AES as Error Correction: Cryptosystems for Reliable Communication
Open accessAbstract
In this letter, we show that the Advanced Encryption Standard (AES) cryptosystem can be utilized as an error-correcting code to obtain reliability over noisy communication and data systems. Specifically, we show that by composing a simple padding followed by encrypting with AES, we can achieve error-correcting performance similar to random codes. We conduct an empirical comparison between our AES-based error correction approach, Random Linear Codes, and CA-Polar codes to assess their performance in practical scenarios. Our results show that the AES-based approach achieves nearly identical performance to the other well-established error-correcting codes.
AI snapshot
Written from the paper’s full text.
TL;DR
The paper shows that cryptosystems with uniform or pseudouniform outputs, such as AES, can be used directly as error-correcting codes by padding the message and decoding with GRAND. Empirically, AES with zero padding performs almost the same as random linear codes and CA-Polar codes for a length-128, rate-0.91 setting.
Key findings
- The deterministic CEC scheme can asymptotically transmit at rate \(k/n\) with arbitrarily low error probability if \(k/n < C\), where \(C\) is the BSC or AWGN capacity, when the cryptosystem is a random function.
- The probabilistic CEC scheme can asymptotically transmit at rate \(k/(n+r)\) if \((k+r_0)/(n+r) < C\); the paper notes this bound “might not be tight.”
- In simulations with \(n=128\), \(k=116\), zero padding \(n-k=12\), and rate \(R=0.91\), AES as an error-correcting code achieved BER and BLER “almost the same” as RLC and CA-Polar codes.
- With hard-decision GRAND, RLC and AES achieved the same performance; CRC-based schemes using GRAND were slightly better at SNR higher than \(\sim 8\) dB.
- The proposed input plaintext counter (IP-CTR) mode avoids CBC-style error propagation and the CTR issue that plaintext is not an AES input, while obtaining the same error-correcting performance as the ECB-based scheme in Section IV-A.
Methods
- Theoretical: defines cryptosystems, random functions, and random codes, then proves that padding followed by a random-function cryptosystem yields a random code and thus approaches capacity under the noisy-channel coding theorem.
- Empirical: simulates an AWGN channel with BPSK modulation, comparing AES-FIPS-197 with zero padding, random linear codes, and CRC-aided polar codes, all of length \(n=128\) and rate \(R=0.91\).
- Decoding: uses a modified joint decryption-decoding scheme with GRAND and the soft-information variant ORBGRAND.
Limitations
- The paper states that security guarantees of the padding approach require further analysis and leaves this direction for future work.
- The paper states that the bound in Theorem 2 might not be tight, so higher rates than stated might be achievable.
- Evident from the text: the empirical evaluation is limited to AWGN/BPSK, a single short block length \(n=128\), rate \(R=0.91\), and does not include a security analysis; the paper also notes ECB mode does not hide raw-data patterns.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
0 of 12 cited sources checked · 0 supported · 0 flagged
- Findingown data
In practical scenarios, AES as error correction and Random Linear Codes and CA-Polar codes achieve almost the same performance.
“We empirically contrast the performance of the proposed approach using AES as error correction with that of Random Linear Codes and CA-Polar codes and show that in practical scenarios, they achieve almost the same performance.”
Quote in the paperSubstantial claim- Rests on an assumption: The AWGN/BPSK simulation setup represents practical scenarios.
- Findingown data
In IP-CTR mode, the same error correcting performance as shown in Section IV-A is obtained.
“We note that, as shown in Figure 6, in IP-CTR mode we obtain the same error correcting performance as shown in Section IV-A.”
Quote in the paperNot a substantial claim - Findingcitation
Coding schemes with CRC and using GRAND decoder can obtain slightly better performance in SNR higher than about 8 dB.
“However, we note that the coding schemes with CRC and using GRAND decoder can obtain slightly better performance in SNR higher than ∼ 8dB.”
Quote in the paperNot a substantial claim - Findingown data
The BER and BLER performance of AES as an error-correcting code is almost the same as those of an RLC and CA-Polar code.
“Indeed, we show that the bit error rate (BER) and block error rate (BLER) performance of AES as an error-correcting code is almost the same as those of an RLC and CA-Polar code.”
Quote in the paperSubstantial claim- Rests on an assumption: The AWGN/BPSK simulation setup represents practical scenarios.
- Findingown data
A simple padding followed by AES is an effective mechanism for error correction when combined with Guessing Additive Noise Decoding (GRAND).
