Pith. sign in

REVIEW 4 major objections 6 minor 56 references

PoSyn: Secure Power Side-Channel Aware Synthesis

T0 review · 4 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read PoSyn claims a synthesis-time cell-mapping scheme that drives power side-channel leakage of cryptographic hardware toward zero while preserving function and area.

desk verdict PoSyn's synthesis-time cell-selection heuristic is worth a serious look, but its 'theoretical guarantee' is a circular assertion and the experimental narrative has unresolved tensions. read the letter →

arxiv 2506.08252 v1 pith:52W75N5Q submitted 2025-06-09 cs.CR cs.AR

Amisha Srivastava , Samit S. Miftah , Hyunmin Kim , Debjit Pal , Kanad Basu This is my paper
classification cs.CRcs.AR
keywords powerside-channelattackslogicsynthesisstandardcellmappingbipartitematchingdifferentialanalysiscorrelationmutualinformationcryptographichardware
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

PoSyn is a logic-synthesis method that changes how vulnerable RTL operations are translated into standard cells, with the aim of producing netlists whose power consumption carries almost no information about the secret key. The paper's central claim is that this cell-mapping choice can be optimized, without changing the cipher's function, so that the mutual information between key and observed power traces approaches zero. It backs the claim with measured DPA and CPA success rates as low as 3% and 6% across AES, RSA, PRESENT, and two post-quantum ciphers at 65nm, 45nm, and 15nm, plus TVLA results below the standard threshold. A sympathetic reader would care because today's standard countermeasure, masking, is expensive in area and can be weakened by synthesis optimization; PoSyn reports comparable or better attack resistance while cutting area overhead by up to 3.79x relative to masking schemes.

What carries the argument

The load-bearing object is the bipartite matching graph whose left vertices are the RTLIL cells of vulnerable components and whose right vertices are the valid standard-cell combinations generated for each cell. Each edge carries the cost function $C(A,S)=\sum(\alpha \cdot SV/DS_i + \beta \cdot IO \cdot C_i + \gamma \cdot F \cdot DS_i)$, where $SV$ marks sensitive variables, $IO$ counts intensive operations, $F$ is fanout, $DS$ is cell driving strength, and $C_i$ is cell capacitance; the weights are chosen by empirical search. A minimum-cost bipartite matching algorithm selects the assignment with lowest total cost, and simulated annealing is used earlier to prune the space of candidate cell combinations. This machinery does the paper's work because it turns an abstract reduce-leakage goal into a concrete, solvable optimization problem over the technology library.

What would settle it

Take a PoSyn-synthesized AES netlist, run a CPA attack with 400,000 traces instead of 4,000, and compute TVLA on the post-layout netlist; if the attack success rate rises substantially or any t-value exceeds +/-4.5, the claimed negligible leakage does not hold under stronger measurement. Equivalently, simulate the netlist with a glitch-aware power model and estimate $I(K,L)$ directly; if it stays bounded away from zero while $C(M)$ is minimized, the cost-to-leakage link is broken.

Watch

Extended reading notes

Core claim

The paper's core discovery claim is that side-channel resistance can be engineered at the RTL-to-netlist mapping step rather than by adding countermeasure logic. PoSyn identifies leak-prone RTL blocks, such as sensitive variables, encryption operations, and high-fanout components, translates the design to an intermediate logic representation, and enumerates functionally equivalent standard-cell combinations through direct and decomposed mappings. A cost function $C(A,S)=\sum(\alpha \cdot SV/DS_i + \beta \cdot IO \cdot C_i + \gamma \cdot F \cdot DS_i)$ scores each mapping, and minimum-cost bipartite matching selects the globally cheapest assignment. The paper argues that because power leakage depends on switching activity, capacitance, and driving strength, minimizing this cost increases the conditional entropy of the key given the leakage $H(K|L)$, driving $I(K,L)$ toward zero. It reports that resulting netlists pass equivalence checking and show negligible TVLA leakage across all benchmarks and libraries.

Load-bearing premise

The load-bearing premise is that the weighted sum of driving-strength, capacitance, and fanout terms in the cost function is a faithful stand-in for true information leakage, so that whatever minimizes that sum also drives the key-leakage mutual information to zero; the weights themselves are fitted empirically rather than derived.

