An Algorithm for Fast Generation of Bivariate Poisson Random Vectors

Copulas are used in finance and insurance for modeling stochastic dependency. They comprehend the entire dependence structure, not only the correlations. Here they are estimated from measured samples of random vectors. The copula and the marginal distributions of the vector elements define a multivariate distribution of the sample which can be used to generate random vectors with this distribution. This can be applied as well to time series. A programmed algorithm is proposed. It is fast and allows for random vectors with high dimension, for example 100.

Parallel VLSI generation of random analog vectors with controlled statistics from deterministic chaos is the key to applications such as analog encryption and secure communications, analog built-in self-test, stochastic neural networks, and simulated annealing optimization as well as perturbation learning in neural hardware. We present a class of analog additive cellular automata which generate parallel streams of statistically independent, uniformly distributed random analog values. The underlying noise-shaping mechanism is essentially that of a MASH cascade of delta-sigma modulators. We present theory on the noise-shaping properties, scalable parallel VLSI architectures, and include experimental results from an analog VLSI prototype with 65 channels. The cell for each channel implements a switched-capacitor delta-sigma modulator, and measures 100 /spl mu/m/spl times/120 /spl mu/m in 2 /spl mu/m CMOS technology. The 65 cells are connected in a chain on a 2-D grid, and can be rearranged for use in various VLSI applications that require a parallel supply of random analog vectors.

Facial Parts Detection (FPD) approach in conjunction with Vector Quantization (VQ) algorithm are proposed in this paper for face recognition. There are three phases in the proposed system, namely, Preprocessing, Feature Extraction, and Classification. Detecting facial parts, which are nose, both eyes, and mouth, and choosing appropriate dimensions for each part are done in the preprocessing phase. In the feature extraction phase, four groups for each person, one group for each detected part, are constructed for dimensionality reduction and feature discrimination by considering all parts of all training poses. For further data compaction, VQ algorithm employing Kekre Fast Codebook Generation (KFCG) approach for codebook initialization is applied to each of the four groups. Finally, Euclidean distance criterion is used to obtain the recognition rates. Four databases, namely, ORL, YALE, FERET, and FEI that have different facial variations, such as illuminations, rotations, makeups, facial expressions, etc. are used to evaluate the proposed system. Experimental work is performed to evaluate the performance of the proposed technique and the state-of-the-arts approaches. Then, K-Fold Cross Validation (CV) is used to analyze the results. The proposed system consistently improved the recognition rates as well as the storage requirements. Sample results are given.

Pattern matching algorithms have gained a lot of importance in computer science, primarily because they are used in various domains such as computational biology, video retrieval, intrusion detection systems, and fraud detection. Finding one or more patterns in a given text is known as pattern matching. Two important things that are used to judge how well exact pattern matching algorithms work are the total number of attempts and the character comparisons that are made during the matching process. The primary focus of our proposed method is reducing the size of both components wherever possible. Despite sprinting, hash-based pattern matching algorithms may have hash collisions. The Efficient Hashing Method (EHM) algorithm is improved in this research. Despite the EHM algorithm’s effectiveness, it takes a lot of time in the preprocessing phase, and some hash collisions are generated. A novel hashing method has been proposed, which has reduced the preprocessing time and hash collision of the EHM algorithm. We devised the Hashing Approach for Pattern Matching (HAPM) algorithm by taking the best parts of the EHM and Quick Search (QS) algorithms and adding a way to avoid hash collisions. The preprocessing step of this algorithm combines the bad character table from the QS algorithm, the hashing strategy from the EHM algorithm, and the collision-reducing mechanism. To analyze the performance of our HAPM algorithm, we have used three types of datasets: E. coli, DNA sequences, and protein sequences. We looked at six algorithms discussed in the literature and compared our proposed method. The Hash-q with Unique FNG (HqUF) algorithm was only compared with E. coli and DNA datasets because it creates unique bits for DNA sequences. Our proposed HAPM algorithm also overcomes the problems of the HqUF algorithm. The new method beats older ones regarding average runtime, number of attempts, and character comparisons for long and short text patterns, though it did worse on some short patterns.

