A non-audible murmur (NAM), a very weak whisper sound produced without vocal fold vibration, has been researched in the development of a silent-speech communication tool for functional speech disorders as well as human-to-human/machine interfaces with inaudible voice input. The NAM can be detected using a specially designed microphone, called a NAM microphone, attached to the neck. However, the detected NAM signal has a low signal-to-noise ratio and severely suppressed high-frequency component. To improve NAM clarity, the mechanism of a NAM production must be clarified. In this work, an air flow through a glottis in the vocal tract was numerically simulated using computational fluid dynamics and vocal tract shape models that are obtained by a magnetic resonance imaging (MRI) scan for whispered voice production with various strengths, i.e. strong, weak, and very weak. For a very weak whispering during the MRI scan, subjects were trained, just before the scanning, to produce the very weak whispered voice, or the NAM. The numerical results show that a weak vorticity flow occurs in the supraglottal region even during a very weak whisper production; such vorticity flow provide aeroacoustic sources for a very weak whispering, i.e. NAM, as in an ordinary whispering.

View full abstract

This work extends the optimum Neymann-Pearson methodology to detection of a subspace signal in the correlated additive Gaussian noise when the noise power may be different under the null (H₀) and alternative (H₁) hypotheses. Moreover, it is assumed that the noise covariance structure and power under the null hypothesis are known but under the alternative hypothesis the noise power can be unknown. This situation occurs when the presence of a small point (subpixel) target decreases the noise power. The conventional matched subspace detector (MSD) neglects this phenomenon and causes a consistent loss in the detection performance. We derive the generalized likelihood ratio test (GLRT) for such a detection problem comparing it against the conventional MSD. The designed detector is theoretically justified and numerically evaluated. Both the theoretical and computer simulation results have shown that the proposed detector outperforms the conventional MSD. As to the detection performance, it has been shown that the detectivity of the proposed detector depends on the additional adaptive corrective term in the threshold. This corrective term decreases the value of presumed threshold automatically and, therefore, increases the probability of detection. The influence of this corrective term on the detector performance has been evaluated for an example scenario.

View full abstract

In this paper, a novel algorithm, Cross Low-dimension Pursuit, based on a new structured sparse matrix, Permuted Block Diagonal (PBD) matrix, is proposed in order to recover sparse signals from incomplete linear measurements. The main idea of the proposed method is using the PBD matrix to convert a high-dimension sparse recovery problem into two (or more) groups of highly low-dimension problems and crossly recover the entries of the original signal from them in an iterative way. By sampling a sufficiently sparse signal with a PBD matrix, the proposed algorithm can recover it efficiently. It has the following advantages over conventional algorithms: (1) low complexity, i.e., the algorithm has linear complexity, which is much lower than that of existing algorithms including greedy algorithms such as Orthogonal Matching Pursuit and (2) high recovery ability, i.e., the proposed algorithm can recover much less sparse signals than even l¹-norm minimization algorithms. Moreover, we demonstrate both theoretically and empirically that the proposed algorithm can reliably recover a sparse signal from highly incomplete measurements.

View full abstract

The dynamic encoding algorithm for searches (DEAS) is a recently developed algorithm that comprises a series of global optimization methods based on variable-length binary strings that represent real variables. It has been successfully applied to various optimization problems, exhibiting outstanding search efficiency and accuracy. Because DEAS manages binary strings or matrices, the decoding rules applied to the binary strings and the algorithm's structure determine the aspects of local search. The decoding rules used thus far in DEAS have some drawbacks in terms of efficiency and mathematical analysis. This paper proposes a new decoding rule and applies it to univariate DEAS (uDEAS), validating its performance against several benchmark functions. The overall optimization results of the modified uDEAS indicate that it outperforms other metaheuristic methods and obviously improves upon older versions of DEAS series.

View full abstract

Because of its simple structure, many reports on the logistic map have been presented. To implement this map on computers, finite precision is usually used, and therefore rounding is required. There are five major methods to implement rounding, but, to date, no study of rounding applied to the logistic map has been reported. In the present paper, we present experimental results showing that the properties of sequences generated by the logistic map are heavily dependent on the rounding method used and give a theoretical analysis of each method. Then, we describe why using the map with a floor function for rounding generates long aperiodic subsequences.