“We empirically demonstrate our idea by showing that a simple padding followed by AES, is an effective mechanism for error correction when combined with Guessing Additive Noise Decoding (GRAND) [11]- [13], which can decode any code, including the nonlinear construct yielded by AES.”
Quote in the paperSubstantial claim- Rests on an assumption: AES's output is sufficiently random or pseudorandom to act as a random function for error correction.
- Findingown data
Using GRAND for hard-decision decoding, both the RLC code and AES as an error-correcting code achieve the same performance.
“Using GRAND for hard-decision decoding, both the RLC code and AES as an error-correcting code achieve the same performance as presented in Section III.”
Quote in the paperSubstantial claim - Findingown data
The performance of the error-correction scheme utilizing padding and AES is similar to that of Random Linear Codes and CA-Polar codes.
“Moreover, the performance of our errorcorrection scheme, utilizing padding and AES, is similar to that of Random Linear Codes (RLC), which are known to be capacity-achieving [14], and with CRC-Aided Polar (CA-Polar) codes [15].”
Quote in the paperSubstantial claim- Rests on an assumption: The AWGN/BPSK simulation setup represents practical scenarios.
- Findingown data
In practical scenarios, a cryptosystem such as AES, whose outcome is considered pseudo-random, has an error-correcting performance similar to RLCs.
“That is, we show that in practical scenarios, a cryptosystem such as AES, whose outcome is considered pseudo-random, has an error-correcting performance which is similar to random linear codes (RLCs).”
Quote in the paperSubstantial claim- Rests on an assumption: AES's output is sufficiently random or pseudorandom to act as a random function for error correction.
- Rests on an assumption: The AWGN/BPSK simulation setup represents practical scenarios.
- Findingreasoning
The deviation on the probability of error for a random code is bounded by P(P_e(C) >= 2cε) <= 1/c.
“P (P e (C) ≥ 2cε) ≤ E[P e (C] 2cε ≤ 2ε 2cε = 1 c”
Quote in the paperSubstantial claim - Findingown data
All schemes tested obtain almost the same performance for decoding with a soft-information decoder as proposed by ORBGRAND.
“All the schemes tested obtain almost the same performance for decoding with a soft-information decoder as proposed by ORBGRAND.”
Quote in the paperSubstantial claim - Premisecitation
Common block cipher modes proposed by NIST suffer from error propagation due to the ciphertext in one iteration being used as an input in the next.
“Common block cipher modes proposed by NIST suffer from error propagation due to the ciphertext in one iteration, being used as an input in the next.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [27]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Interpretationreasoning
If Crypt_pk is a random function and k/n < C, the deterministic CEC scheme can asymptotically transmit with arbitrarily low probability of error at rate k/n.
“Theorem 1.Let C be the capacity of the underlying BSC or AWGN channel, and suppose that Crypt p k is a random function.Then, if k n < C, the deterministic CEC scheme can asymptotically transmit with an arbitrarily low probability of error at a rate of k n .”
Quote in the paperSubstantial claim- The authors state this as an assumption: Crypt_pk is a random function.
- Interpretationreasoning
A modified counter mode of operation, input plaintext counter mode, allows AES to be utilized for multiple blocks while retaining its error-correcting capabilities.
“we present a modified counter mode of operation, named input plaintext counter mode, in order to utilize AES for multiple blocks while retaining its error correcting capabilities.”
Quote in the paperSubstantial claim- Rests on an assumption: AES's output is sufficiently random or pseudorandom to act as a random function for error correction.
- Interpretationreasoning
Simple padding followed by a cryptosystem with uniform or pseudouniform outputs can approach the error-correcting performance of random codes.
“we show that simple padding followed by a cryptosystem with uniform or pseudouniform outputs can approach the error-correcting performance of random codes.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Crypt_pk is a random function.
- Interpretationreasoning
In the CEC scheme, the code C = Crypt_pk(Pad(F^k)) is a random code.
“Since in the CEC scheme the cryptosystem Crypt is a random function as given in Definition 2, the code C is a random code as given in Definition 3.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Crypt_pk is a random function.
- Premisereasoning
ECB mode does not hide raw-data patterns.
“However, the main drawback of this mode is that it does not hide raw-data patterns, as illustrated in Figure 4.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [21]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Interpretationreasoning
If Crypt_pk is a random function and (k+r0)/(n+r) < C, the probabilistic CEC scheme can asymptotically transmit with arbitrarily low probability of error at rate k/(n+r).
