A data processing functions are expected as a key-issue of knowledge-intensive service functions in the Cloud computing environment. Cloud computing is a technology that evolved from technologies of the field of virtual machine and distributed computing. However, these unique technologies brings unique privacy and security problems concerns for customers and service providers due to involvement of expertise (such as knowledge, experience, idea, etc.) in data to be processed. We propose the cryptographic protocols preserving the privacy of users and confidentiality of the problem solving servers.

In this paper, we propose a knapsack cryptosystem based on three knapsacks. Although one of the three secret knapsacks is superincreasing, the other two are non-superincreasing. On the encryption, a ciphertext is formed by multiplying the two non-superincreasing knapsacks together and then adding it to the superincreasing knapsack. Due to this structure of the cipher text, our knapsack cryptosystem is thought to be secure against all existing attacks, i.e., the low density attack and Shamir attack.

The linear Piece In Hand (PH, for short) matrix methods are general prescriptions which can be applicable to any type of multivariate public key cryptosystems (MPKCs, for short) for the purpose of enhancing their security. Among them, the primitive linear PH matrix method was introduced in our previous work [S. Tsujii, K. Tadaki, and R. Fujita, Cryptology ePrint Archive, Report 2004/366, December 2004] to explain the notion of the PH matrix methods in general in an illustrative manner and not for a practical use to enhance the security of any given MPKC. In 2005, for the purpose of enhancing the security of the primitive linear PH matrix method to a practical level, Akasaki proposed a variant of the primitive linear PH matrix method, called the two-sided linear PH matrix method. In this paper we show that the two-sided linear PH matrix method is reduced to the primitive linear PH matrix method. Based on this, we show that the two-sided linear PH matrix method cannot be more secure than the primitive linear PH matrix method.

The linear Piece In Hand (PH, for short) matrix method with random variables was proposed in our previous work [S. Tsujii, K. Tadaki, and R. Fujita, IEICE Transactions on Fundamentals, E90-A, No.5, pp.992-999, 2007]. It is a general prescription which can be applicable to any type of multivariate public key cryptosystems for the purpose of enhancing their security. Actually, we showed, in an experimental manner, that the linear PH matrix method with random variables can certainly enhance the security of HFE against the Gröbner basis attack, where HFE is one of the major variants of multivariate public key cryptosystems. In the linear PH matrix method with random variables, the three secret matrices: the PH matrix M, S and R play a central role in the key-generation. In our former work [K. Tadaki and S. Tsujii, IEICE Transactions on Fundamentals, E93-A, No.6, pp.1102-1110, 2010], we presented a key-generation algorithm which generates these matrices. In this paper, we make more clear the specification of the linear PH matrix method with random variables. In the linear PH matrix method, a matrix A is introduced to the public key of the enhanced multivariate public key cryptosystem internally in order to prevent an eavesdropper from forging the PH matrix. In this paper, we first investigate a necessary condition for the matrix A to satisfy by considering the immunity against a variant of the attack by Courtois, Daum, and Felke against HFEv-, on which the signature scheme Quartz is based. Based on this investigation, we then present a complete and concise key-generation algorithm for the linear PH matrix method with random variables, which generates the complete secret key (M; S; R; A) of the method.

This paper deals with the problem of noise suppression for white and colored noises. Kalman filter based noise suppression is well known as effective approach, and usually performs the parameter estimation algorithm of AR (auto-regressive) system and then the Kalman filter algorithm. In this paper, we propose Kalman filter for robust noise suppression without the conception of AR system. The algorithm aims to achieve robust noise suppression using only Kalman filter theory from the canonical state space models with (i) a state equation composed of the speech signal and (ii) an observation equation composed of the speech signal and additive noise. We also show the effectiveness of the proposed method, which utilizes Kalman filter theory for the proposed canonical state space model with the colored driving source, using numerical results and subjective evaluation results.

The concept of information security is "the dynamic process for establishing an integrated and complete system designed to form, without infringing freedom broadened by IT (information technology), and with closer coordination among technologies, administration and management techniques, legal and social systems and information morals in order to make compatible improved usability and efficiency and enhanced security, protected privacy and minimized surveillance, or monitoring, over people". This paper considers the paradigm of information security as an interdisciplinary comprehensive science and the way to develop human resources, both of which are required to establish information security.

In this paper, we discuss a blind channel equalization (BCE) considered where the output data is corrupted by additive noise. As one of BCE methods, there is a method of permuting the parameter estimation problem of finite impulse response (FIR) channel model to that of infinite impulse response (IIR) system model. The feature of this method is the ability to use several existing adaptive algorithm. However, in the actual system, it is considered that the output data includes an additive noise. Under the influence of noise, asymptotic bias arises in the presumed value of frequency characteristics of channel. Then, we propose the modified BCE method using a recursive bias compensated least-squares (RBCLS) algorithm. The performances of the proposed algorithm are shown by computer simulation.

Subspace methods with second-order statistics based on principal component analysis (PCA) basically need to calculate the eigenvalues and the eigenvectors of the autocorrelation matrix of the received signal. However, the calculation of the eigenvalues and the eigenvectors of the matrix requires much computational complexity. We propose a new algorithm based on PCA without solving the eigenvalues and the eigenvectors of the matrix. Moreover, we perform the proposed method under the condition that noise-variance is known, but we confirm that the proposed method is effective to a certain degree when noise-variance is unknown. We show the effectiveness of the proposed method by numerical examples.

The normalized LMS algorithm is widely used because of its simplicity. However, this algorithm behaves unstably when all elements of an input vector is very small. This property is caused by the division in the procedure of the normalized LMS algorithm. One of the modifications on this problem is taken that coefficients of the adaptive filter are not updated when a norm of an input vector is smaller than a threshold. This paper shows effects of interrupting adaptive filter's coefficients update by the threshold derived from the error analysis. Also, it presents the guarantee value and the stochastic fastest convergence step gain.