Editorial extensions

If this is right

  • Cryptographic hardware can gain power side-channel resistance during logic synthesis itself, so no masking hardware or RTL rewriting is needed.
  • The reported DPA and CPA success rates fall to 3% and 6%, and TVLA maximum t-values stay under 4.5 across all benchmarks, so standard first-order attacks would need far more traces or fail outright.
  • The method transfers across algorithm families and technology nodes, suggesting it is a library-level rather than algorithm-specific fix.
  • Compared with first-order masking and shuffling, PoSyn claims up to 72% lower attack success rates and up to 3.79x better area, at the cost of timing overhead up to 22% and memory overhead up to 15%.
  • Because the framework accepts any standard-cell library and validates functional equivalence with a logic-equivalence checker, it slots into existing synthesis flows rather than requiring a new design methodology.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The reported success rates depend on the 4000-trace attack setup; if the same netlists are attacked with orders of magnitude more traces or with glitch-aware leakage models, the practical margin may shrink even if the cost minimization is doing real work.
  • The theoretical claim that $I(K,L)\to 0$ is conditional on leakage being fully captured by the three cost terms; extending the same synthesis approach to electromagnetic or timing side channels would require adding analogous terms, which the paper does not do.
  • A plausible next experiment is combining PoSyn with masking; the paper's own hybrid result shows a small success-rate increase over PoSyn alone, which is counterintuitive under independent-leakage assumptions and worth investigating rather than treating as noise.
  • If confirmed, the method implies a general design principle: leakage can be treated as a synthesis objective on a Pareto frontier with area and delay, so security-aware EDA could expose a tunable leakage knob to designers.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. PoSyn is a logic-synthesis framework that maps vulnerable RTL components to standard-cell combinations via bipartite matching, using the cost function C(A,S) of Eq. (4) to select mappings that allegedly minimize power side-channel leakage. The paper claims a theoretical guarantee that PoSyn minimizes mutual information I(K,L) between the key and leakage, driving I(K,L) to zero, and validates the approach through DPA and CPA success rates and TVLA on AES, RSA, PRESENT, Saber, and Kyber across 65nm, 45nm, and 15nm libraries. Reported results include DPA success rates as low as 3%, CPA rates as low as 6%, and large reductions relative to conventional synthesis and to masking/shuffling countermeasures.

Significance. If the security claim were established, a synthesis-level countermeasure with low area overhead and broad applicability across ciphers and technology nodes would be a valuable contribution to hardware security. The paper has concrete strengths: a wide benchmark suite (AES variants, RSA, PRESENT, two PQC schemes), evaluation at three technology nodes, comparison with masking and shuffling, and use of LEC and TVLA as validation tools. However, the central theoretical claim in Section VI is not a derivation: it defines leakage as a function of the very cost function being minimized and then asserts convergence of mutual information to zero. The nonzero attack success rates in Table II and the anomalous masking-plus-PoSyn result in Table V are in tension with the advertised 'negligible leakage' guarantee. As it stands, the contribution is a heuristic mapping technique with empirical support, not a proven side-channel security mechanism.

major comments (4)
  1. [Section VI, Eqs. (7)-(8)] The claimed theoretical guarantee is an assertion, not a proof. Equation (7) posits L = f(C(M)) with the function f left completely unspecified, and Eq. (8) then concludes H(K|L) -> H(K) and I(K,L) -> 0. No quantitative relationship is established between the weighted structural cost C(A,S) of Eq. (4) and the conditional entropy H(K|L) of the key given physical power traces. Minimizing a weighted sum of driving strength, capacitance, operation count, and fanout does not, by itself, imply that power traces become statistically independent of the key. Because the weights alpha, beta, and gamma are selected by grid search (Section IV-D3), the claimed guarantee is not parameter-free, and no sensitivity or robustness analysis is provided.
  2. [Section VI, Eq. (7)] The argument is circular. Leakage L is defined as a function of C(M), which is exactly the objective that PoSyn minimizes through the Hungarian-algorithm-based bipartite matching. The conclusion that minimizing C(M) drives I(K,L) to zero therefore reduces, by the paper's own definitions, to the statement that minimizing the cost function reduces a quantity defined as a function of that cost function. A genuine security proof would need an independent leakage model, for example a stochastic power-trace distribution conditioned on key-dependent intermediate values, and a derivation showing that the PoSyn mapping minimizes the relevant mutual information or a proven upper bound on it.
  3. [Table II and Table V] The empirical results contradict the theoretical claim of negligible leakage. Table II reports PoSyn DPA success rates between 3% and 16% and CPA success rates between 6% and 22% across benchmarks and technology nodes; a nonzero key-recovery rate is inconsistent with the assertion that I(K,L) -> 0 and that 'the side-channel leakage conveys negligible information regarding the cryptographic key.' Additionally, Table V reports that combining first-order masking with PoSyn increases the DPA success rate from 3% to 5% relative to PoSyn alone; this unexpected result is not explained and further undermines confidence in the leakage-reduction mechanism. The authors should reconcile these measurements with the strong theoretical statements in the abstract, Section VI, and the conclusion.
  4. [Section III vs. Section VI] The threat model defines the adversary as observing power traces during encryption, but the theoretical analysis uses a leakage model L = f(C(M)) in which L is a deterministic function of a static structural cost. No connection is made between this deterministic structural quantity and the stochastic, measurement-noise-limited power traces that the adversary actually observes. The paper needs at least an explicit power model (e.g., Hamming-weight or toggle-count based) and a statement of how the mapping cost bounds the statistical dependence between that power model and the key; without this, the security analysis is disconnected from the attack model.
minor comments (6)
  1. [Abstract and Section I] The abstract and the contribution list cite '3.79 times' area improvement, while the contribution bullet in Section I states '3.70× reduction'; Table IV shows 3.79x only for the 3-Share AES implementation, not as a general result. Please unify these numbers and state the specific comparison baseline.
  2. [Section V] The paper states that LEC with Synopsys Formality is performed to verify functional equivalence, but Section VII reports no LEC results. Please either include the verification outcomes or clarify where they appear.
  3. [Section VII-C and Figure 5] The TVLA presentation reports only the maximum absolute t-value per benchmark-library combination. A proper TVLA assessment should report the number of traces, the grouping procedure (fixed vs. random), and the distribution of t-values over time samples, since a single maximum can be sensitive to outliers or to alignment choices.
  4. [Section VI, Eq. (5)] Mutual information is conventionally written I(K;L), not I(K,L); the comma notation is nonstandard and could be confused with the joint entropy or a bivariate function.
  5. [Throughout] There are minor typographical and consistency issues, including 'exibit' in Section VII-C, inconsistent capitalization of 'PoSyn' versus 'Posyn' in the Figure 4 caption, and the phrase 'PoSyn achieves notably lowers the success rates' in the abstract. These should be corrected in a revision.
  6. [Table II and Eq. (9)] Success rates are reported without confidence intervals or the number of attack attempts, despite Eq. (9) defining success rate as a fraction of attempts. Reporting the number of trials and, ideally, bootstrap confidence intervals would strengthen the empirical claims.

