A detailed primer on cryptographic principles, methods, and their essential role in modern cybersecurity – from symmetric keys to digital signatures.
Depending on the algorithm, cryptography can provide confidentiality, integrity, authenticity, and non-repudiation. The two most widespread cryptosystems are symmetric and asymmetric cryptography. Even if we do not notice it, cryptography is literally everywhere in the digital era. End-to-end cryptography is becoming the norm for certain types of communication such as instant messaging. Lastly, cryptography is not magic: it is possible to break it, in particular when it is not sufficiently well implemented.
Cryptography provides confidentiality, integrity, authenticity and non-repudiation
Cryptography refers to practices, means, methods, and techniques that transform data to provide confidentiality, integrity, authentication, and non-repudiation, or a mix of those. In contrast, cryptology is the broader science of secret messages, encompassing both cryptography and other disciplines such as the science of breaking cryptography, also known as cryptanalysis. Cryptography encompasses encryption and decryption. Encryption is the implementation of a cryptographic algorithm (cipher) to transform intelligible data (plaintext) into unintelligible data (ciphertext). Decryption is the algorithm that reverts it to its intelligible form. In cryptographic jargon: a cryptographic method uses a cipher to turn plaintext into ciphertext, and vice-versa.
Symmetric and asymmetric cryptography are the two main cryptosystems
There is a wide variety of cryptographic methods, also known as cryptosystems, with different strength and weaknesses. This diversity allows users to choose the most appropriate method for their needs, context, resources, and other constraints. Cryptographic methods involve techniques rooted in mathematics. The two most common families of cryptographic methods, symmetric and asymmetric cryptography. Other methods include homomorphic cryptography, as well as zero-knowledge proofs, secure-multi- party computation (SMPC), lattice-based cryptography, secret sharing, obfuscation, etc.
Both symmetric and asymmetric cryptography use a simple string of bits (zeros and ones) called a cryptographic key, as an input variable provided to a cryptographic algorithm to transform data in a unique manner. The size and security of the key are critical conditions for the security of the entire cryptographic process, although not the only ones.
With symmetric cryptography, the data is encrypted and decrypted using the same secret key, just like the same key can both lock and unlock the same door. Symmetric cryptography is reversible: decrypting ciphertext simply requires reversing the encryption process using a copy of the same key. The secret key is sometimes called a shared secret because the sending and receiving parties in a communication have to share it. Symmetric encryption ensures the confidentiality of information in transit or at rest but does not ensure its integrity, authenticity and non-repudiation. Symmetric cryptography is also called secret key cryptography. It has been used since antiquity in many parts of the world.
With asymmetric cryptography, a pair of keys mathematically related ensures the confidentiality, integrity, authenticity and non-repudiation of information in transit or at rest. One key is “public”, i.e. not secret, and the other is private, i.e. must be kept secret. Asymmetric cryptography is also called public key cryptography
Symmetric and asymmetric cryptography are complementary and generally serve different purposes in different contexts. Among their differences, asymmetric cryptography is more complex technically, organisationally, and operationally than symmetric cryptography.
To provide confidentiality, the data is encrypted by the sender with the receiver’s public key and decrypted by the receiver with her private key. To provide authenticity and integrity of data such as a message or file, the sender (or owner if the data is at rest) generates a digital signature by encrypting with her private key a mathematically generated unique digital code based on the message and appending the results, called a hash value, to the message before sending it to the recipient(s).
The recipient(s) can then decrypt the digital signature with the sender’s public key. If the public key decrypts the digital signature, they can trust the integrity and authenticity of the data. Otherwise, it is a proof that the authenticity and integrity are broken.
The digital signature also ensures that the sending party cannot deny having “signed” the message (i.e. non-repudiation) because the encryption process to digitally sign the data provides a very high level of assurance that it was really performed by whomever holds the private key. The confidence stems from the mathematical properties of the key pair and encryption/decryption algorithms, which make it quasi-impossible to tamper with if there are no exploitable weaknesses in the implementation of the cryptosystem.
There is no 100% risk-free cryptographic method. All cryptographic methods come with pros and cons, and when a method seems to eliminate a weakness compared to another one, it often shifts the risk elsewhere. It is therefore essential to analyse and understand the risk related to the use of a particular cryptographic method in each use case prior to choosing a cryptographic method, and to treat the risk associated with the chosen method to reduce it to an acceptable level.
A systematic and comprehensive risk analysis of both the cryptographic technique and how it would be implemented in each case is essential to choose the most appropriate one, and the implementation of a systematic and comprehensive risk management cycle is also crucial when cryptography is deployed and operated in practice. Naturally, participants also need to consider other factors such as cost, usability, and technical feasibility.
Asymmetric cryptography is well-suited for encrypting small amounts of data because its performance drops in proportion to the amount of data. This is why symmetric and asymmetric cryptography are often used in combination, typically in the key establishment (or agreement) process whereby two parties use asymmetric cryptography to agree on a symmetric key that they will then use during their communication. This is typically the case with hybrid protocols such as Transport Layer Security (TLS).
Digital Signatures and Non-Repudiation
Ensuring Integrity and Authenticity
To provide authenticity and integrity of data, the sender generates a **digital signature** by encrypting a mathematically generated unique digital code (a **hash value**) of the message with her **private key**. The recipient can then decrypt the digital signature with the sender’s **public key**.
If the public key successfully decrypts the digital signature, it is proof that the integrity and authenticity of the data are intact. Otherwise, it is a sign of tampering. The digital signature also ensures that the sending party cannot deny having “signed” the message (**non-repudiation**), due to the mathematical certainty that only the private key holder could have generated that signature.
The Importance of Risk Management
There is no 100% risk-free cryptographic method. All cryptographic methods come with pros and cons, and when a method seems to eliminate a weakness compared to another one, it often shifts the risk elsewhere. It is therefore essential to analyze and understand the risk related to the use of a particular cryptographic method in each use case prior to choosing one.
A **systematic and comprehensive risk analysis** of both the cryptographic technique and how it would be implemented in each case is crucial. The subsequent implementation of a systematic and comprehensive **risk management cycle** is also vital when cryptography is deployed and operated in practice.