“Theorem 2. Let C be the capacity of the underlying BSC or AWGN channel, and suppose that Crypt p k is a random function.Then, if k+r0 n+r < C, the probabilistic CEC scheme can asymptotically transmit with an arbitrarily low probability of error at a rate of k n+r .”
Quote in the paperSubstantial claim- The authors state this as an assumption: Crypt_pk is a random function.
- Interpretationnone given
AES belongs to the family of cryptosystems that can serve as error-correcting as defined in Section III.
“This is a common cryptosystem that belongs to the family of cryptosystems that can serve as error-correcting as defined in Section III.”
Quote in the paperNot a substantial claim- Rests on an assumption: AES's output is sufficiently random or pseudorandom to act as a random function for error correction.
- Premisecitation
Common block cipher modes like CBC suffer from error propagation.
“Common modes, like cipher block chaining (CBC), suffer from error propagation”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [16]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Interpretationreasoning
For large n, random codes are good with high probability.
“Thus, for large n, random codes are good with high probability.”
Quote in the paperSubstantial claim - Interpretationreasoning
By utilizing mere padding to manage code rate, followed by encryption with a cryptosystem with uniformly distributed outputs, one can build a high-performance error correction mechanism.
“We show that, by utilizing a mere padding in order to manage the code rate, followed by an encryption with a cryptosystem with uniformly distributed outputs, we can build a high performance error correction mechanism.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Crypt_pk is a random function.
- Interpretationown data
The AES cryptosystem can be used as an error-correcting code to obtain reliability over noisy communication and data systems.
“we show that the Advanced Encryption Standard (AES) cryptosystem can be used as an errorcorrecting code to obtain reliability over noisy communication and data systems.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Crypt_pk is a random function.
- Rests on an assumption: AES's output is sufficiently random or pseudorandom to act as a random function for error correction.
- Premisecitation
Counter (CTR) mode, because it does not input the plaintext into AES, is not well suited for the technique.
“while counter (CTR) mode, because it does not input the plaintext into AES, is not well suited for our technique.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [16]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Interpretationreasoning
A family of computational cryptosystems can be used as well-performing error-correcting codes.
“we characterize a family of computational cryptosystems that can potentially be used as well performing error correcting codes.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Crypt_pk is a random function.
- Premisenone given
In CTR mode, the plaintext itself is not an input of the AES algorithm.
“In CTR mode, however, the plaintext itself is not an input of the AES algorithm.”
Quote in the paperSubstantial claim - Premisecitation
No information is gained from the padding bits as they are used solely as a redundancy check.
“Thus, no information is gained from the padding bits as they are used solely as a redundancy check [20].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [20]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Interpretationreasoning
In CTR mode, padding with zeroes does not provide good error correction capabilities.
“Thus, in this case, padding with zeroes does not provide good error correction capabilities.”
Quote in the paperSubstantial claim - Premisenone given
Error propagation does not occur for CTR mode.
“This error propagation does not occur for CTR mode.”
Quote in the paperSubstantial claim - Premisereasoning
In CBC mode, any error in decoding the first message is propagated to the second message.
“Thus, any error in decoding the first message is propagated to the second message.”
Quote in the paperSubstantial claim - Interpretationcitation
For a random code C with rate R = k/n < C, as n goes to infinity, the error probability P_e(C) goes to 0.
“Let C be the capacity of the underlying BSC or AWGN channel, and C be a random code with rate R = k n < C.Then, as n → ∞, the error probability P e (C) → 0.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [1]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Interpretationcitation
Cryptosystems with random or pseudo-random outputs, such as AES, can serve as error correction codes that can approach capacity.
“Cryptosystems that have outputs that are random or pseudo-random, say AES or as in [10], can serve as error correction codes that can approach capacity.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [10]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- The authors state this as an assumption: Crypt_pk is a random function.
- May combine two claims. 2
- Classified as probably not a substantial claim. 7
May combine two claims.
PremisecitationIn the asymptotic regime, random codes are capacity achieving for the BSC and AWGN channels.
“In the asymptotic regime, i.e., when n goes to infinity, random codes are capacity achieving for the BSC and AWGN channels.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [1]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
May combine two claims.
InterpretationreasoningIP-CTR mode avoids both error propagation and the plaintext being used as an input to AES.