Circularity Check

2 steps flagged · score 7.0 of 10

Section VI's 'theoretical proof' that PoSyn minimizes mutual information leakage is forced by the paper's own definitions: Eq. (7) sets L=f(C(M)) with f unspecified, where C(M) is exactly the cost the Hungarian algorithm minimizes, and Eq. (8) then asserts H(K|L)→H(K) without any bound linking the fitted structural cost C(A,S) to conditional entropy.

  1. self definitional [Section VI, 'Mutual Information and Leakage Reduction' through 'Impact on Entropy and Mutual Information', Eqs. (6)-(8)]
    "Since power side-channel leakage L arises from variations in power consumption across different operations, it depends on the underlying physical implementation of the design. ... Therefore, we model leakage as a function of the total mapping cost: L=f(C(M)) (7) ... As a result of the targeted reductions in C(M), PoSyn increases H(K|L) such that: H(K|L)→H(K)⇒I(K, L)→0 (8)"

    Eq. (7) defines leakage L as a function of C(M), and Eq. (6) defines C(M) as the total cost PoSyn's Hungarian algorithm minimizes. The guarantee of Eq. (8) — minimizing C(M) makes H(K|L)→H(K) and I(K,L)→0 — is a restatement of that defining identification, not a derivation: leakage is characterized as whatever C measures, so minimizing C 'minimizes leakage' by construction. No bound relates the structural cost C(A,S) (Eq. 4) to the conditional entropy H(K|L) of the physical power-leakage process, and f is unspecified. The three bullet justifications ('Term #1... decreasing the correlation between power traces and key-dependent activity') assert the conclusion rather than derive it. Hence the Abstract's claim that PoSyn 'theoretically establishes...

  2. fitted input called prediction [Section IV-D3, 'Derivation of the Cost Function' and Section VI, 'Mutual Information and Leakage Reduction']
    "The weighting factors α, β, γ are introduced to appropriately scale the impact of each term based on its significance. These factors are determined through empirical methods such as grid search or optimization based on design-specific power and performance requirements."