View full abstract

The advantages of the orthogonal frequency division multiplexing (OFDM) are high spectral efficiency, resiliency to RF interference, lower multi-path distortion and others. To further utilize the vast channel capacity of the multiuser OFDM, one has to find the efficient adaptive subcarrier and bit allocation among users. In this paper, we propose a 0-1 integer programming model formulating the optimal subcarrier and bit allocation problem of the multiuser OFDM. We proved that the continuous relaxation of our formulation is tighter than the previous convex optimization formulation based on perspective function and the Lagrangian dual bound of our formulation is equivalent to the linear programming relaxation bound. The proposed Lagrangian dual is seperable with respect to subcarriers and allows an efficient dual maximization algorithm. We compared the performance of the integer programming formulation and the Lagrangian dual of our formulation and the continuous relaxation and the primal heuristic proposed in [3]. Computer simulation on a system employing M-ary quadrature amplitude modulation (MQAM) assuming a frequency-selective channel consisting of three independent Rayleigh multipaths is carried out with the optimal subcarrier and bit allocation solution generated by the 0-1 integer programming model.

View full abstract

The marking construction problem (MCP) of Petri nets is defined as follows: “Given a Petri net N, an initial marking M_i and a target marking M_t, construct a marking that is closest to M_t among those which can be reached from M_i by firing transitions.” MCP includes the well-known marking reachability problem of Petri nets. MCP is known to be NP-hard, and we propose two schemas of heuristic algorithms: (i) not using any algorithm for the maximum legal firing sequence problem (MAXLFS) or (ii) using an algorithm for MAXLFS. Moreover, this paper proposes four pseudo-polynomial time algorithms: MCG and MCA for (i), and MCHF_k and MC_feideq_a for (ii), where MCA (MC_feideq_a, respectively) is an improved version of MCG (MCHFk). Their performance is evaluated through results of computing experiment.

View full abstract

In this paper, a new method for 2-D frequency estimation of multiple damped sinusoids in additive white Gaussian noise is proposed. The key idea is to combine the subspace-based technique and projection separation approach. The frequency parameters in the first dimension are estimated by the MUSIC-based method, and then a set of projection separation matrices are constructed by the estimated frequency parameters. In doing so, the frequency parameters in the second dimension can be separated by the constructed projection separation matrix. Finally, each frequency parameter in the second dimension is estimated by multiple 1-D MUSIC-based methods. The estimated frequency parameters in two dimensions are automatically paired. Computer simulations are included to compare the proposed algorithm with several existing methods.

View full abstract

In this paper, we propose an Estimation of Signal Parameter via Rotational Invariance Techniques (ESPRIT) based algorithm for estimating the two-dimensional-direction-of-arrivals (2D-DOA) of signals impinging on a uniform rectangular array (URA). The basic idea of the proposed algorithm is to successively apply two rounds of one-dimensional ESPRIT (1D-ESPRIT) algorithm for 2D-DOA estimation. The first round 1D-ESPRIT is applied on columns of the URA whereas the other round 1D-ESPRIT is on the rows of the URA. In between, a grouping technique is developed to produces signal groups each containing signals with distinguishable spatial signatures. The grouping technique is performed by using the subspace projection method where the needed spatial information is provided by the first round 1D-ESPRIT algorithm. Computer simulations show that, in addition to having significantly reduced computational complexity, the proposed algorithm possesses better estimation accuracy as compared to the conventional 2D-ESPRIT algorithm.

View full abstract

Location of a source is of considerable interest in wireless sensor networks, and it can be estimated from passive measurements of the arrival times. A novel algorithm for source location by utilizing the time of arrival (TOA) measurements of a signal received at spatially separated sensors is proposed. The algorithm is based on total least-squares (TLS) method, which is a generalized least-squares method to solve an overdetermined set of equations whose coefficients are noisy, and gives an explicit solution. Comparisons of performance with standard least-squares method are made, and Monte Carlo simulations are performed. Simulation results indicate that the proposed TLS algorithm gives better results than LS algorithm.

View full abstract

This letter studies learning of the binary neural network and its relation to the logical synthesis. The network has the signum activation function and can approximate a desired Boolean function if parameters are selected suitably. In a parameter subspace the network is equivalent to the disjoint canonical form of the Boolean functions. Outside of the subspace, the network can have simpler structure than the canonical form where the simplicity is measured by the number of hidden neurons. In order to realize effective parameter setting, we present a learning algorithm based on the genetic algorithm. The algorithm uses the teacher signals as the initial kernel and tolerates a level of learning error. Performing basic numerical experiments, the algorithm efficiency is confirmed.