“Thus, IP-CTR mode avoids both issues mentioned above.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationA pseudorandom function is computationally indistinguishable from a random function.
“It is important to note that a pseudorandom function, as usually utilized in classical cryptosystems, is computationally indistinguishable from a random function [17].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [17]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Classified as probably not a substantial claim.
Interpretationown dataThe empirical results validate the theoretical result given in Section III.
“We empirically validate the theoretical result given in Section III.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationReversing the ordering to encoding then encrypting was proposed in a setting where reliability and security can be obtained by partial encryption after error correction encoding.
“More recently, however, the reverse ordering of encoding and then encrypting was proposed in a setting where reliability and security can be obtained by partial encryption after error correction encoding [9].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [9]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Premisenone givenAny padding P of size n-k serves the purpose.
“Any padding P of size n-k serves our purpose.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationStructured codes can slightly outperform random codes in the tested rate and code-size regime when one utilizes a hard-decision decoder.
“This is settled with the known results in the literature, which show that structured codes can slightly outperform the performance of random codes in the tested rate and code-size regime when one utilizes a hard-decision decoder.”
Quote in the paperNot a substantial claimClassified as probably not a substantial claim.
PremisecitationThe traditional encoding process consists of two separated stages: encryption followed by an error correcting code.
“Thus, the traditional encoding process consists of two separated stages/systems: encryption followed by an error correcting code.”
Quote in the paperSubstantial claim0 of 2 sources checked- cites [7]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites [8]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Classified as probably not a substantial claim.
Premisenone givenThe output of ECB mode is the same whenever a message is repeated.
“This is because the output of the ECB mode is the same whenever a message is repeated.”
Quote in the paperSubstantial claim
The authors state this as an assumption.
Crypt_pk is a random function.
“suppose that Crypt p k is a random function”
The paper relies on this without support.
AES's output is sufficiently random or pseudorandom to act as a random function for error correction.
“whose outcome is considered pseudo-random”
The paper relies on this without support.
The AWGN/BPSK simulation setup represents practical scenarios.
“in practical scenarios”
“Our simulations employ a Gaussian channel model with AWGN and Binary Phase Shift Keying (BPSK) modulation.”
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Number of codewords per message uses output expansion instead of random-bit count
MinorThe paper allows r 0 < r and defines r 0 as the number of random input bits. Those inputs can produce at most 2^r0 codewords for a fixed message, not the stated 2^r when r 0 < r.
“In the probabilistic CEC scheme for every message M ∈ F k 2 the code C has 2 r codewords corresponding to M .”
“with r 0 ≤ r, where r 0 corresponds to the amount of random bits (r 0 = 0 for deterministic schemes) used by the algorithm and r corresponds to a possible increase in the output space.”
Cites
A Mathematical Theory of Communication
Shannon, C. E. (1948) · Bell System Technical Journal · doi:10.1002/j.1538-7305.1948.tb01338.x · Cited by 88,486 · see citing works
A Mathematical Theory of Communication
Shannon, C. E. (1948) · Bell System Technical Journal · doi:10.1002/j.1538-7305.1948.tb00917.x · Cited by 49,574 · see citing works
Arıkan, E. (2009) · IEEE Transactions on Information Theory · doi:10.1109/tit.2009.2021379 · Cited by 4,470 · cited 1× · see citing works
“• CRC-Aided Polar (CA-Polar) Codes [11], [16], which are Polar codes [17], [18] with an outer CRC code.”
BackgroundDaemen, J., & Rijmen, V. (2002) · Information Security and Cryptography · doi:10.1007/978-3-662-04722-4 · Cited by 2,437 · see citing works
Elements of Information Theory
Nalewajski, R. F. (2011) · Springer eBooks · doi:10.1007/978-3-642-20180-6_8 · Cited by 2,310 · see citing works
Tal, I., & Vardy, A. (2015) · IEEE Transactions on Information Theory · doi:10.1109/tit.2015.2410251 · Cited by 1,824 · see citing works
Error Control Coding: Fundamentals and Applications
Darnell, M. (1985) · IEE Proceedings F Communications Radar and Signal Processing · doi:10.1049/ip-f-1.1985.0011 · Cited by 991 · see citing works
Advanced Encryption Standard (AES)
(2001) · doi:10.6028/nist.fips.197 · Cited by 700 · see citing works
Recommendation for block cipher modes of operation :
Dworkin, M. J. (2001) · doi:10.6028/nist.sp.800-38a · Cited by 661 · see citing works
Cyclic Codes for Error Detection
Peterson, W. W., & Brown, D. (1961) · Proceedings of the IRE · doi:10.1109/jrproc.1961.287814 · Cited by 618 · see citing works
Tal, I., & Vardy, A. (2011) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit.2011.6033904 · Cited by 538 · cited 3× · see citing works
“• CRC-Aided Polar (CA-Polar) Codes [11], [16], which are Polar codes [17], [18] with an outer CRC code.”