    The cost function whose minimization Section VI equates with driving I(K,L)→0 is an empirically fitted heuristic: α, β, γ are determined by grid search/optimization, not derived from any leakage model. Section VI then states the MI result 'Based on derivation of Equation 4', so the 'proof' rests on a fitted equation whose parameters are never tied to H(K|L) by a stated argument. The validation numbers (DPA/CPA success rates, TVLA) are measured on netlists produced with this same fitted cost, so the claimed guarantee co-varies with the empirical tuning; absent a stated grid-search objective, a validation split, or a parameter-free bound, the Section VI result is a description of the fitted model expressed in information-theoretic language rather than an independent prediction.

full rationale

PoSyn's Section VI 'theoretical proof' is the paper's stated second contribution ('We theoretically prove that PoSyn is secure, minimizing mutual information leakage'). Tracing the derivation: Section IV-D3 builds C(A,S) as a weighted sum of library/design features with grid-searched weights α, β, γ; Section VI defines the total cost C(M) (Eq. 6), then defines leakage L=f(C(M)) with f unspecified (Eq. 7), and concludes H(K|L)→H(K), i.e., I(K,L)→0 (Eq. 8). The conclusion is generated by that definition: if leakage is declared to be a function of the very cost being minimized, then 'PoSyn minimizes leakage' holds by construction, and the mutual-information statement is an assertion rather than a theorem, since no estimate relates C(A,S) to the physical leakage distribution or to H(K|L). This is the self-definitional pattern, aggravated by the empirical fitting of α, β, γ (fitted input called prediction). The same-author citation [28] for leaky-module identification is a preprocessing input, not the load-bearing step, so it adds no score. The paper does contain genuine external evidence: DPA/CPA success rates (Table II) and TVLA t-values (Fig. 5) are measured, not derived, which supports the heuristic's practical effect and prevents a full 9-10 circularity verdict. Separately, the nonzero success rates (3-16% DPA, 6-22% CPA) and the Table V anomaly (masking+PoSyn raises DPA from 3% to 5%) are consistency tensions with the I→0 claim, and the Section VII-D throughput disclosure is unrelated to circularity. Net: the headline theoretical security guarantee reduces by construction to the paper's own Eq. (7), so the circularity score is 7.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central claim rests on a heuristic cost function whose weights are grid-search fitted, plus a definitional link between that cost and information leakage. The paper also assumes that leakage is confined to user-identified vulnerable components and that TVLA non-detection implies no leakage, both unverified. No invented physical entities are introduced.

free parameters (4)
  • α (cost weight for SV/DS term) = not reported
    Determined via grid search per Section IV-D3; no values given; directly affects cell selection and thus leakage results.
  • β (cost weight for IO·C term) = not reported
    Same grid-search tuning; affects capacitance selections.
  • γ (cost weight for F·DS term) = not reported
    Same grid-search tuning; affects high-fanout drive strength selections.
  • combination exploration threshold (max cells per indirect mapping) = not reported
    Adaptively adjusted by simulated annealing (Section IV-C2); not specified; controls the search space and area/leakage balance.
assumptions (4)
  • ad hoc to paper Minimizing the weighted cost function C(A,S) monotonically reduces actual power side-channel information I(K,L).
    Section VI models leakage as L=f(C(M)) (Eq. 7) and then asserts I(K,L)→0 (Eq. 8) without derivation or empirical calibration; this is the load-bearing premise of the security guarantee.
  • domain assumption Leakage is dominated by the user-identified vulnerable components (Sboxes, round keys, high-fanout nets), and non-vulnerable parts synthesized conventionally introduce no exploitable leakage.
    Section IV-A states vulnerability identification is 'partially user-defined' and only those components are protected; the paper does not assess leakage from the remainder of the netlist.
  • domain assumption TVLA t-values below ±4.5 with 4000 traces imply negligible leakage.
    Standard TVLA practice (Refs [42]-[45]), but it is a statistical heuristic and depends on trace count; nonzero DPA/CPA success rates in Table II suggest leakage detection is incomplete.
  • standard math Hungarian algorithm and simulated annealing find globally optimal or near-optimal mappings for the stated cost.
    Standard combinatorial optimization results invoked in Sections IV-C2 and IV-D; not problematic.

how reviews work

0 comments
Cite this review

Pith. "Pith review of PoSyn: Secure Power Side-Channel Aware Synthesis." pith.science (2026). https://pith.science/paper/52W75N5Q

@misc{pith2026250608252,
  author       = {Pith},
  title        = {Pith review of: PoSyn: Secure Power Side-Channel Aware Synthesis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/52W75N5Q}},
  note         = {Machine review of arXiv:2506.08252}
}
read the original abstract

