Alejandro Cohen,Rafael G. L. D'Oliveira
Abstract:Linear hashes are known to possess error-correcting capabilities. However, in most applications, non-linear hashes with pseudorandom outputs are utilized instead. It has also been established that classical non-systematic random codes, both linear and non-linear, are capacity achieving in the asymptotic regime. Thus, it is reasonable to expect that non-linear hashes might also exhibit good error-correcting capabilities. In this paper, we show this to be the case. Our proof is based on techniques from multiple access channels. As a consequence, we show that Systematic Random Non-Linear Codes (S-RNLC) are capacity achieving in the asymptotic regime. We validate our results by comparing the performance of the Secure Hash Algorithm (SHA) with that of Systematic Random Linear Codes (SRLC) and S-RNLC, demonstrating that SHA performs equally.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is: **Does the non - linear cryptographic hash function (NL - CHF) have error - correction ability?** Specifically, the author explores whether the widely - used non - linear hash functions (such as SHA) can be used not only for error detection but also for error correction when transmitted through a noisy channel.
### Background and Motivation of the Main Problem
1. **Limitations of Existing Schemes**:
- In most current practical systems, in order to achieve error correction, a two - stage method is usually adopted: first apply a non - linear hash function, and then use a linear error - correction coding scheme.
- However, this two - stage method increases the complexity and latency of the system.
2. **Theoretical Basis**:
- Previous research has shown that linear hash functions have error - correction ability, and classical random codes (whether linear or non - linear) can reach the channel capacity asymptotically.
- However, there has been little research on the error - correction ability of non - linear hash functions before.
### Core Contributions of the Paper
1. **Prove that Non - linear Hash Functions Have Error - correction Ability**:
- The author uses the technology of multi - access channels to prove that non - linear hash functions (such as SHA) can reach the channel capacity asymptotically and thus have good error - correction ability.
2. **Propose a New Joint Error - correction and Hash - checking Scheme**:
- A highly - efficient decoding algorithm based on Guessing Random Additive Noise Decoding (GRAND) is proposed, which is combined with the hash function for authentication, thus achieving joint error - correction and hash - checking.
3. **Experimental Proof**:
- Through experiments, the performance of SHA1, Systematic Random Linear Codes (S - RLC) and Systematic Random Non - Linear Codes (S - RNLC) on the Additive White Gaussian Noise (AWGN) channel is compared. The results show that the error - correction performance of SHA1 is comparable to that of S - RLC and S - RNLC.
### Formula Summary
- **Channel Capacity** \( C \):
\[
C=\max_{p(x)}I(X; Y)
\]
where \( I(X; Y) \) is the mutual information quantity.
- **Typical Sequence Set** \( T^{(n)}_\epsilon \):
\[
T^{(n)}_\epsilon=\left\{(m_1, d_1, y, z):\left|-\frac{1}{n}\log p(m_1, d_1, y, z)-H(M, D, Y, Z)\right|\leq\epsilon\right\}
\]
- **Error Probability Analysis**:
\[
P_e^{(n)}\leq Q(E^c_{11})+Q(E_{i1})+Q(E_{1j})+Q(E_{ij})
\]
where \( Q(\cdot) \) represents the conditional probability given Alice's transmission sequence \((m_1, d_1)\).
### Conclusion
This paper proves that non - linear cryptographic hash functions (such as SHA) have error - correction ability and proposes an efficient joint error - correction and hash - checking scheme. This provides new ideas and methods for designing simpler and more efficient communication systems in the future.