Search

Current prediction models for network traffic cannot accurately depict the multi-properties of the Internet traffic. This paper proposes a wavelet-based hybrid model prediction method for network traffic called CLWT model and proposes a prediction method for traffic based on this model. The traffic time series can be rapidly decomposed respectively into approximate time series and detail time series with LF and HF response. The approximate time series predicts by making use of Least Squares Support Vector Machine and proceeds error calibration by using Generalized Recurrent Nerve Network. The detail time series predict it by making use of self-adaption chaotic prediction methods after the medium-soft threshold noise reduction. Finally the prediction value of time series is got by making use of promoting wavelet reconstitution. The effectiveness for the prediction methods mentioned in the paper has been validated by simulation experiment. High prediction accuracy is obtained compared with the existing methods.

This article selects HOL theorem proving systems for hardware Trojan detection and gives the symbol and meaning of theorem proving systems, and then introduces the symbol table, item and the meaning of HOL theorem proving systems. In order to solve the theorem proving the application of the system in hardware Trojan detection requirements, this article analyses basic hardware Trojan detection methods which applies for theorem proving systems and introduces the implementation methods and process of theorem proving about hardware Trojan detection.

For hardware Trojan detection process is difficult to effectively extract the exact equivalent logic function of the target chip logic function is proposed based on contradictory truth table equivalent extraction scheme and key Enabling technologies. Designed to target chip waveforms attribute recognition model, through the analysis of multiple properties of the model and proof, the work is given waveform attribute recognition model parameter setting principles.

A lattice basis reduction multisequence synthesis (LBRMS) algorithm was presented with a new, intuitive and vector-form model. In this paper, a refined version of the LBRMS algorithm is deduced, from which Massey's conjectured algorithm can be derived. Moreover, if we modify the lattice basis reduction algorithm used in the LBRMS algorithm, a new synthesis algorithm is deduced, which is equivalent to the generalization of the Berlekamp-Massey algorithm proposed by Feng and Tzeng (1989). Therefore, the LBRMS algorithm provides a unified approach to the multisequence shift-register synthesis problem.

To solve the problem that it is difficult to detect the failure-type hardware Trojan, this paper selects the failure-type hardware Trojan as the research subjects, putting forward the detection principles of failure-type hardware Trojan based on state transition diagram. It also designs and realizes the detection algorithm as well as provides the relative demonstrations of its correctness. The experiment testing shows that the proposed detection algorithm can effectively check out the hardware Trojan in the combinational logic and timing logic hardware chips, with a detection accuracy of 100% ideally, actual success rate of 92.8% and false retrieval rate of 7.14%.

In this paper we present a new framework for automatically proving the security of public-key cryptographic schemes in computational model. The framework uses the sequence-of-games approach to construct security proof. A probabilistic polynomial-time process calculus is designed to describe the attack games and the game transformations are executed with the help of observational equivalence. The framework has been implemented as a automated prover tested on a series of examples, including encryption and signature schemes. As an example, we illustrate the use of our framework with the proof of IND-CPA security of ElGamal encryption.

In the field of malware study, the automatic generation of malicious behavior pattern based on system call trace is important to malware detection. This paper studied the generation of malicious behavior distinguishing pattern, and used attributes order reduction method to reduce the patterns. It gave the ordering rules based on mutual information and the size of pattern. After studying the quick reduction algorithm based on divide and conquer method, it improved the algorithm according to the characteristic of datasets. Finally, test results verifies that the improved algorithm can decrease the reduction time, the ordering method can obtain relatively good reduction results, and the classification result is relatively good.

The automatic generation of malicious behavior pattern based on system call trace is important to malware detection. This paper studied the existing generation method of malicious behavior specification. In order to reduce the complexity of pattern generation, it constructs graph which vertex label is unique, and uses these graphs to mine the pattern. To address the issue of limitation of the minimal contrast subgraph mining method, it uses multiple positive and negative samples, and proposes a mining method to mine distinguishing patterns based on mutual information. It designs the overall framework of mining process, and gives the mining algorithm. Finally, validation results demonstrate the effectiveness.

RSA is one of the most popular and widely used public key cryptosystems. In this paper, an application of low private exponent attack on it is presented. More precisely, let (N, e) be an RSA public key with corresponding private key d, then N can be factored provided that there exists a proper integer k such that ek is relatively small and dk mod φ(N) is small (or large) enough. Experiment shows that our generalization of low private exponent attack can seek out some new weak keys of RSA.

Formal methods and automatic tools are always necessary for analyzing security protocols, and model checking has become a hot spot of research for its ability of verifying finite-state concurrent system automatically. In this paper, we propose a model of security protocols based on MSR (multi-set rewriting), coupled with PLTL describing security properties. Based on the model, a series of key automatic analysis algorithms close to the implementation are designed for our own tool. By using the tool, we illustrate the method’s effectiveness with some typical protocols and give the experimental results.

Aiming at the technique crux of system failure testing and function reconstruction, we carry on a research to the distilling method of the Boolean relation data. First, start with the analysis to the working wave character of target chip in the waveform confined space of system, and then on this foundation, the corresponding logic relation expression between signals of different logic types is given. While aiming at the combination logic waveform, and also integrating with the analysis to the fixed time relation of the waveform and the feature of its waveform data, we have put forward and carried out the logic relation distilling algorithm of the combination circuit signals based on credibility interval. While aiming at the timing logic waveform and on the foundation of the analysis to the different timing logic types, as well as the analysis to the waveform timing feature, we have put forward and carried out the logic relation distilling algorithm of the timing circuit signals based on waveform decomposition. Through the extraction of corresponding logic relation between different signals, the implicit logic relation of the circuit inner is translated into the limited set value of the logic relation expression which the logic synthesis can identify. And then from the angle of corresponding relation of the different signal logic values, the logic function of circuit is described.

TA4SP is a state-of-art tool of AVISPA that can automatically verify security protocol with unbounded number of parallel sessions. But it still has some limitations and can’t verify hierarchy of authentication automatically. In this paper, we use an approximation-based model to define security protocol and design an algorithm close to the real implementation to calculate the fix-point tree automata based on the tool TA4SP. A method is proposed for analyzing hierarchy of authentication properties as extension of TA4SP. We illustrate the effectiveness of this model with the example of NSPK protocol.

Fair exchange protocols have been studied intensively in recent years. But a lot of methods are still performed manually. In this paper an automatic method is proposed for analyzing fair exchange protocols. In this method we formalize security properties of fairness and non-repudiation in TLA (temporal logic of action) and define common predicates to make the analysis automatic. An "end" rule is introduced to solve the different type of channels. Protocol of fair Zhou-Gollmann has been analyzed in our own implemented tool and an attack trace can be found.

Adaptive steganography which avoids changing the sensitive area in carrier, can efectively improve the security of steganography. Firstly, the schemes of side information synchronization in adaptive steganography are summarized and analyzed. Secondly, using the freedom degree which is extented by the previous one-way texture functions, a new orderpreserving texture function as side information of adaptive steganography is constructed. Finally, by combining wet paper codes and the method of multi-layered embedding, ± 1 algorithm and ± 2 algorithm are proposed. The proposed scheme can efciently enhance the security of adaptive steganography while maintaining fast side information synchronization between communicators. Theoretical analysis and experimental results show the efciency of the proposed scheme.