Power Side-Channel (PSC) attacks exploit power consumption patterns to extract sensitive information, posing risks to cryptographic operations crucial for secure systems. Traditional countermeasures, such as masking, face challenges including complex integration during synthesis, substantial area overhead, and susceptibility to optimization removal during logic synthesis. To address these issues, we introduce PoSyn, a novel logic synthesis framework designed to enhance cryptographic hardware resistance against PSC attacks. Our method centers on optimal bipartite mapping of vulnerable RTL components to standard cells from the technology library, aiming to minimize PSC leakage. By utilizing a cost function integrating critical characteristics from both the RTL design and the standard cell library, we strategically modify mapping criteria during RTL-to-netlist conversion without altering design functionality. Furthermore, we theoretically establish that PoSyn minimizes mutual information leakage, strengthening its security against PSC vulnerabilities. We evaluate PoSyn across various cryptographic hardware implementations, including AES, RSA, PRESENT, and post-quantum cryptographic algorithms such as Saber and CRYSTALS-Kyber, at technology nodes of 65nm, 45nm, and 15nm. Experimental results demonstrate a substantial reduction in success rates for Differential Power Analysis (DPA) and Correlation Power Analysis (CPA) attacks, achieving lows of 3% and 6%, respectively. TVLA analysis further confirms that synthesized netlists exhibit negligible leakage. Additionally, compared to conventional countermeasures like masking and shuffling, PoSyn significantly lowers attack success rates, achieving reductions of up to 72%, while simultaneously enhancing area efficiency by as much as 3.79 times.

Figures

Figures reproduced from arXiv: 2506.08252 by the authors.

Figure 1
Figure 1. Overview of the proposed side-channel aware synthesis [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Optimal Bipartite Matching: The grey lines from Set 1 [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. DPA results for Side-Channel Resistant vs Conventional Netlist: Graphs showcase the success rates for the attack across [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: CPA results for Side-Channel Resistant vs Conventional Netlist: Graphs showcase the success rates for the attack across [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 5
Figure 5. Figure 5: Maximum absolute t-values from TVLA for crypto [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

56 extracted references · 54 canonical work pages

  1. [1]

    Power side-channel attack analysis: A review of 20 years of study for the layman,

    M. Randolphet al., “Power side-channel attack analysis: A review of 20 years of study for the layman,”Cryptography, 2020

    work page 2020

  2. [2]

    Physical side-channel attacks on cryptographic systems,

    N. P. Smart, “Physical side-channel attacks on cryptographic systems,” Software Focus, vol. 1, no. 2, pp. 6–13, 2000

    work page 2000

  3. [3]

    Amasive: an adaptable and modular autonomous side-channel vulnerability evaluation framework,

    S. A. Huss, M. St ¨ottinger, and M. Zohner, “Amasive: an adaptable and modular autonomous side-channel vulnerability evaluation framework,” inNumber Theory and Cryptography: Papers in Honor of Johannes Buchmann on the Occasion of His 60th Birthday. Springer, 2013, pp. 151–165

    work page 2013

  4. [4]

    Scare: Side- channel analysis based reverse engineering for post-silicon validation,

    X. Wang, S. Narasimhan, A. Krishna, and S. Bhunia, “Scare: Side- channel analysis based reverse engineering for post-silicon validation,” in2012 25th International Conference on VLSI Design. IEEE, 2012, pp. 304–309

    work page 2012

  5. [5]

    Aes-based security coprocessor ic in 0.18-muhboxm cmos with resistance to differential power analysis side-channel attacks,

    D. D. Hwang, K. Tiri, A. Hodjat, B.-C. Lai, S. Yang, P. Schaumont, and I. Verbauwhede, “Aes-based security coprocessor ic in 0.18-muhboxm cmos with resistance to differential power analysis side-channel attacks,” IEEE Journal of Solid-State Circuits, vol. 41, no. 4, pp. 781–792, 2006

    work page 2006

  6. [6]

    A probing attack on aes,

    J.-M. Schmidt and C. H. Kim, “A probing attack on aes,” inInformation Security Applications: 9th International Workshop, WISA 2008, Jeju Island, Korea, September 23-25, 2008, Revised Selected Papers 9. Springer, 2009, pp. 256–265

    work page 2008

  7. [7]

    Test vector leakage assessment (tvla) methodology in practice,

    G. Becker, J. Cooper, E. DeMulder, G. Goodwill, J. Jaffe, G. Kenworthy, T. Kouzminov, A. Leiserson, M. Marson, P. Rohatgiet al., “Test vector leakage assessment (tvla) methodology in practice,” inInternational Cryptographic Module Conference, vol. 1001. sn, 2013, p. 13

    work page 2013

  8. [8]

    Power side channel attack analysis and detection,

    N. Gattuet al., “Power side channel attack analysis and detection,” in Proceedings of the 39th International Conference on Computer-Aided Design, 2020

    work page 2020

Show all 56 references
  1. [9]

    An implementation of des and aes, secure against some attacks,

    M.-L. Akkar and C. Giraud, “An implementation of des and aes, secure against some attacks,” inCryptographic Hardware and Embedded Systems—CHES 2001: Third International Workshop Paris, France, May 14–16, 2001 Proceedings 3. Springer, 2001, pp. 309–318