BackgroundMethodReport on the development of the Advanced Encryption Standard (AES)
Nechvatal, J., Barker, E., Bassham, L. E., Burr, W. E., Dworkin, M. J., Foti, J., & Roback, E. (2001) · Journal of Research of the National Institute of Standards and Technology · doi:10.6028/jres.106.023 · Cited by 480 · cited 2× · see citing works
“, Advanced Encryption Standard (AES) [5], [6], are used in a pre-process to encrypt the information [7], [8].”
BackgroundCapacity-Achieving Guessing Random Additive Noise Decoding
Duffy, K. R., Li, J., & Medard, M. (2019) · IEEE Transactions on Information Theory · doi:10.1109/tit.2019.2896110 · Cited by 223 · cited 10× · see citing works
“For decoding, we use a modified version of a joint decryptiondecoding scheme with GRAND decoder [13], [15] as presented in Section III.”
BackgroundMethodInfluentialSelf-citationChannel polarization: A method for constructing capacity-achieving codes
Arıkan, E. (2008) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit.2008.4595172 · Cited by 165 · cited 1× · see citing works
“• CRC-Aided Polar (CA-Polar) Codes [11], [16], which are Polar codes [17], [18] with an outer CRC code.”
BackgroundOrdered Reliability Bits Guessing Random Additive Noise Decoding
Duffy, K. R., An, W., & Médard, M. (2022) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2022.3203251 · Cited by 131 · see citing works
Self-citationSoft Maximum Likelihood Decoding using GRAND
Solomon, A., Duffy, K. R., & Médard, M. (2020) · International Conference on Communications · doi:10.1109/icc40277.2020.9149208 · Cited by 117 · cited 1× · see citing works
“To decode the transmitted message, we employ both hard- and soft-detection variants of the recently developed Guessing Additive Noise Decoding (GRAND) [10], [11], [12],”
MethodSelf-citationThe random coding bound is tight for the average code (Corresp.)
Gallager, R. G. (1973) · IEEE Transactions on Information Theory · doi:10.1109/tit.1973.1054971 · Cited by 104 · see citing works
An Introduction to Mathematical Cryptography
Silverman, J. H., Pipher, J., & Hoffstein, J. (2008) · Undergraduate texts in mathematics · doi:10.1007/978-0-387-77993-5 · Cited by 102 · see citing works
Concise Encyclopedia of Coding Theory
Huffman, W. C., Kim, J.-L., & Solé, P. (2021) · Chapman & Hall/CRC eBooks · doi:10.1201/9781315147901 · Cited by 93 · see citing works
Ordered Reliability Bits Guessing Random Additive Noise Decoding
Duffy, K. R. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9414615 · Cited by 78 · cited 4× · see citing works
“We also use a soft-information GRAND variant, Ordered Reliability Bits Guessing Random Additive Noise Decoding (ORBGRAND) [14], which is well suited to short, high-rate codes.”
MethodInfluentialSelf-citationEmpirical evidence concerning AES
Hellekalek, P., & Wegenkittl, S. (2003) · ACM Transactions on Modeling and Computer Simulation · doi:10.1145/945511.945515 · Cited by 76 · cited 1× · see citing works
“perform well as a pseudo-random number generator [25].”
MethodHigh-Throughput and Energy-Efficient VLSI Architecture for Ordered Reliability Bits GRAND
Abbas, S. M., Tonnellier, T., Ercan, F., et al. (2022) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2022.3153605 · Cited by 74 · cited 1× · see citing works
“in-silicon implementations [16], [17] and syntheses [18], [19],”
5G and Beyond: Fundamentals and Standards
(2021) · doi:10.1007/978-3-030-58197-8 · Cited by 67 · see citing works
Guessing noise, not code-words
Duffy, K. R., Li, J., & Médard, M. (2018) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit.2018.8437648 · Cited by 64 · cited 2× · see citing works
“For decoding, we use a modified version of a joint decryptiondecoding scheme with GRAND decoder [13], [15] as presented in Section III.”
