Chiplet Fingerprinting Through PDN Response for Tamper Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Heterogeneous die SoCs face increased security risks due to the flexibility of chiplet configurations, which allows for attacks like die swapping and interface tampering, and the power distribution network (PDN) complexity makes it difficult to detect tampering events.
Innovation Solution
A multi-chiplet system with a transmitter die providing a test signal to perturb the PDN, a receiver die receiving a response signal, and a machine learning classification algorithm to analyze the response signal and determine the authenticity of the chiplet, using a signature generated from time delays and amplitudes measured by time-to-digital converters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If heterogeneous die SoCs are used to increase design flexibility and overcome Moore's law limits, then design versatility is improved, but security vulnerability increases due to susceptibility to probing and die swap attacks at interconnects
Solution Approach 1:
The patent introduces an intermediary verification mechanism using the power distribution network as a sensing medium. The PDN acts as a mediator that detects physical tampering events (probing, die swap) by measuring changes in electrical characteristics, thereby providing security verification without adding complex active security components to the heterogeneous die system
2Adaptability or versatility
If the power distribution network complexity increases to support heterogeneous die architectures, then design capability is improved, but difficulty of detecting tampering events increases
Solution Approach 1:
The patent enables the power distribution network to perform dual functions: its primary function of power delivery and its secondary function of tampering detection. By utilizing the PDN's inherent electrical characteristics and having it 'self-report' through measurable parameter changes, the system eliminates the need for separate detection infrastructure, thereby reducing the difficulty of tampering detection despite increased system complexity
Solution Approach 2:
The patent replaces complex mechanical or electronic tampering detection systems with electrical measurements of the power distribution network. By substituting physical intrusion detection mechanisms with electrical characteristic analysis (voltage, current, impedance measurements), the system simplifies the detection process while maintaining effectiveness against various tampering attacks
Data Source
AI summary
The present disclosure describes a heterogeneous integration (HI) system including an interposer and a plurality of dies coupled by the interposer. The plurality of dies include a die, a transmitter die, and a receiver die. The transmitter die is configured to provide a test signal to the die to generate a perturbation in a power distribution network (PDN) of the HI system. The receiver die is configured to measure a response signal in response to the perturbation in the PDN and to determine an authenticity of the die based on a machine learning classification algorithm applied on the response signal. The disclosure also describes a method of testing the authenticity of the die and a method of training the machine learning classification algorithm.