    2001

  2. [10]

    Provably secure masking of aes,

    J. Bl ¨omeret al., “Provably secure masking of aes,” inInternational workshop on selected areas in cryptography. Springer, 2004

    2004

  3. [11]

    Multiplicative masking and power analysis of aes,

    J. D. Goli ´c and C. Tymen, “Multiplicative masking and power analysis of aes,” inCryptographic Hardware and Embedded Systems-CHES 2002: 4th International Workshop Redwood Shores, CA, USA, August 13–15, 2002 Revised Papers 4. Springer, 2003, pp. 198–212

    2002

  4. [12]

    A side- channel analysis resistant description of the aes s-box,

    E. Oswald, S. Mangard, N. Pramstaller, and V . Rijmen, “A side- channel analysis resistant description of the aes s-box,” inFast Software Encryption: 12th International Workshop, FSE 2005, Paris, France, February 21-23, 2005, Revised Selected Papers 12. Springer, 2005, pp. 413–423

    2005

  5. [13]

    Securing the aes finalists against power analy- sis attacks,

    T. S. Messerges, “Securing the aes finalists against power analy- sis attacks,” inInternational Workshop on Fast Software Encryption. Springer, 2000, pp. 150–164

    2000

  6. [14]

    Masking and leakage- resilient primitives: One, the other (s) or both?

    S. Bela ¨ıd, V . Grosso, and F.-X. Standaert, “Masking and leakage- resilient primitives: One, the other (s) or both?”Cryptography and Communications, vol. 7, pp. 163–184, 2015

    2015

  7. [15]

    On the cost of lazy engineering for masked software implementations,

    J. Balasch, B. Gierlichs, V . Grosso, O. Reparaz, and F.-X. Standaert, “On the cost of lazy engineering for masked software implementations,” inSmart Card Research and Advanced Applications: 13th International Conference, CARDIS 2014, Paris, France, November 5-7, 2014. Revised...

    2014

  8. [16]

    Maskver: a tool helping designers detect flawed masking implementations,

    M. Tempelmeier and G. Sigl, “Maskver: a tool helping designers detect flawed masking implementations,” in2016 1st IEEE International Verification and Security Workshop (IVSW). IEEE, 2016, pp. 1–6

    2016

  9. [17]

    Glitch-resistant masking revisited: Or why proofs in the robust probing model are needed,

    T. Moos, A. Moradi, T. Schneider, and F.-X. Standaert, “Glitch-resistant masking revisited: Or why proofs in the robust probing model are needed,”IACR Transactions on Cryptographic Hardware and Embedded Systems, pp. 256–292, 2019

    2019

  10. [18]

    Automated gener- ation of masked hardware,

    D. Knichel, A. Moradi, N. M ¨uller, and P. Sasdrich, “Automated gener- ation of masked hardware,”Cryptology ePrint Archive, 2021

    2021

  11. [19]

    A simple power-analysis (spa) attack on implementa- tions of the aes key expansion,

    S. Mangard, “A simple power-analysis (spa) attack on implementa- tions of the aes key expansion,” inInformation Security and Cryptol- ogy—ICISC 2002: 5th International Conference Seoul, Korea, November 28–29, 2002 Revised Papers 5. Springer, 2003, pp. 343–358

    2002

  12. [20]

    Differential power analysis,

    P. Kocher, J. Jaffe, and B. Jun, “Differential power analysis,” in Advances in Cryptology—CRYPTO’99: 19th Annual International Cryp- tology Conference Santa Barbara, California, USA, August 15–19, 1999 Proceedings 19. Springer, 1999, pp. 388–397

    1999

  13. [21]

    Introduction to differential power analysis,

    P. Kocher, J. Jaffe, B. Jun, and P. Rohatgi, “Introduction to differential power analysis,”Journal of Cryptographic Engineering, vol. 1, pp. 5–27, 2011. 13

    2011

  14. [22]

    Correlation power analysis with a leakage model,

    E. Brier, C. Clavier, and F. Olivier, “Correlation power analysis with a leakage model,” inCryptographic Hardware and Embedded Systems- CHES 2004: 6th International Workshop Cambridge, MA, USA, August 11-13, 2004. Proceedings 6. Springer, 2004, pp. 16–29

    2004

  15. [23]

    Balance power leakage to fight against side-channel analysis at gate level in fpgas,

    X. Fang, P. Luo, Y . Fei, and M. Leeser, “Balance power leakage to fight against side-channel analysis at gate level in fpgas,” in2015 IEEE 26th International Conference on Application-specific Systems, Architectures and Processors (ASAP). IEEE, 2015, pp. 154–155

    2015

  16. [24]

    Shuffling against side-channel attacks: A comprehensive study with cautionary note,

    N. Veyrat-Charvillon, M. Medwed, S. Kerckhof, and F.-X. Standaert, “Shuffling against side-channel attacks: A comprehensive study with cautionary note,” inAdvances in Cryptology–ASIACRYPT 2012: 18th International Conference on the Theory and Application of Cryptology and Infor...

