Group signatures (GS, Chaum and van Heyst, EUROCRYPT 1991) are digital signatures that allow a signer to anonymously prove the membership and also allow the special authority called the opener can identify the signer. Group signatures with message-dependent opening (GS-MDO, Sakai et al., Pairing 2012) weakened the power of the opener by introducing another authority called the admitter who issues a message-dependent token. It would be a natural research topic to clarify whether cryptographic primitives that are required to construct GS-MDO are stronger than those of GS or not, according to the enhanced functionality of GS-MDO. In this paper, we propose a generic construction of timed-release encryption (TRE) from GS-MDO. Note that Sakai et al. have shown that GS-MDO implies identity-based encryption (IBE), and Nakai et al. (IWSEC 2009) and Matsuda et al. (Pairing 2010) demonstrated generic constructions of TRE from IBE. Thus, we do not show any new result from the viewpoint of feasibility. We show that (1) GS-MDO directly implies TRE without employing the generic constructions of TRE from IBE, and (2) the proposed TRE construction provides public verifiability, that is not usually supported by TRE, because a TRE ciphertext is a group signature in our construction. We also introduce a new security notion which we call token unforgeability where no adversary can forge a token even the adversary has the opener’s secret key, and prove that token unforgeability is implied by opener anonymity which is a fundamental security notion of GS-MDO. Our result implies that GS-MDO is a very strong cryptographic primitive.

Ateniese et al. (CRYPTO 2019/JoC 2021) introduced a cryptographic primitive which they call matchmaking encryption (ME), and Identity-based ME (IB-ME) is its identity-based variant. IB-ME supports an equality matching where a sender (encryptor) indicates a receiver’s (decryptor’s) identity (rcv) in addition to their own ID (σ), and a receiver indicates a sender’s identity (snd) in addition to the own identity (ρ). A ciphertext is decrypted if (σ, ρ) = (snd, rcv). In this paper, we pay attention to the search condition of public key authenticated encryption with keyword search (PAEKS) (Huang-Li, Information Sciences 2017) is reminiscent of the equality matching. We introduce a public key variant of ME which we call PK-ME, and propose a generic construction of PK-ME from PAEKS. As a conceptual contribution, our work lies in revealing a connection between ME and public key searchable encryption, which were independently researched so far. Due to the generic construction of PAEKS (Li-Boyen, IACR CiC 2024), we can instantiate the proposed generic construction from pairings or lattices. We also introduce a weaker version of authenticity and show that it can be instantiated from several complexity assumptions. Finally, we discuss the advantage/disadvantage of PK-ME compared to IB-ME.

Hopfield networks have been extended to multistate Hopfield networks, since complex-valued Hopfield networks (CVHNs) were proposed. Multistate Hopfield networks have often been applied to storage of multilevel data, such as images. Cyclic-group-ring-valued Hopfield networks (CGRVHNs) are multistate Hopfield networks proposed in the framework of group-ring-valued Hopfield networks. In this paper, we prove that a CGRVHN decomposes to a CVHN and a hyperbolic-valued Hopfield network (HVHN), where the decomposition is not unique. Therefore, the CGRVHNs are equivalent to mixed Hopfield networks of CVHNs and HVHNs. Moreover, we introduce hebbian rule for CGRVHNs, and prove that it also decomposes to those for CVHNs and HVHNs. These decompositions are expected to lead to improvement in the learning algorithms and analysis of the inherent properties.

This paper studies higher-order color Voronoi diagrams in a one-dimensional space. A recent paper by Bae, Oliver, and Papadopoulou [1], presented at the 41st International Symposium on Computational Geometry (SoCG 2025), proved tight upper bounds on the complexity of higher-order color Voronoi diagrams in the plane and presented simple iterative algorithms that compute them. In this paper, we prove exact upper bounds on the complexity of its one-dimensional variants, and present efficient algorithms to compute them. Our results are based on new observations on arrangements and levels of a collection of planar curves that satisfy certain conditions.

