Post-Quantum Cryptography (PQC)

Multivariate cryptography is based on the difficulty of solving systems of multivariate quadratic polynomials over finite fields. The security hinges on the MQ problem, which is also NP-hard.

• Typical Use: Primarily used for constructing digital signatures.
• Characteristics: Produces very short signatures and fast verification, but often has large public keys.
• Examples: Rainbow (a signature scheme that was a NIST finalist but later had its security parameters reduced due to cryptanalysis) and GeMSS. These schemes often involve generating a complex public map that is easy to evaluate but hard to invert without a secret 'trapdoor'.

Isogeny-based cryptography is a newer family that uses the mathematical theory of elliptic curves and the difficulty of computing isogenies (maps between curves) between supersingular elliptic curves.

• Core Problem: Supersingular Isogeny Diffie-Hellman (SIDH).
• Key Advantage: Exceptionally small key sizes compared to other PQC families.
• Recent Development: The original SIDH proposal (SIKE) was broken in 2022 using a new mathematical attack. Active research is focused on developing secure variants, such as CSIDH (which operates in the commutative setting), but these are less efficient and not yet ready for standardization.

While not a 'public-key' family, symmetric algorithms (AES, SHA-3) are generally considered quantum-resistant, though Grover's algorithm forces a doubling of key size for equivalent security. Therefore, PQC migration focuses on replacing public-key algorithms.

• Hybrid Deployment: The recommended practical approach is hybrid key exchange, where a classical algorithm (e.g., ECDH) and a PQC algorithm (e.g., Kyber) are both used. The final shared secret is derived from both results.
• Benefit: Maintains security even if one of the cryptographic systems is broken, providing a robust transition path. Major protocols like TLS 1.3 and SSH are adopting hybrid modes.

A hardware-enforced security mechanism that ensures a device executes only cryptographically signed and verified firmware during its initial startup sequence. PQC is critical for its long-term security.

• Process: The bootloader verifies the digital signature of the next-stage firmware using a public key anchored in hardware before execution.
• Quantum Threat: Classical digital signatures (ECDSA, RSA-PSS) used in boot chains are vulnerable to quantum attacks. PQC digital signature algorithms (e.g., SPHINCS+, Falcon) are required for future-proofing.
• TinyML Impact: Guarantees the integrity of the machine learning model binary and inference runtime loaded onto the device.

A security exploit that extracts secret information (e.g., encryption keys) by analyzing indirect physical emissions from a device during cryptographic operations, such as power consumption, electromagnetic radiation, or timing.

• Relevance to PQC: New PQC algorithms have different computational profiles than classical crypto, potentially introducing novel side-channel vulnerabilities (e.g., via secret-dependent branches or memory access patterns).
• Common Types: Simple Power Analysis (SPA), Differential Power Analysis (DPA), and timing attacks.
• Mitigation: Constant-time implementation, algorithmic masking, and dedicated secure hardware are essential for PQC on exposed embedded devices.

A cryptographic mode of operation that simultaneously provides confidentiality (encryption), integrity, and authenticity for data. It is a fundamental primitive for securing data-in-transit and data-at-rest for TinyML.

• Classical Standard: AES-GCM is widely used but relies on AES, which, while quantum-resistant as a symmetric cipher, requires larger key sizes post-quantum (256-bit).
• PQC Modes: NIST is standardizing PQC authenticated key exchange and considering composite schemes that combine classical symmetric authenticated encryption with PQC key establishment.
• Use Case: Securing sensor data streams or model updates sent to/from a microcontroller.

A security process where a device cryptographically proves the integrity and authenticity of its currently running firmware and software state to a remote verifier.

• Mechanism: The device generates a signed report (attestation) of measured software hashes, using a key certified by a hardware root of trust.
• PQC Transition: The digital signature in the attestation report must be quantum-resistant. This requires integrating a PQC signature algorithm into the attestation protocol.
• TinyML Application: Critical for verifying that an on-device machine learning model has not been tampered with or replaced by a malicious version, enabling trust in distributed edge intelligence.