receive the answer. This work considers square regions with side lengths of one kilometer (1,000 x 1,000 meters). The nodes are identical, static, and with symmetrical radio. Each node has a GPS to inform its location. The geographical coordinates used are in the standard decimal degrees notation, with four decimal places. It gives an accuracy of about 11m in the equator (worst case). The sensor nodes are deployed in a uniformly random distribution. We use the node 1 position as the reference point. All the other points in the network (location or point in a polygon) are based on this reference point.

  1. Wireless Sensor Networks

Because of its spatial coverage and multiplicity in sensing aspect and modality, a sensor network is ideally suited for the information collaboration of disasters traversing larger areas monitoring a large number of events simultaneously (e.g. forest fires) or detecting low-observable events (e.g. seismic waves). The detection, classification, and non-local low-observable events require non-local collaboration among sensors. Aggregation of a multitude of sensor data can improve accuracy. Informed selective collaboration of sensors, in contrast to flooding data requests to all sensors, can reduce latency. Moreover, sensor collaboration can minimize bandwidth consumption (translating into energy savings) and mitigate the risk of network node/link failures. For low-observable events like seismic waves, sensor collaboration can selectively aggregate multiple sources of information to improve detection accuracy.

The benefits of sensor collaboration can be measured as improvement in one or more of the following capabilities:

  1. Detection quality: detection resolution, sensitivity, and dynamic range; misses and false alarms; response latency

  2. Track quality: tracking errors, track length, robustness against sensing gaps

  3. Scalability: size of network, number of events, number of active queries

  4. Survivability: robustness against node/link failures

  5. Resource usage: power/bandwidth consumption

A data distribution system is responsible for information retrieval and extraction. Distributed sensor networks are one of the most advanced technologies used in disaster management. Networked sensors are used to collaborate information during any disaster which is considered to be a difficult task. Distributed systems are far more scalable in practical deployment and may be the only way to achieve the large-scale accuracy and precision needed during disasters. A sensor network is designed to perform a set of high-level information processing tasks such as detection, tracking, or classification. Measures of performance for these tasks are well defined, including detection, false alarms or misses, classification errors, and track quality. The data analysis technologies that we have reviewed for disaster-related situations include the following:

  • Information extraction (IE): disaster management data must be extracted from the heterogeneous sources and stored in a common structured (e.g., relational) format that allows further processing.
  • Information retrieval (IR): users should be able to search and locate disaster-related information relevant to their needs, which are expressed using appropriate queries (e.g., keyword queries).
  • Information filtering (IF): as disaster-related data arrives from the data producers (e.g., media, local government agencies), it should be filtered and directed to the right data consumers (e.g., other agencies, businesses). The goal is to avoid information overload.
  • Data mining (DM): collected current and historic data must be mined to extract interesting patterns and trends (for instance, classify locations as safe/unsafe).
  • Decision support: analysis of the data assists in decision-making. For instance, suggest an appropriate location as ice distribution center.
  1. Decision Support System

Decision support system is an information system for disaster management and relief. It is a central database, where data and information can be made available in an online basis. It is an intelligent system to help planning activities. It is also an electronic-based correspondence system report generator that can be modified according to the user. It is a conglomeration of damage assessment, thematic hazards maps, proposed solutions, early warning, decision support, risk prediction, and situational analysis.

  1. Randomized Multi-Path Delivery

As illustrated in the figure below, we consider a three-phase approach for secure information delivery in a WSN: secret sharing of information, randomized propagation of each information share, and normal routing (e.g., min-hop routing) toward the sink. Specifically, when a sensor node wants to send a packet to the sink, it first breaks the packet into M shares according to a (T,M)-threshold secret sharing algorithm. Each share is then transmitted to some randomly picked neighbor. That neighbor will continue to relay the share it has received to other randomly picked neighbors, and so on. In each information share, there is a TTL field, whose initial value is set by the source node to control the total number of randomized relays. After each relay, the TTL field is reduced by one. When the TTL count reaches zero, the final node receiving this share stops the random propagation phase and begins to route this share toward the sink using normal single-path routing. Once the sink collects at least T shares, it can inversely compute the original information. No information can be recovered from less than T shares.Singla-Kumar-img

There is also a natural disaster management system based on location aware distributed sensor networks to achieve maximum lifetime in WSNs in disaster management applications. It is a system based on hierarchical transmission of packets from sensor nodes to the base station by identifying a path from one head to a subsequent head along the route. It divides the entire sensor network into logical concentric zones based on energy of transmission of the packet. This is transmitted from a head node to one of the head nodes in the next zone with lesser distance. The implementation profoundly uses the location awareness of sensor nodes for better routing and, hence, is applicable only to those situations where such data can be made available at the time of installation. They further provide the concept of multiple memberships of sensor nodes to different heads within its area of reach thereby handling disaster conditions where a head fails without notification to its primary members.


This article shows the comprehensive study of the role of wireless sensor networks in disaster management. It furthermore studied the different types of sensor collaboration measures. The performance are studied based on detection quality, track quality, scalability, survivability, and resource usage. It studied about the randomized multi-path delivery which is based on TTL field value and routed toward the sink. There is also a disaster management system based on energy efficiency to maximize lifetime of wireless sensor networks.


Annie Singla is a post-graduate student in disaster mitigation and management from Indian Institute of Technology in Roorkee, India.


Pankaj Kumar is a post-graduate student in disaster mitigation and management from Indian Institute of Technology in Roorkee, India.