Most Montgomery multiplication algorithms using residue number system (RNS) representations are constructed using two sets of RNS bases (double bases), which involve representation extensions between the bases. In 2024, a new algorithm was proposed for calculating Montgomery multiplication using a single RNS base set (single base). Assuming that the size of the RNS base set is as k, although the typical number of multiplications increases from order 2k² to 3k², the single-base algorithm reduces the processing cycles from 2k of the double-base to k beause the single-base algorithm has more parallelism. This study first refines the single-base algorithm by changing the expression of variables from standard RNS to the RNS normalized by the complementary range, which we name the ξ-expression in this study. We then discuss how the number of multiplications can be reduced in specific multiplications, that is, squaring and constant multiplication. The reduction rate depends on k and asymptotically approaches two-thirds when k increases. These two multiplications are used in a binary exponentiation algorithm that is widely used to implement RSA cryptosystems. Third, we present an algorithm for lazy reduction using a single RNS base, which can be used to implement elliptic-curve cryptography. The lazy-reduction formula for a single-base RNS is suitable for parallel processing and thus requires fewer processing cycles than the formula for a double-base RNS. We evaluate the proposed algorithms from an algorithmic perspective using a simple hardware model.

The Hamiltonian cycle problem is a well-known NP-complete problem in graph theory. This problem relates to lots of practical problems such as designing very large scale integration (VLSI) and travelling salesman problem (TSP). Since it is NP-complete, there is no efficient algorithm to solve the Hamiltonian cycle problem, and hence, its solution is valuable. In this paper, we propose new physical zero-knowledge proof protocols for the Hamiltonian cycle problem, whereby an entity can prove its knowledge of a solution to another entity without leaking any information about the valuable solution. Our protocols are more efficient than the previous protocols. We also propose a physical zero-knowledge proof protocol for TSP, one of whose building blocks is a new representation of an integer commitment with a secure addition protocol.

Convolutional neural networks (CNNs) have emerged as one of the most successful deep learning approaches to image recognition and classification. A recent line of research, which includes zkCNN (ACM CCS ’21), vCNN (IEEE Trans. Dependable Secur. Comput. 21(4)), and ZEN (Cryptology ePrint Archive 2021/087), aims at protecting the privacy of CNN models by developing publicly verifiable proofs of correct classification which do not leak any information about the underlying CNN models themselves. A shared feature of these schemes is that they require the entity constructing the proof to have access to both the model and the input in the clear. In other words, a client holding a potentially sensitive input is required to reveal this input to the entity holding the CNN model, thereby sacrificing his privacy, to be able to obtain a verifiable proof of correct classification. This is in contrast to the security guarantees provided by secure classification considered in privacy-preserving machine learning, which does not require the client to reveal his input to obtain a (non-verifiable) classification. In this paper, we propose a privacy-preserving verifiable CNN scheme that overcomes this limitation of the previous schemes by allowing the client to obtain a classification proof without having to reveal his input. The obtained proof allows the client to selectively reveal properties of the obtained classification and his input, which will be verifiable to any third-party verifier. Our scheme is based on the recent notion of collaborative zk-SNARKs by Ozdemir and Boneh (USENIX Security ’22). Specifically, we construct a new collaborative zk-SNARK based on Bulletproofs achieving an efficient maliciously secure proof generation protocol. Based on this, we then present an optimized approach to CNN evaluation. Finally, we demonstrate the feasibility of our approach by measuring the performance of our scheme on a CNN for classifying the MNIST dataset. Our measurement and estimation show that our protocol can prove the correctness of a classification of LeNet (Proc. IEEE 86(11)) within 40 minutes.

Card-based cryptography is a research field for realizing secure computation using physical cards. In card-based protocols, the efficiency of a protocol is mainly measured by the number of cards and the number of shuffles. Shinagawa and Nuida proposed a batching technique, which is a technique for reducing the number of pile-scramble shuffles (PSS) by combining multiple PSSs into a single PSS. In this paper, we propose three new methods for the batching technique. The first method uses a new shuffle called a generalized PSS; unlike a PSS, this allows the number of cards in each pile to be uneven. The second method uses a shuffle called a branching PSS, which is a PSS for a sequence including face-up cards. The third method uses a shuffle called a branching generalized PSS, which is a generalized PSS for a sequence including face-up cards. Our methods are efficient in terms of the number of cards and shuffles compared to the existing batching technique. As applications, we use our methods to reduce the number of shuffles for AND protocols and XOR protocols.

We study the Fewest Clues Problem (FCP), a framework that asks whether fixing parts of a certificate forces uniqueness for problems in NP. While several FCP versions of NP-complete problems are known to be Σ₂P-complete, a general method has been lacking. We present a meta theorem: if P is NP-complete and FCP-P is Σ₂P-complete, then any parsimonious reduction (i.e., preserves the number of solutions by a one-to-one correspondence between solutions) from P to Q whose certificates satisfy a simple alignment (each target coordinate is either a constant or an injective image of exactly one source coordinate) implies that FCP-Q is Σ₂P-complete. As a case study, we prove that FCP-Minesweeper Consistency Problem (FCP-MCP) is Σ₂P-complete: first by a direct parsimonious reduction adapted from the classical MCP construction, and second by verifying that the same reduction meets the meta theorem’s conditions. To make the method practical, we add a short “cookbook” checklist and work out two additional instantiations beyond MCP: FCP POSITIVE PLANAR 1-IN-3 SAT and FCP 1-IN-3 SAT⁺. Our theorem thus provides a systematic route to transfer Σ₂P-completeness once a suitable parsimonious, certificate-aligned reduction is available.