View full abstract

This paper presents pulse-coupled piecewise constant spiking oscillators (PWCSOs) consisting of two PWCSOs and a coupling method is master-slave coupling. The slave PWCSO exhibits chaos because of chaotic response of the master one. However, if the parameter varies, the slave PWCSO can exhibit the phenomena as a periodicity in the phase plane. We focus on such phenomena and corresponding bifurcation. Using the 2-D return map, we clarify its mechanism.

View full abstract

In this study, we propose a multi-layer perceptron with a glial network which is inspired from the features of glias in the brain. All glias in the proposed network generate independent oscillations, and the oscillations propagate through the glial network with attenuation. We apply the proposed network to the two-spiral problem. Computer simulations show that the proposed network gains a better performance than the conventional multi-layer perceptron.

View full abstract

For a prime p with p ≡ 3(mod 4) and an odd number m, the Bentness of the p-ary binomial function $f_{a,b}(x)={\\ m Tr}_{1}^n(ax^{p^m-1})+{\\ m Tr}_{1}^2(bx^{\\frac{p^n-1}{4}})$ is characterized, where n=2m, $a\\in \\bF_{p^n}^*$, and $b\\in \\bF_{p^2}^*$. The necessary and sufficient conditions of ƒ_a,b(x) being Bent are established respectively by an exponential sum and two sequences related to a and b. For the special case of p=3, we further characterize the Bentness of the ternary function ƒ_a,b(x) by the Hamming weight of a sequence.

View full abstract

This paper presents a high-speed SHA-1 implementation. Unlike the conventional unfolding transformation, the proposed unfolding transformation technique makes the combined hash operation blocks to have almost the same delay overhead regardless of the unfolding factor. It can achieve high throughput of SHA-1 implementation by avoiding the performance degradation caused by the first hash computation. We demonstrate the proposed SHA-1 architecture on a FPGA chip. From the experimental results, the SHA-1 architecture with unfolding factor 5 shows 1.17Gbps. The proposed SHA-1 architecture can achieve about 31% performance improvements compared to its counterparts. Thus, the proposed SHA-1 can be applicable for the security of the high-speed but compact mobile appliances.

View full abstract

In this note, we go further on the “basis exchange” idea presented in [2] by using Mobious inversion. We show that the matrix S₁(ƒ)S₀(ƒ)^-1 has a nice form when ƒ is chosen to be the majority function, where S₁(ƒ) is the matrix with row vectors υ_k(α) for all α ∈ 1_ƒ and S₀(ƒ)=S₁(ƒ ⊕ 1). And an exact counting for Boolean functions with maximum algebraic immunity by exchanging one point in on-set with one point in off-set of the majority function is given. Furthermore, we present a necessary condition according to weight distribution for Boolean functions to achieve algebraic immunity not less than a given number.

View full abstract

In this correspondence, a new method to extend the number of quaternary low correlation zone (LCZ) sequence sets is presented. Based on the inverse Gray mapping and a binary sequence with ideal two-level auto-correlation function, numbers of quaternary LCZ sequence sets can be generated by choosing different parameters. There is at most one sequence cyclically equivalent in different LCZ sequence sets. The parameters of LCZ sequence sets are flexible.

View full abstract

In this letter, a new class of almost binary sequence pairs with a single zero element and three autocorrelation values is presented. The new almost binary sequence pairs are based on cyclic difference sets and difference set pairs. By applying the method to the binary sequence pairs, new binary sequence pairs with three-level autocorrelation are constructed. It is shown that new sequence pairs from our constructions are balanced or almost balanced and have optimal three-level autocorrelation when the characteristic sequences or sequence pairs of difference sets or difference set pairs are balanced or almost balanced and have optimal autocorrelations.

View full abstract

In this paper, we propose a simple, yet effective, multiuser detection scheme for a two-hop cooperative CDMAs. In phase 1, the minimum mean square error (MMSE) detector at the destination is used to identify reliable decisions of direct transmissions from the sources and return them to the relays. Then, in phase 2, based on the reliable decisions, the relays and the destination successively utilize the maximum likelihood (ML) detectors to estimate the residual symbols. Due to the destination estimating the symbols separately from direct transmissions and the relaying signals, as a result the destination does not need the information about the relays' decision performance for the construction of the ML detector. Hence, the proposed scheme is more feasible than existing approaches for practical implementation. In addition, due to the ML detectors in phase 2 only estimating the residual symbols, the number of computations performed by the ML detectors can be reduced significantly. The results of simulations and complexity analysis demonstrate the efficiency and effectiveness of the proposed scheme.

View full abstract