AI Crypto Processor for Untrusted Chiplet Anomaly Response
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As chiplet-based systems integrate untrusted chiplets from various sources, there is a risk of counterfeit or malicious components being introduced, which can compromise system security and performance.
Innovation Solution
A chiplet-based system that includes a trusted AI accelerator and cryptographic processor to monitor the state of trusted and untrusted chiplets, detect anomalies, and execute corrective actions through reinforcement learning algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If untrusted chiplets are integrated into the system to improve flexibility and functionality, then system adaptability is improved, but system security and reliability deteriorate due to potential counterfeit or malicious components
Solution Approach 1:
A secure processor acts as an intermediary between trusted and untrusted chiplets. The secure processor monitors communications and operations involving untrusted chiplets, detecting anomalies and preventing malicious activities while allowing legitimate functionality to proceed. This mediator approach enables the system to integrate diverse untrusted components without compromising overall security.
Solution Approach 2:
The system implements continuous monitoring of chiplet operations with feedback mechanisms that detect anomalies in real-time. When suspicious behavior is detected from untrusted chiplets, the system responds by adjusting security parameters, isolating problematic components, or terminating communications. This feedback loop maintains system reliability while allowing flexible integration of untrusted components.
2Reliability
If monitoring and security mechanisms are added to detect anomalies in untrusted chiplets, then system security is improved, but device complexity increases
Solution Approach 1:
Security functions are segmented and distributed across multiple components rather than centralized in a single complex unit. The secure processor handles specific security tasks while other system components maintain their primary functions. This segmentation reduces overall system complexity by dividing security responsibilities into manageable, specialized modules that can operate independently.
3Measurement precision
If continuous monitoring of chiplet states is performed to detect anomalies, then measurement precision is improved, but energy consumption increases
Solution Approach 1:
Instead of continuous monitoring, the system employs periodic sampling of chiplet states at strategically chosen intervals. The secure processor checks for anomalies at key transition points and after specific operation sequences, rather than constantly monitoring all operations. This periodic approach maintains adequate detection precision while significantly reducing energy consumption compared to continuous monitoring.
Data Source
AI summary
A chiplet-based system may include a first chiplet mounted to an interposer that is designated as being from one or more trusted sources, a second chiplet mounted to the interposer that is designated as not being from the one or more trusted sources, and an artificial intelligence (AI) accelerator. The AI accelerator may be programmed to monitor a state of the first chiplet, where the state may indicate an anomaly associated with the second chiplet. The AI accelerator may then select an action from a plurality of actions based at least in part on the state of the first chiplet, cause the action to be performed by the chiplet-based system, and execute a reinforcement learning algorithm update the plurality of actions based on a result of the action being performed.