Identity-based matchmaking encryption (IB-ME) is an advanced encryption scheme that allows both the sender and the receiver to specify their respective identities. We study the notion of hierarchical IB-ME (HIB-ME), which augments IB-ME with delegation capabilities. Specifically, we first formalize HIB-ME and construct it based on hierarchical identity-based encryption and hierarchical identity-based signature. Moreover, as applications of the HIB-ME, we show two chosen ciphertext secure (H)IB-ME constructions for different security levels.

In this paper, we investigate the computational complexity of a series of puzzles. We are given a set of n centers of circles and n unit disks. Each disk is based on a unit circular shape, but a part of it can be bitten by other disks, which is called a lune. We investigate three variants of this puzzle. First, we investigate the classic packing puzzle of disks and lunes. We note that the centers of disks are given as a part of input, and some disks can be lunes. Therefore, essentially, we can only choose and rotate the lunes to pack them. (In our reduction, the assignment of lunes is easy to find. Therefore, only rotations of them are crucial.) Even under this strong constraint, the packing problem is still NP-complete. Next we turn to the combinatorial reconfiguration variant of this puzzle. That is, we are given two nonoverlapping arrangements of disks and lunes on a given set of centers on a board. Each disk is pinned at the center, and thus we can just rotate it. The problem asks if we can transform one to the other by just rotations of disks and lunes without overlapping. We show that this puzzle is PSPACE-complete in general. Lastly, we focus on the cases in which the puzzle can be solved in polynomial time. The first tractable case is a one-dimensional packing puzzle. The second one is the screw-type variant of this puzzle. In this variant, each disk or lune is realized by a thick screw. We are given a nonoverlapping arrangement of them. The operation we can do is that we can screw a disk or lune if its orbit is not blocked by any other disk or lune. We prove that this variant can be solved in polynomial time.

Deciding flat foldability of a rectangular piece of paper with a given mountain-valley pattern is known to be NP-complete. One special case known to be solvable in linear time is when the creases are parallel to each other and to one side of the rectangular piece of paper; this case reduces to a purely one-dimensional folding problem. A linear-time algorithm for this problem was given by Arkin et al. in 2004. Recently, this case was generalized to a parallelogram of paper: if the piece of paper is a parallelogram and creases are parallel to each other and to one side of the paper, then it can be solved in linear-time by a group of the authors of this paper (Demaine et al. 2020). In this paper, we try to extend the result to a more general shape of paper. A polygonal piece of paper with parallel creases is said to be axis-convex if any straight line parallel or perpendicular to the creases shares a continuous part (i.e., a straight line segment) with the polygon. We show that under the restriction of that all creases are parallel to each other, if the shape of the paper is any axis-convex polygon, the flat-foldability problem with mountain-valley assignment can be solved in linear-time. For obtaining the linear-time algorithm, we redefine the operation SEPARATE, which was presented in the algorithm for the parallelogram cases, under new conditions.

This study addresses a common problem: people often use similar walking routes, resulting in monotony in their daily strolling experience. In this study, we propose a method for generating diverse routes of lengths within a specified range. We define the “similarity” between two routes as the total length of the shared edges, and formulate the problem as one of minimizing the total similarity across all pairs in the generated route set. This means minimizing the average shared edge length between consecutive routes when routes are selected based on random permutations. As the problem is NP-hard, we introduce a heuristic algorithm based on a local search that reconnects routes using a bidirectional breadth-first search. Experimental results using real-world road network data from Tokyo and Kyoto demonstrate that the proposed method effectively generates diverse route sets with well-distributed edge usage patterns. The proposed approach has several practical applications, including daily strolling route recommendation, test scenario generation for autonomous driving systems, and other route scheduling problems requiring diversity.