BackgroundMethodSelf-citationA Fixed Latency ORBGRAND Decoder Architecture With LUT-Aided Error-Pattern Scheduling
Condo, C. (2022) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2022.3150583 · Cited by 63 · cited 1× · see citing works
“[20] in both hard and soft-detection settings have established the flexibility and energy efficiency of GRAND strategies.”
BackgroundHigh-Throughput VLSI Architecture for GRAND
Abbas, S. M., Tonnellier, T., Ercan, F., et al. (2020) · SiPS ... design and implementation - IEEE Workshop on Signal Processing Systems · doi:10.1109/sips50750.2020.9195254 · Cited by 62 · see citing works
Guessing Random Additive Noise Decoding With Symbol Reliability Information (SRGRAND)
Duffy, K. R., Médard, M., & An, W. (2021) · IEEE Transactions on Communications · doi:10.1109/tcomm.2021.3114315 · Cited by 59 · see citing works
Self-citationA Public-Key Encryption Scheme with Pseudo-random Ciphertexts
Möller, B. (2004) · Lecture notes in computer science · doi:10.1007/978-3-540-30108-0_21 · Cited by 58 · see citing works
CRC Codes as Error Correction Codes
An, W., Médard, M., & Duffy, K. R. (2021) · International Conference on Communications · doi:10.1109/icc42927.2021.9500279 · Cited by 55 · see citing works
Self-citationMulti-Code Multi-Rate Universal Maximum Likelihood Decoder using GRAND
Riaz, A., Bansal, V., Solomon, A., et al. (2021) · Proceedings of ESSCIRC · doi:10.1109/esscirc53450.2021.9567867 · Cited by 55 · cited 2× · see citing works
“in-silicon implementations [16], [17] and syntheses [18], [19],”
BackgroundSelf-citationKeep the Bursts and Ditch the Interleavers
An, W., Médard, M., & Duffy, K. R. (2022) · IEEE Transactions on Communications · doi:10.1109/tcomm.2022.3171798 · Cited by 47 · see citing works
Self-citationAny code of which we cannot think is good
Coffey, J. T., & Goodman, R. M. (1990) · IEEE Transactions on Information Theory · doi:10.1109/18.59944 · Cited by 45 · cited 3× · see citing works
“which are theoretically known to be good codes [21], [22], and CRC-Assisted Polar (CA-Polar) codes, which have recently been introduced into the 5G New Radio communications standard for all control channel communications [23].”
BackgroundShort Codes with Near-ML Universal Decoding: Are Random Codes Good Enough?
Papadopoulou, V., Hashemipour-Nazari, M., & Balatsoukas‐Stimming, A. (2021) · SiPS ... design and implementation - IEEE Workshop on Signal Processing Systems · doi:10.1109/sips52927.2021.00025 · Cited by 13 · see citing works
Partial Encryption after Encoding for Security and Reliability in Data Systems
Cohen, A., D’Oliveira, R. G. L., Duffy, K. R., & Médard, M. (2022) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit50566.2022.9834772 · Cited by 12 · cited 4× · see citing works
“We provide a new efficient communication system for encoding by AES as error correction and joint decryption-decoding based on the decoding scheme proposed recently in [9].”
