Password Theft Occurs Through the Interception of Keystrokes by the Exploit
“Incorporate an additional hazard into the roster of dangers you expose yourself to when utilizing your phone for business transactions at the neighborhood coffee shop.”
Researchers from Chinese and Singaporean universities have identified a security vulnerability that enables malicious individuals to capture your passwords by monitoring your keystrokes.
This innovative cyberattack, known as Wiki-Eve, is being hailed as “the first keystroke eavesdropping system via WiFi without hacking.” The researchers demonstrated this attack, which takes advantage of a feature in wireless communication called BFI (beamforming feedback information). BFI allows devices to transmit feedback about their location more precisely, directing signals specifically to the intended routers rather than broadcasting them in all directions.
Transmitting Data in Plain Text Eliminates the Need for Physical Hacking or Encryption Key Decryption
However, a weakness of BFI, a component of the 802.11ac WiFi standard (also known as WiFi 5), is that it transmits data in plain text, eliminating the need for physical hacking or decryption of encryption keys.
The researchers devised a method to identify a user’s device and intercept the plain text transmissions. Unlike older side-channel attacks, Wiki-Eve doesn’t require the installation of rogue programs or additional links to capture a user’s keystrokes.
The researchers reported that, “Since BFI is transmitted from a smartphone to an AP [access point] in cleartext, it can be overheard by any other Wi-Fi devices switching to monitor mode.”
Wiki-Eve’s Impressive Accuracy
Wiki-Eve achieves impressive accuracy, with an 88.9% success rate for individual keystrokes and up to 65.8% accuracy in the top 10 for stealing passwords of mobile applications. Keystroke inference, which determines the pressed key based on BFI data and variations in wireless signals, is aided by deep-learning models.
Numerical passwords were used in tests, as they are easier to decipher than alphanumeric ones. The researchers successfully demonstrated Wiki-Eve by extracting WeChat Pay passwords from a subject in a nearby conference room.
Wiki-Eve joins a list of side-channel attack methods, including acoustic cryptanalysis (interpreting sounds emitted during transmission), cache attacks (probing access patterns), electromagnetic analysis (using radiation for decryption), and thermal attacks (detecting temperature variations to reveal activities).
The study assumed that users were operating on an unprotected network, a common scenario in public spaces such as coffee shops, airports, train stations, and other locations offering free WiFi.
To defend against Wiki-Eve, the researchers suggest encrypting data traffic as the most direct defense strategy, preventing attackers from intercepting BFI in plain text.
This study, titled “Password-Stealing without Hacking: Wi-Fi Enabled Practical Keystroke Eavesdropping,” was presented on the preprint server arXiv and involved researchers from Hunan University, Fudan University (both in China), and Nanyang Technological University in Singapore.
Read the original article on: Tech XPlore
Read more: AI: 2023’s Dominant Tech