    2012

  17. [25]

    Perfectly secure multi- party computation and the computational overhead of cryptography,

    I. Damg ˚ard, Y . Ishai, and M. Krøigaard, “Perfectly secure multi- party computation and the computational overhead of cryptography,” inAnnual international conference on the theory and applications of cryptographic techniques. Springer, 2010, pp. 445–465

    2010

  18. [26]

    Synthesis of dual-rail adiabatic logic for low power security applications,

    M. Morrison and N. Ranganathan, “Synthesis of dual-rail adiabatic logic for low power security applications,”IEEE Transactions on Computer- Aided Design of Integrated Circuits and Systems, vol. 33, no. 7, pp. 975–988, 2014

    2014

  19. [27]

    Rtl-psc: Automated power side-channel leakage assess- ment at register-transfer level,

    M. Heet al., “Rtl-psc: Automated power side-channel leakage assess- ment at register-transfer level,” inIEEE VTS, 2019

    2019

  20. [28]

    Scar: Power side-channel analysis at rtl level,

    A. Srivastava, S. Das, N. Choudhury, R. Psiakis, P. H. Silva, D. Pal, and K. Basu, “Scar: Power side-channel analysis at rtl level,”IEEE Transactions on Very Large Scale Integration (VLSI) Systems, pp. 1–14, 2024

    2024

  21. [29]

    Yosys open synthesis suite,

    C. Wolf, “Yosys open synthesis suite,” 2016

    2016

  22. [30]

    Simulated annealing,

    D. Bertsimas and J. Tsitsiklis, “Simulated annealing,”Statistical science, vol. 8, no. 1, pp. 10–15, 1993

    1993

  23. [31]

    Optimization by simulated annealing,

    S. Kirkpatrick, C. D. Gelatt Jr, and M. P. Vecchi, “Optimization by simulated annealing,”science, vol. 220, no. 4598, pp. 671–680, 1983

    1983

  24. [32]

    An optimal algorithm for on-line bipartite matching,

    R. M. Karp, U. V . Vazirani, and V . V . Vazirani, “An optimal algorithm for on-line bipartite matching,” inProceedings of the twenty-second annual ACM symposium on Theory of computing, 1990, pp. 352–358

    1990

  25. [33]

    Finding all minimum-cost perfect matchings in bipartite graphs,

    K. Fukuda and T. Matsui, “Finding all minimum-cost perfect matchings in bipartite graphs,”Networks, vol. 22, no. 5, pp. 461–468, 1992

    1992

  26. [34]

    Equivalence checking using,

    S. Formality, “Equivalence checking using,” 2010

    2010

  27. [35]

    55/65/90nm tech- nologies,

    United Microelectronics Corporation (UMC), “55/65/90nm tech- nologies,” https://www.umc.com/en/Product/technologies/Detail/55 65 90nm, accessed: 2024-04-29

    2024

  28. [36]

    Openroad-flow-scripts: Nangate45 platform,

    T. O. Project, “Openroad-flow-scripts: Nangate45 platform,” https://github.com/The-OpenROAD-Project/OpenROAD-flow-scripts/ tree/master/flow/platforms/nangate45, 2024, accessed: 2024-04-29

    2024

  29. [37]

    Open cell library in 15nm freepdk technology,

    M. Martins, J. M. Matos, R. P. Ribas, A. Reis, G. Schlinker, L. Rech, and J. Michelsen, “Open cell library in 15nm freepdk technology,” in Proceedings of the 2015 Symposium on International Symposium on Physical Design, 2015, pp. 171–178

    2015

  30. [38]

    Verilog designs,

    “Verilog designs,” http://www.aoki.ecei.tohoku.ac.jp/crypto/web/cores. html, 2024

    2024

  31. [39]

    Higher-order glitch resistant implementation of the present s-box,

    T. De Cnuddeet al., “Higher-order glitch resistant implementation of the present s-box,” inBalkanCryptSec. Springer, 2015

    2015

  32. [40]

    Design space exploration of saber in 65nm asic,

    M. Imranet al., “Design space exploration of saber in 65nm asic,” in Proceedings of the 5th Workshop on Attacks and Solutions in Hardware Security, 2021

    2021

  33. [41]

    A hardware accelerator for polynomial multiplication operation of crystals-kyber pqc scheme,

    F. Yamanet al., “A hardware accelerator for polynomial multiplication operation of crystals-kyber pqc scheme,” in2021 IEEE DATE

  34. [42]

    Towards sound and optimal leakage detection procedure,

    A. A. Ding, L. Zhang, F. Durvaux, F.-X. Standaert, and Y . Fei, “Towards sound and optimal leakage detection procedure,” inSmart Card Research and Advanced Applications: 16th International Conference, CARDIS 2017, Lugano, Switzerland, November 13–15, 2017, Revised Selected Pap...