BackgroundMethodInfluentialSelf-citationTal, I., & Vardy, A. (2012) · arXiv (Cornell University) · doi:10.48550/arxiv.1206.0050 · Cited by 12
Batteryless, Wireless, and Secure SoC for Implantable Strain Sensing
Abdelhamid, M. R., Ha, U., Banerjee, U., et al. (2022) · IEEE Open Journal of the Solid-State Circuits Society · doi:10.1109/ojsscs.2022.3230000 · Cited by 9 · see citing works
Capacity-achieving decoding by guessing noise
Duffy, K., Li, J.-G., & Médard, M. (2018) · arXiv.org · doi:10.48550/arxiv.1802.07010 · Cited by 4
Self-citationPartial Encryption after Encoding for Security and Reliability in Data Systems
Cohen, A., D’Oliveira, R. G. L., Duffy, K. R., et al. (2022) · Maynooth University ePrints and eTheses Archive (Maynooth University) · doi:10.48550/arxiv.2202.03002
Self-citationA Sub-0.8pJ/b 16.3Gbps/mm2 Universal Soft-Detection Decoder Using ORBGRAND in 40nm CMOS
Riaz, A., Yasar, A., Ercan, F., et al. (2023) · 2023 IEEE International Solid- State Circuits Conference (ISSCC) · doi:10.1109/isscc42615.2023.10067519
Good codes can be produced by a few permutations
Ahlswede, R., & Dueck, G. (1982) · IEEE Transactions on Information Theory · doi:10.1109/tit.1982.1056514
Large Deviations Behavior of the Logarithmic Error Probability of Random Codes
Tamir, R., Merhav, N., Weinberger, N., et al. (2020) · IEEE Transactions on Information Theory · doi:10.1109/tit.2020.2995136
Flagged first, then by citations
Research works printed but not linked: 5
- 2023, GRAND-Code
- 2014, AES
- 2019, NR Multiplexing and Channel Coding
- 2022, Eb/N0—Wikipedia the Free Encyclopedia
- stallings, 2006, Cryptography and Network Security 4/E
Cited by
Sorted by citations
Soft-Output (SO) GRAND and Iterative Decoding to Outperform LDPC Codes
Yuan, P., Médard, M., Galligan, K., et al. (2025) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2025.3530880 · Cited by 27 · see citing works
“Since this procedure does not depend on codebook structure, GRAND can decode any moderate redundancy code, even non-linear ones [29].”
Self-citationSoft decoding without soft demapping with ORBGRAND
An, W., Médard, M., & Duffy, K. R. (2023) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit54713.2023.10206762 · Cited by 23 · see citing works
Self-citationUsing channel correlation to improve decoding - ORBGRAND-AI
Duffy, K. R., Grundei, M., & Médard, M. (2023) · Globecom · doi:10.1109/globecom54140.2023.10437763 · Cited by 13 · see citing works
“Note that while we have used the basic ORBGRAND pattern generator that has been implemented in hardware [27] for the results shown here, there is no difficulty in using the more sophisticated statistical model that provides enhanced results at higher SNR [25].”
MethodSelf-citationSoft-Output Guessing Codeword Decoding
Duffy, K. R., Yuan, P., Griffin, J., et al. (2024) · IEEE Communications Letters · doi:10.1109/lcomm.2024.3516486 · Cited by 9 · see citing works
Self-citationGuessing Random Additive Noise Decoding of Network Coded Data Transmitted Over Burst Error Channels
Chatzigeorgiou, I., & Savostyanov, D. (2024) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2024.3383546 · Cited by 5 · see citing works
“Since its inception, GRAND has been extended to non-linear codes [12], hard-input iterative decoding [13], soft detection decoding [14] and soft-input soft-output decoding [15].”
MethodSOGRAND Assisted Guesswork Reduction
Rapp, L., Médard, M., & Duffy, K. R. (2025) · IEEE Communications Letters · doi:10.1109/lcomm.2025.3607769 · Cited by 3 · see citing works
“Guessing Random Additive Noise Decoding (GRAND) [1] is a class of algorithms that can decode any moderate redundancy error correction code of any structure, including nonlinear codes, e.g., [2].”
Self-citationWang, Z., Shao, M., Xu, W., et al. (2024) · Symmetry · doi:10.3390/sym16050582 · Cited by 3 · see citing works
Laplacian-ORBGRAND: Decoding for Impulsive Noise
Feng, J., Duffy, K. R., & Médard, M. (2024) · IEEE Military Communications Conference (MILCOM) · doi:10.1109/milcom61039.2024.10773790 · Cited by 2 · see citing works
“Laplacian pdf has higher value near the mean and a heavier tail compared to Gaussian as shown in the large plot and the zoom-in. variants of GRAND provide near Maximum Likelihood de-coder and are computationally efficient for any moderate redundancy code of any structure [27], [28].”
Self-citationIGRAND Query Reduction using Even Codes
Riaz, A., Feng, J., Galligan, K., et al. (2025) · SiPS ... design and implementation - IEEE Workshop on Signal Processing Systems · doi:10.1109/sips66314.2025.11261257 · Cited by 2 · see citing works
“This approach is practical for any code of any structure with moderate redundancy, including non-linear ones [9].”
Self-citationHighly Reliable and Secure System With Multi-Layer Parallel LDPC and Kyber for 5G Communications
Nguyen, L., Phan, Q. B., & Nguyen, T. T. (2024) · IEEE Access · doi:10.1109/access.2024.3485875 · Cited by 2 · see citing works
“Various cryptographic techniques, including the data encryption standard (DES) and advanced encryption standard (AES) [36], [37], have been instrumental in safeguarding data privacy and confidentiality.”
