Snapdragon to use machine learning to fight malware

1 min read

The forthcoming Qualcomm Snapdragon 820 processor will feature a machine learning based approach said to support 'accurate and effective detection' of zero day malware threats. Called Snapdragon Smart Protect, the approach is also said to be the first application to use Qualcomm Zeroth technology, in which conventional anti malware solutions are augmented using an advanced cognitive computing behavioural engine.

The forthcoming Qualcomm Snapdragon 820 processor will feature a machine learning based approach said to support 'accurate and effective detection' of zero day malware threats. Called Snapdragon Smart Protect, the approach is also said to be the first application to use Qualcomm Zeroth technology, in which conventional anti malware solutions are augmented using an advanced cognitive computing behavioural engine.

"With consumers storing more personal information on their devices, data leakage incidents and malware are on the rise," said Asaf Ashkenazi, Qualcomm Technologies' director of product management. "Qualcomm Technologies is equipped to address these issues with our ability to access lower layers of the software stack and dedicated security hardware, to create a device-based, behavioural analysis approach for mobile security."

Smart Protect is said to complement signature based anti malware solutions by analysing and identifying new threats prior to new signature updates. OEMs and mobile anti malware application providers can also use Snapdragon Smart Protect APIs to perform causal analysis, offering users and the mobile industry real-time information of identified threats.

"Snapdragon Smart Protect supports deep on device monitoring for nearly instantaneous notifications of detected privacy violations and malicious activity while also providing great system performance and battery life," Ashkenazi added.

Smart Protect is expected be available in consumer devices featuring the Snapdragon 820 in 2016.