The output error method (OEM) IIR adaptive filters (ADF) can not guarantee their stability and global convergence. We present a sufficient condition for the stability of time-varying IIR filters, and prove that the IIR ADF restricted by the condition can approximate arbitrary unknown systems, provided that the degree of ADF is high enough. Then, we propose a new adaptive method, the "Rolling-Stone scheme" and apply it to the constrained OEM IIR ADF. It is proved that the new OEM IIR ADF is guaranteed to be stable and globally convergent to a global minima.>

This paper presents rigorous conditions, without the independence assumption on the input signal, for the first and second moment convergence of the gradient adaptive algorithms including FIR and IIR (equation error method) LMS and IV algorithms.>

Discrete linear time-varying (TV) systems have been playing an important role in many real-time signal processing and control systems. The stability criteria of these TV systems require information of the impulse response sequence of the state transition matrices in an infinite time interval which is often difficult to obtain in practice. This paper reports some results on this problem. We first presents sufficient conditions for exponential, l/sup p/ and l/sup p/-BIBO stabilities. A detailed analysis is then given on this condition and the distribution domain of the instantaneous poles meeting the condition.>

A novel IIR adaptive gradient instrumental variable echo canceler (GIVE) is presented. Its features include adaptive controllability during double-talk periods in acoustic conference systems; guarantee of global convergence; low computational cost (the same order as the IIR LMS algorithm of the equation error method); and flexible structures (parallel or series-parallel structures). We also show a convergence analysis for gradient adaptive algorithms including GIVE. Based on this analysis, the optimum stepsize for GIVE and three suboptimum algorithms are proposed to accelerate convergence and reduce misadjustment. In addition, a simple method that guarantees the stability of IIR filters and a configuration of GIVE applicable to closed loop systems are presented. These proposals are extensively studied by computer simulations.>

A noninteractive ID-based key sharing scheme based on a method of iterative elimination of additive random numbers is proposed. With this method, a common key is generated step by step, leaving no opportunity for conspiracy attacks. The integrity of the scheme is discussed, and a method for improving computational efficiency is proposed.>

A finite wordlength floating-point error analysis for the RLS (recursive least squares) algorithm using UD factorization is presented. It is known that this algorithm is suitable for parallel architectures; however, it is not known how the wordlength of the algorithm should be chosen for hardware implementation. An investigation is conducted of the relation between the word length and the convergence characteristics of the algorithm, and a sufficient condition for the wordlength of the algorithm so that the proposed algorithm works when noises exist is given. Computer simulation indicates that the algorithm with 5-6 bit operation has almost the same convergence characteristics as the usual RLS algorithm using 64 bit operation.>

The authors propose a block adaptive algorithm for estimating the impulse response of discrete-time systems and discuss some of its properties. This algorithm, in which the performance measure is the block mean square error (BMSE), consists of block processing and orthogonal projection arithmetic and updates the adaptive filter's coefficients once for each block of data. That is, the algorithm is based on orthogonal projection to the subspace spanned by input vectors corresponding to the block index. Some fundamental theorems are derived with respect to this algorithm, and they are substantiated by computer simulations.>

In a modern network system, data security technologies such as cryptosystems, signature schemes, etc., are indispensable for reliable data transmission. In particular, for a large-scale network, ID-based systems such as the ID-based cryptosystem, the ID-based signature scheme, or the ID-based key distribution system are among the better countermeasures for establishing efficient and secure data transmission systems. The concept of an ID-based cryptosystem has been proposed by A. S?hamir (1985), and it is advantageous to public-key cryptosystems because a large public-key file is not required for such a system. An ID-based cryptosystem based on the discrete logarithm problem is proposed which is one of the earliest realizations in Shamir's sense. The security against a conspiracy of some entities in the proposed system is considered, along with the possibility of establishing a more secure system.>

A three-port echo canceler (EC) configuration is proposed which observes the signal of the near-end side on a two-wire circuit in addition to the four-wire circuit signals. Embedding these signals on hybrid ports into a three-dimensional autoregressive process, echo path and innovations of near- and far-end speeches can be estimated through a three-channel lattice filter. The new configuration is then able to track echo path time variance, even during double talk (DT), and requires no changeover at either the beginning or end of DT, thus eliminating the need for DT detection. Two echo synthesizers utilizing inverse lattice and the echo path estimate possess guaranteed stability without the need for testing.>

IIR (infinite-impulse-response) echo cancellers (EC) and howling cancellers (HC) have to freeze adaptation during 'double talk' (DT). However, a fixed series-parallel structure induces signal distortion. Current methods, using a fixed parallel structure or equalizer, need a stability test which costs O(N/sup 2/) multiplications per sample period. A configuration of IIR EC and HC, consisting of a maximum-entropy two-channel lattice filter and its inverse filter, is proposed. It can cancel echoes and during DT eliminate the signal distortion by the fixed inverse lattice. Its stability is automatically guaranteed without the necessity of tests. The computational cost is O(N).>

IIR type adaptive algorithms are classified into two categories according to their structures: series-parallel type and parallel type. Though parallel type algorithms show better convergence performance than series-parallel type, they present strong tendency to oscillate. Therefore, only Landau's method which is based on hyper-stability theory has been known as the parallel type. However, this method requires a lot of computation because it involves matrix operations. This paper proposes a new parallel type algorithm which utilizes the fast least squares method. Its computational complexity is much less than that of Landau's method. We also present by both analysis and computer simulation that the convergence performance of our method is better than that of the series-parallel type fast least squares method.