91福利

When Dalia Hanna was working towards her PhD with professor Alex Ferworn at 91福利鈥檚 Computational Public Safety (CPS) Lab, her research focused on using drones and predictive modelling to locate missing dementia patients. Back then, student Zahidul Islam helped out with the project, playing the role of test subject in simulations with law enforcement.

Fast forward, and Zahidul Islam is a newly minted PhD candidate in the , supervised by lab member Dalia Hanna, Director of IT Services at Sheridan College and professor Ferworn. New research has built on that earlier work to develop the potentially life-saving idea of using Bluetooth Low Energy (BLE) signals to turn mobile phones into detection beacons in disaster zones.

鈥淭he CPS lab was formed to address the need to provide effective response to calamity,鈥� professor Ferworn said. 鈥淭hrough applied research, we hope to provide practical and effective tools to those who save lives in disasters. This project is also an extension of previous work with first responders dealing with the realities of urban search and rescue and the need to find trapped victims quickly in the aftermath of the structural collapse of buildings.鈥�

Designed for low power consumption and cost-effectiveness

Different from classic Bluetooth, BLE is designed for low power consumption and cost-effectiveness. It uses radio waves, not WiFi or cellular technology, meaning it doesn鈥檛 rely on either. Ideal at transmitting small bursts of data, it鈥檚 built into most modern phones and tablets, as well as fitness trackers, heart rate monitors and smartwatches.

The idea to harness BLE is the brainchild of Zahidul Islam. Born and raised in Bangladesh, he was affected by the deadly 2013 tragedy at Dhaka鈥檚 Rana Plaza, when structural failure caused a building that housed several garment factories to collapse. Some 2,500 people were rescued from the rubble, but more than 1,110 others died.

鈥淚t was live on national television,鈥� he recalled of the tragedy. 鈥淧eople were watching what was going on for four or five whole days, every second.鈥�

According to official statistics cited in Zahidul Islam鈥檚 recent research paper, response time is critical for those trapped in rubble or debris. Victims have an 80 per cent chance of survival if they鈥檙e found on the first day, but those rates plummet thereafter, falling to 30 per cent by day two and barely 10 per cent by day three.

鈥淲e鈥檙e trying to find efficiency in rescuing people the fastest we can.鈥�
 

鈥淭he major factor we look at is time, because that saves lives,鈥� Dalia Hanna said. 鈥淭hat鈥檚 the whole big thing. With everything, we鈥檙e trying to find efficiency in rescuing those people the fastest we can.鈥�

Typical search and rescue methods rely on listening equipment, trained sniffer dogs and ground-penetrating radar to locate survivors. While such techniques can be effective, there are caveats. Transporting specialized equipment to a disaster area can be time-consuming, and certain machines won鈥檛 operate without electricity or internet service, both of which are often interrupted following a disaster.

In contrast, the BLE method removes the need for expensive devices, power or internet service. Also, because BLE is less power-hungry than WiFi, a victim鈥檚 phone acts as a beacon far longer. Tests carried out in abandoned buildings in suburban Toronto showed more than two days of discoverable signal strength.

In those tests, researchers flew a drone over a simulated disaster area to identify BLE signals and create a 3D map indicating the area of greatest signal concentration. In a real-world setting, rescuers would be able to focus on that spot in the hopes of finding the most survivors. The map also means search crews don鈥檛 need to risk injury looking through potentially dangerous rubble where no one is trapped. 

鈥淲ithin an hour, we can find the pattern (of highest density),鈥� said Zahidul Islam. That rapid turnaround time is critical in a scenario where hours and minutes can be so key to survival.  

The next steps are improving the accuracy of the 3D map, then increasing the size of the test area. Once he achieves an 80 per cent accuracy threshold in small spaces, Zahidul Islam plans to scale up the testing to larger areas.