It鈥檚 not just the flu that thrives aboard a busy flight or within a hectic workplace. Computer worms could spread in remarkably similar fashion, according to a new mathematical model.
Several proof-of-concept worms have been created and two, Cabir and Commwarrior, successfully spread in the 鈥渨ild鈥� although not very efficiently.
Nonetheless, as portable devices become more fully functional, the risk of a full-blown outbreak is only likely to increase, which makes modelling an outbreak important to devising countermeasures.
Advertisement
, an expert on infectious diseases at Imperial College London and , a researcher at BT鈥檚 research lab in Suffolk, UK, created a mathematical model to better understand the way a wireless computer worm would spread between portable devices.
Flu-like spread
In December 2007, another researcher, Steven Myers, at Indiana University, US, developed a model to show how a wireless computer worm might spread between stationary Wi-Fi routers.
The new model adds another dimension by factoring in the movement of people carrying infected or vulnerable devices.
Rhodes and Nekovee鈥檚 model considers a crowd of people carrying Bluetooth-enabled smartphones, each with a fixed range for connecting to other phones in the crowd.
Members of the crowd move in a straight line, and at a fixed speed. If a phone is infected with a worm it then has a fixed probability of infecting other devices while they are within range.
The work shows that the spread of a wireless worm can be represented using a common tool for representing virus outbreaks, called a standard mass-action mixing model. 鈥淭he transmission model is actually very similar to something like influenza,鈥� Rhodes told 51动漫.
鈥楽urgical mask鈥�
This means, just like a real virus, a wireless worm would spread most efficiently in a busy setting and could also hop between geographically distant locations, for example aboard a plane.
鈥淎 Bluetooth worm鈥檚 ability to spread is rooted in person-to-person contact,鈥� says , a chief researcher at Symantec Security Response in the US. So a mass-action mixing model provides a good approximation.
鈥淜nowledge that person-to-person contact, or rather device to device contact, represents a major factor in how a Bluetooth worm spreads is definitely important,鈥� Chien adds.
He also suggests that a simple mitigation strategy would be to disable non-essential Bluetooth communications during an outbreak: 鈥淭his reduces the contact occurrences and would be analogous to wearing a surgical mask in areas of potential infection.鈥�
Bruce Schneier of security company describes the work as 鈥済reat and interesting reserach.鈥�
鈥淲hat鈥檚 interesting is that malware depends a lot on the characteristics of the underlying network,鈥� he says. 鈥淭he phone network is different [to existing networks] and the spreading characteristics will be different.鈥�
However, he doubts whether it will lead to any radically new strategies for defending against such threats. 鈥淧eople subscribe to automatic update, because worms spread faster,鈥� he says. 鈥淔ast spreading is a characteristic of wireless worms, too.鈥�
Reference:
