Seeded Random Memory Scanning for Escrow-Resistant Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing tampering detection systems face vulnerabilities to escrow attacks, with sequential scans being computationally minimal but susceptible, and random scans requiring significant calculation.
Innovation Solution
A detection system utilizing a node and verification server that employs a shuffle algorithm for random scanning, generating random values based on a seed to ensure robustness against escrow attacks with reduced computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If sequential scanning is used to generate a digest, then the computational complexity is minimal (O(m)), but the system becomes vulnerable to escrow attacks
Solution Approach 1:
The patent applies dynamics by transforming the static sequential scanning approach into a dynamic random scanning approach. The verifier generates random challenge values that dynamically determine which memory locations to scan, making the verification process adaptive and unpredictable. This dynamic approach increases security against escrow attacks while maintaining computational efficiency through selective random sampling rather than exhaustive scanning.
Solution Approach 2:
The patent changes the fundamental parameter of the scanning approach from sequential (fixed pattern) to random (variable pattern). By introducing randomness as a parameter change, the system achieves better security properties against escrow attacks where an attacker might predict or cache sequential access patterns. The random scanning with challenge-response mechanism changes the verification paradigm while keeping computational complexity manageable.
2Reliability
If random scanning is used to generate a digest, then the system becomes robust against escrow attacks, but the computational complexity increases to O(mlog(m))
Solution Approach 1:
The patent applies partial action by performing random scanning on a selected subset of memory locations rather than requiring complete verification of all m memory cells. The challenge-response mechanism allows the verifier to randomly sample specific locations based on challenge values, achieving sufficient security through partial verification. This reduces computational complexity from O(mlog(m)) to a more manageable level while maintaining robustness against escrow attacks.
Solution Approach 2:
The patent uses copying in the sense that the verifier maintains a reference copy of the original digest computed from the trusted initial state. This reference digest is compared against the current digest computed through random scanning. The copying approach allows verification without requiring the verifier to store or process all original data, reducing computational overhead while maintaining security through cryptographic comparison of digest values.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A detection system includes a node (50) and a verification server (100). The node (50) generates a plurality of random values based on a seed received from the verification server (100), generates a response based on information on an area of the first storage content specified by the plurality of generated random values, and transmits the generated response to the verification server (100). The verification server (100) generates a plurality of random values based on the seed, and generates a verification response based on information on an area of the second storage content specified by the plurality of generated random values. The verification server (100) determines whether a response received from the node matches the verification response, and outputs a determination result.