    2017

  35. [43]

    Simpler, faster, and more robust t-test based leakage detection,

    A. A. Ding, C. Chen, and T. Eisenbarth, “Simpler, faster, and more robust t-test based leakage detection,” inConstructive Side-Channel Analysis and Secure Design: 7th International Workshop, COSADE 2016, Graz, Austria, April 14-15, 2016, Revised Selected Papers 7. Springer, 20...

    2016

  36. [44]

    Leakage assessment methodology: Ex- tended version,

    T. Schneider and A. Moradi, “Leakage assessment methodology: Ex- tended version,”Journal of Cryptographic Engineering, vol. 6, pp. 85– 99, 2016

    2016

  37. [45]

    Fobos 3: An open-source platform for side-channel analysis and benchmarking,

    E. Ferrufino, L. Beckwith, A. Abdulgadir, and J.-P. Kaps, “Fobos 3: An open-source platform for side-channel analysis and benchmarking,” in Proceedings of the 2023 Workshop on Attacks and Solutions in Hardware Security, 2023, pp. 5–14

    2023

  38. [46]

    A very compact “perfectly masked

    “A very compact “perfectly masked” s-box for aes,” https://faculty.nps. edu/drcanrig/pub/acns2008corr.pdf

  39. [47]

    New first-order secure aes performance records,

    A. R. Shahmirzadi, D. Bo ˇzilov, and A. Moradi, “New first-order secure aes performance records,”Cryptology ePrint Archive, 2021

    2021

  40. [48]

    Guard- ing the first order: The rise of aes maskings,

    A. Askeland, S. Dhooghe, S. Nikova, V . Rijmen, and Z. Zhang, “Guard- ing the first order: The rise of aes maskings,” inInternational Conference on Smart Card Research and Advanced Applications. Springer, 2022, pp. 103–122

    2022

  41. [49]

    Masking aes with shares in hardware,

    T. De Cnudde, O. Reparaz, B. Bilgin, S. Nikova, V . Nikov, and V . Rijmen, “Masking aes with shares in hardware,” inInternational Con- ference on Cryptographic Hardware and Embedded Systems. Springer, 2016, pp. 194–212

    2016

  42. [50]

    Trade-offs for threshold implementations illustrated on aes,

    B. Bilgin, B. Gierlichs, S. Nikova, V . Nikov, and V . Rijmen, “Trade-offs for threshold implementations illustrated on aes,”IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 34, no. 7, pp. 1188–1200, 2015

    2015

  43. [51]

    A first-order sca resistant aes without fresh randomness,

    F. Wegener and A. Moradi, “A first-order sca resistant aes without fresh randomness,” inConstructive Side-Channel Analysis and Secure Design: 9th International Workshop, COSADE 2018, Singapore, April 23–24, 2018, Proceedings 9. Springer, 2018, pp. 245–262

    2018

  44. [52]

    Re-consolidating first-order masking schemes: Nullifying fresh randomness,

    A. R. Shahmirzadi and A. Moradi, “Re-consolidating first-order masking schemes: Nullifying fresh randomness,”IACR Transactions on Crypto- graphic Hardware and Embedded Systems, pp. 305–342, 2021

    2021

  45. [53]

    3-share threshold implementation of aes s-box without fresh randomness,

    T. Sugawara, “3-share threshold implementation of aes s-box without fresh randomness,”IACR Transactions on Cryptographic Hardware and Embedded Systems, pp. 123–145, 2019

    2019

  46. [54]

    An aes-based encryption algorithm with shuffling

    A. Yahya, A. M. Abdalla, H. Arabnia, and K. Daimi, “An aes-based encryption algorithm with shuffling.” inSecurity and Management, 2009, pp. 113–116

    2009

  47. [55]

    An area-efficient shuffling scheme for aes imple- mentation on fpga,

    Y . Wang and Y . Ha, “An area-efficient shuffling scheme for aes imple- mentation on fpga,” in2013 IEEE International Symposium on Circuits and Systems (ISCAS). IEEE, 2013, pp. 2577–2580

    2013

  48. [56]

    Ascad: Anssi sca database,

    “Ascad: Anssi sca database,” https://github.com/ANSSI-FR/ASCAD

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.