Security Challenges in the Internet of Things

Solid security is a prerequisite to the Internet of Things becoming more pervasive in today’s world of remote-controlled pacemakers and driverless cars. Here, RT Insights International Technology Editor Dr. Opher Etzion discusses the potentially murderous results of hacked sensors in today’s systems.

I was recently asked by one of the security companies in the industry to give a talk to their forum of technical leaders about the security challenges of the Internet of Things (IoT). I am still preparing the material for my talk and am realizing that the challenges here are quite extensive.

One can talk dramatically about “electronic murder” and, indeed, this may become reality. Some examples of “electronic murder” are when a pacemaker that is remotely controlled by a physician is turned off by a hacker, when an intravenous (IV) drip that is remotely controlled by a nurse stops flowing the liquid thanks to a hacker, or when the vision sensors of a driverless car are hacked to show the wrong information, leading to a car crash. In general, a virus in any transplanted device in the human body can cause a person to die at the hand of computer virus. This type of virus has the potential to become a bigger threat than a biological virus.

These examples stem from the fact that either there are autonomic actions that are based on sensor information or there is an actuator that can be remotely controlled by using the connection to the network. Sometimes both situations exist within a system.

Indeed, one of the known sources of sensors’ inaccuracy is malicious intervention in the sensing process. I have written before about the “Twitter hoax case” in which a malicious notification on the social network caused a five-minute crash of the stock market. When talking about the IoT, the malicious intervention is done either through manipulating a physical sensor to report an event that did not really happen or by manipulating a sensor to ignore an event that did really happen. In both cases, damage can be inflicted.

While the trend today is to keep a system’s logic within easily modifiable rules, this ease in modification can be used by a hacker to change the system’s behavior, making it do things that it never really intend to. Hackers’ modifications can get the system to make wrong decisions, with damaging actions.

The security community will have to find good solutions to these issues as solid security is one of the prerequisites of getting the IoT to be more pervasive. I am optimistic that this is doable.

Want more? Check out our most-read content:

Frontiers in Artificial Intelligence for the IoT: White Paper
Research from Gartner: Real-Time Analytics with the Internet of Things
How Real-Time Railroad Data Keeps Trains Running
Operational Analytics: Five Tips for Better Decisions
Why Gateways and Controllers Are Critical for IoT Architecture

Liked this article? Share it with your colleagues!