In the complexity estimation for an attack that reduces a cryptosystem to solving a system of polynomial equations, the degree of regularity and an upper bound of the first fall degree are often used in cryptanalysis. While the degree of regularity can be easily computed using a univariate formal power series under the semi-regularity assumption, determining an upper bound of the first fall degree requires investigating the concrete syzygies of an input system. In this paper, we investigate an upper bound of the first fall degree for a polynomial system over a sufficiently large field. In this case, we prove that the first fall degree of a non-semi-regular system is bounded above by the degree of regularity, and that the first fall degree of a multi-graded polynomial system is bounded above by a certain value determined from a multivariate formal power series. Moreover, we provide a theoretical assumption for computing the first fall degree of a polynomial system over a sufficiently large field.

Symmetric functions are an important class of functions that includes many interesting functions. Constructing efficient card-based protocols for symmetric functions is an important research topic in card-based cryptography. In 2020, Ruangwises and Itoh designed a protocol for symmetric functions f: {0, 1}ⁿ → R using 2n + 2 cards of a two-colored deck. Whether the number of cards in this protocol could be reduced is an open problem. In this paper, we design protocols for symmetric functions using 2n + 1 cards of a two-colored deck, which solve the above problem affirmatively. We also propose protocols for symmetric functions using 2n cards of a three-colored deck, while the existing 2n-card protocol requires a four-colored deck. By further restricting the class of functions to {0}-partially doubly symmetric functions, we can construct even more efficient protocols. In particular, we construct protocols for {0}-partially doubly symmetric functions using 2n cards of a two-colored deck, while the existing protocol requires 2n + 1 cards.

Shi et al. (Scientific Reports 2025 (SJR Q1 Journal)) proposed a public key searchable encryption scheme, which they call secure public key searchable encryption (SPKSE). In this short note, we show that a trapdoor can be generated from a ciphertext without using the receiver’s secret key. We also show that the test algorithm can be run without using the server’s secret key, although the scheme employs a designated tester setting. Our analysis indicates that the scheme must be completely redesigned.

Wang, Wu, Li, Du, and Luo proposed a public key searchable encryption scheme (Electric Power Systems Research 2022 (SJR Q1 Journal)). They claimed that no keyword information is revealed by either a ciphertext or a trapdoor. In this short note, we demonstrate that keyword information is indeed revealed from both a ciphertext and a trapdoor.

Han, Han, Meng, Lai, and Wu (Journal of Systems Architecture 2024 (SJR Q1 Journal)) proposed a public key searchable encryption scheme. They claimed that the scheme provides keyword blindness where a trapdoor generation center generates a trapdoor without knowing the keyword. In this short note, we show that their scheme does not provide keyword blindness.

The longest common subsequence (LCS) problem, consisting of finding an arbitrary LCS of given two strings, has been studied for a long time due to its many applications in fields. As a variant of this problem, we consider the problem of enumerating all LCSs. In LCS enumeration, there is a tradeoff between the delay time to find each LCS and the size of the space used to shorten it. For any positive constant α with α < 1, we propose an O(n^{2 - α})-space algorithm that outputs all LCSs each in O(n^{1 + α} log² n) time after an O(n²)-time preprocessing, where n is the length of the input strings.

This paper mainly focuses on the machine learning integration of battery storage system in renewable energy microgrid. In the architecture of renewable energy microgrid, battery storage system has some challenges, such as weak adaptability to output fluctuation, lagging charging and discharging strategies, and rapid life loss, while the existing regulation methods are deficient in stability and energy efficiency. In view of these situations, an innovative machine learning integrated optimization scheme, namely EMS-ml algorithm, combines multi-model fusion and real-time data-driven technology to optimize the renewable energy consumption and charging and discharging scheduling process to ensure the safe operation of battery storage system. The experimental results show that the scheme effectively improves the response speed of the battery storage system to the fluctuation of wind and solar output, reduces the response delay to 0.2 s, and in the high fluctuation scenario, that is, when the fluctuation of wind and solar output is ±30%, the absorption rate of renewable energy reaches 92.3%, and also improves the cycle life of the battery. The energy efficiency index reaches 8.72. In the actual microgrid prototype, the charging and discharging accuracy index FCD is 356, which provides a strong support for the stable operation of the renewable energy microgrid.

When IR-drop in LSI testing becomes excessive in a certain area of LSI, it causes excessive delay, resulting in test malfunction (over-testing). This is because power consumption in the test mode is generally much higher than in the normal mode. Excessive IR-drop does not occur in the entire area of a circuit, but in certain areas where a large number of switching activities occur (such areas are called hotspots in this work). In order to avoid test malfunction, it is important to develop a method to reduce or control IR-drop in the hotspots. Locating hotspots is a necessary technique to reduce or control IR-drop effectively and efficiently. In this work, we propose a method to locate hotspots caused in the test mode by switching probability calculation and functional analysis with partial truth table. Experimental results for IWLS2005 OpenCores circuits demonstrate the proposed method can support to locate hotspots.

