Glossary of Key Terms
This glossary provides definitions for essential homomorphic encryption terminology.
- Homomorphic Encryption (HE)
- A form of encryption allowing computations directly on encrypted data (ciphertext) without decryption. The decrypted result matches the result of operations on plaintext.
- Plaintext
- The original, unencrypted data or message.
- Ciphertext
- The encrypted form of plaintext, rendered unintelligible without the correct decryption key.
- Encryption
- The process of converting plaintext into ciphertext using an algorithm and a key.
- Decryption
- The process of converting ciphertext back into plaintext using an algorithm and a key.
- Partially Homomorphic Encryption (PHE)
- HE schemes supporting only one type of operation (addition OR multiplication) unlimited times.
- Somewhat Homomorphic Encryption (SHE)
- HE schemes supporting limited different operation types due to noise accumulation, with bounded circuit depth.
- Fully Homomorphic Encryption (FHE)
- HE schemes supporting arbitrary unlimited computations (additions and multiplications) on ciphertexts.
- Bootstrapping
- A technique to refresh a ciphertext in FHE by homomorphically evaluating the decryption circuit, reducing noise and enabling further operations.
- Noise Management
- Managing accumulated noise in ciphertexts during operations. Excessive noise prevents correct decryption.
- Lattice-based Cryptography
- Cryptography based on presumed hardness of lattice problems. Forms the foundation of many modern HE schemes including FHE.
- Learning With Errors (LWE)
- A hard mathematical problem used as security foundation for lattice-based cryptosystems including many HE schemes.
- BFV Scheme
- A popular FHE scheme (Brakerski-Fan-Vercauteren) suited for exact integer arithmetic modulo a prime.
- CKKS Scheme
- An FHE scheme (Cheon-Kim-Kim-Song) designed for approximate arithmetic on real or complex numbers, suitable for machine learning.
- Circuit Depth
- The maximum number of sequential operations that can be performed in a computation. Limited in leveled SHE schemes by noise growth.
- Privacy-Enhancing Technologies (PETs)
- Technologies that protect personal data. HE is one example; others include differential privacy, zero-knowledge proofs, and secure multi-party computation.
Understanding these terms is foundational to engaging with HE concepts and the broader field of homomorphic encryption.