MethodFinite-Length Coding Bounds via Guesswork
Mariona, A., Duffy, K. R., & Médard, M. (2024) · Allerton Conference on Communication, Control, and Computing · doi:10.1109/allerton63246.2024.10735309 · Cited by 1 · see citing works
“GRAND has been used with random linear codes (RLCs) [10], [11] and with coding schemes based on cryptosystems, such as the Advanced Encryption Standard (AES) [13], and Elliptic Curve Cryptography (ECC) [14].”
MethodSelf-citationUsing a Single Parity-Check to Reduce the Guesswork of Guessing Codeword Decoding
Griffin, J., Yuan, P., Duffy, K. R., et al. (2025) · Conference on Information Sciences and Systems · doi:10.1109/ciss64860.2025.10944685 · Cited by 1 · see citing works
“…as in Guessing Random Additive Noise Decoding (GRAND) [1], enables the decoding of any code that has an efficient codebook membership test, and has been demonstrated on code structures including binary linear codes [2], nonlinear codes [3], codes in higher-order fields, and codes based on…”
Self-citationError Correction Capabilities of Non-Linear Cryptographic Hash Functions
Cohen, A., & D’Oliveira, R. G. L. (2024) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit57864.2024.10619370
Self-citationSCAM-QUAVs: Side Channel Attacks Mitigation and Countermeasures of Quantum-Safe UAVs
Kumar, G. H., Chattaraj, D., Saini, D. K. J. B., Kulkarni, N., Tadkal, S., & Ranjima, P. (2025) · IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT) · doi:10.1109/conecct65861.2025.11306697
“Hardware-based strategies include dual-rail precharge logic, which uses complementary logic to balance the power consumption, and asynchronous logic, which eliminates clock signals to reduce timing-based leaks [28].”
Learning Binary Autoencoder-Based Codes with Progressive Training
Ninković, V., & Vukobratović, D. (2025) · Telecommunications Forum (TELFOR) · doi:10.1109/telfor67910.2025.11314356
“In [13], the Advanced Encryption Standard (AES) cryptosystem was shown to function as an error-correcting mechanism, achieving reliability comparable to random linear codes through a simple padding–encryption composition.”
The proposed reliable network 3-dimension RSA cryptosystem
Ajeena, R. K. K. (2026) · AIP conference proceedings · doi:10.1063/5.0317990
Guessing random additive noise decoding for digital data communication
Duffy, K. R., & Médard, M. (2026) · Foundations and Trends® in Integrated Circuits and Systems · doi:10.1108/ftics-09-2025-0017
Self-citationSadikin, A. S. B. A. H., & Susanto, H. (2026) · Frontiers in Big Data · doi:10.3389/fdata.2026.1826953
“Previous studies have also examined the relationship between encryption reliability, steganography, and error correction in secure communication systems (Li et al., 2011; Cohen et al., 2022).”
PipSORE: pipelined AES hardware design with shared one-round key expansion
Mahmoud, M. M. A., Farraj, S. O. B., & Ali, A. F. (2026) · The Journal of Supercomputing · doi:10.1007/s11227-026-08793-y
- Authors
- Cohen, A., D’Oliveira, R. G. L., Duffy, K. R., Woo, J., & Medard, M.
- Affiliations
- Technion – Israel Institute of Technology (IL)
- Clemson University (US)
- Northeastern University (US)
- Massachusetts Institute of Technology (US)
- Research Laboratory of Electronics, MIT, Cambridge, MA, USA
- Venue
- IEEE Communications Letters
- Published
- 2023-08
- Volume
- 27
- Issue
- 8
- Pages
- 1964-1968
- ISSN
- 1089-7798 (ISSN-L)
- 1558-2558
- 2373-7891
- OpenAlex
- W4380303715
- Cited by
- 19
- References
- 31
- Citations by year
- 2026: 4 · 2025: 6 · 2024: 7 · 2023: 2
- FWCI
- 1.66
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Communications Letters (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Defense Advanced Research Projects Agency (number HR00112120008)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Coding theory and cryptography
- Physical Sciences › Computer Science › Computer Networks and Communications › Error Correcting Code Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Cryptographic Implementations and Security
- Keywords
- Advanced Encryption Standard, error correction, random linear codes, noisy communication, data reliability, padding, random codes
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-23.
Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-23.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