Range arguments are a type of zero-knowledge proofs that aim to prove that a prover’s committed value falls within a specified range for a verifier. Previously, most range arguments were constructed based on the discrete logarithm (DLOG) assumption, and hence, exponentiation operation is required for proof generation and verification. In addition, it is generally known that splitting a zero-knowledge proof protocol into a preprocessing phase and an online phase makes computation after fixing the input efficient. Still, such protocol has yet to be known for range arguments. This paper proposes an efficient range arguments protocol with a preprocessing phase. Our proposal takes a new approach by using arithmetic circuits to express the constraints that the prover must prove. The prover (resp. verifier) can generate (resp. verify) a part of proof based on multiplication and addition operations instead of exponentiation operations. Our range argument is a generic construction that does not rely on any particular mathematical assumptions, which enables us to construct a post-quantum range argument. The implementation evaluation shows that the total computation time for the prover and verifier in the online phase is efficient compared to Bulletproofs, one of the state-of-the-art range proofs. Especially, the prover computation is efficient.

With the rapid development of communication network, users’ traffic demands have shown an unprecedented growth trend. Mobile virtual edge network represents a promising system to meet these requirements. Therefore, this paper studies a virtual edge network. Firstly, Software Defined Network (SDN) and Network Function Virtualization (NFV) techniques are adopted to establish a hyper-fusion virtualization network architecture. In the proposed architecture, mobile virtual network operators (MVNOs) are responsible for virtualizing shared physical layer network infrastructures into virtual network slice resources and allocating these resources to users according to their traffic demands. Secondly, a hyper-fusion orchestration method for the Communication, Caching and Computation (3C) slice resources is presented, and a joint optimization problem including user association, 3C slice resources, and power allocation is constructed. Subsequently, a three-steps resource allocation scheme that includes user association, 3C slice resource allocation method and power allocation optimization algorithm is proposed to solve the formulated problem. Finally, numerical results demonstrate that the proposed scheme significantly improves the delay performance of the network.

In this letter, we address a global adaptive regulation problem for a class of systems subject to unknown input saturation. A notable feature is that the saturation bounds are unknown and time-varying. To handle this unknown input saturation, we propose a novel event-triggered gain-scaling state-feedback controller. It is analytically shown that the resulting closed-loop system achieves global regulation and that Zeno behavior is excluded. An illustrative example with comparative results is provided to demonstrate the effectiveness and validity of the proposed approach.

Efficient resource allocation and task offloading remain critical challenges in intelligent transportation systems. This letter introduces M-MADDPG, a multi-agent deep reinforcement learning algorithm enhanced with multi-head attention, to jointly optimize task offloading, transmission power, and computing frequency in dynamic vehicular environments. The proposed attention mechanism significantly improves feature extraction and agent cooperation. Extensive simulations demonstrate that M-MADDPG consistently outperforms conventional methods in throughput and adaptability, providing a robust and scalable solution for future vehicular networks.

LEO communication mega-constellations with laser Inter-Satellite Link (ISL) provide a promising solution by enabling cross-constellation data relay for remote sensing satellites. This letter first proposes an optimal time-matching algorithm that efficiently establishes temporary Cross-Layer Link (CLL) while satisfying strict antenna pointing constraints. Building upon these links, we design a Dyna-Q-Enhanced Distributed Routing Algorithm (DQDRA), a fully decentralized reinforcement learning(RL) approach that enables each satellite to autonomously make low-delay next-hop decisions without requiring global network state. Simulation results show that, compared with the centralized and baseline distributed method, the proposed scheme significantly reduces end-to-end (E2E) latency and improves the delivery success rate, particularly demonstrating superior performance under high traffic load conditions.

For shipborne high-frequency surface wave radar(HFSWR), the six degrees of freedom (6-DOF) oscillation motion of the platform seriously affect the performance of target detection and parameter estimation. This letter analyzes the antenna displacement caused by the 6-DOF motion, and confirms that the attitude adjustment of the platform could minimize the impact of range bias on the target Doppler spectrum. On this basis, this letter proposes an RD imaging method based on platform steering, and verifies the feasibility and effectiveness of the proposed method through Monte Carlo experiments based on actual RD backgrounds. This helps achieve the detection and tracking of noncooperative targets by shipborne HFSWR.