Mixtures of normal distributions provide a useful modeling extension of the normal distribution—both univariate and multivariate. Unlike the normal distribution, mixtures of normals can capture the kurtosis (fat tails) and nonzero skewness often necessary for accurately modeling a variety of real-world variables. An efficient analytical Monte Carlo method is proposed for considering multivariate mixtures of normal distributions having arbitrary covariance matrices. The method consists of a linear transformation of a multivariate normal having a computed covariance matrix into the desired multivariate mixture of normal distributions. The computed covariance matrix is derived analytically. Among the properties of the multivariate mixture of normals that we demonstrate is that any linear combination of mixtures of normal distributions is also a mixture of normal distributions. Methods of fitting mixtures of normal distributions are briefly discussed. A motivating example carried throughout this paper is the use of multivariate mixtures of normals for modeling daily changes in market variables.

Real-Time Detection of Cracks on Concrete Bridge Decks Using Deep Learning in the Frequency Domain

The Node Weighted Steiner Tree Problem (NW-STP) is a generalization of the Steiner Tree Problem. A lagrangean heuristic presented in EngevallS: StrLBN: 98, and based on the work in Lucena: 92, solves the problem by relaxing an exponential family of generalized subtour elimination constraints and taking into account only the violated ones as the computation proceeds. In EngevallS: StrLBN: 98 the computational results refer to complete graphs up to one hundred vertices. In this paper, we present a branch-and-bound algorithm based on this formulation. Its performance on the instances from the literature confirms the effectiveness of the approach. The experimentation on a newly generated set of benchmark problems, more similar to the real-world applications, shows that the approach is still valid, provided that suitable refinements on the bounding procedures and a preprocessing phase are introduced. The algorithm solves to optimality all of the considered instances up to one thousand vertices, with the exception of 11 hard instances, derived from the literature of a similar problem, the Prize Collecting Steiner Tree Problem.

The paper discusses a general method of computer generation of random vectors based on transformations of uniformly distributed vectors. This method is then applied to build up efficient algorithms for generatingq-exponential random vectors and multivariate normal ort-distributed vectors. Comparisons and connections with other similar algorithms are also presented.

Recurrent neural networks-guided vector-valued synchronized key exchange for secure and privacy-preserving communication in Industrial Internet of Things

On generation of elliptical distributions with Gaussian form covariance matrix

In this paper we present a method to incorporate a content-based retrieval algorithm on compressed images with a digital image transform scheme to achieve a low cost and fast indexing method. The target application is the access and interaction with huge amount of visual data on Internet. The approach exploits a modified Wavelet multiresolution decomposition and reconstruction scheme and a multiresolution algorithm for feature extraction and index generation. The efficacy of the method has been proved by extensive tests on YUV compressed JPEG images and the performance have been compared with other approaches on uncompressed, original images, even with the addition of noise. The results suggest a great opportunity to embed in a unique paradigm a fast retrieval technique and a good compression algorithm of low computational complexity, very suitable for Internet imaging applications.

In rapid prototyping, the support can constraint the parts deformation, and avoid a lot of problems such as collapse, shift and imbalance of the parts. The support automatic generation algorithm has become the key research of rapid prototyping technics software; however, presently the support automatic generation algorithm is complex with low efficiency. Thus, it is of great importance in improving the overall efficiency of rapid prototyping technics software through enhancing the support generation efficiency. Based on this, this paper proposes a discrete-marking support algorithm for treatment of processing on manufactured part model of all Triangle-based discrete-marking on the support plane, which realizes the three-dimensional computation of intersection for each line associated with the triangle automatically, and avoid invalid computation between all the support lines and other triangles. Thus, the algorithm efficiency has reached O(n). Meanwhile, the technics support generation speed has been improved. Furthermore, the new mesh discrete-marking and automatic support generation algorithm has been successfully applied to the Sterolithography apparatus (SLA) and Selective Laser Melting (SLM) of HRPS Series. Practical application indicates that the new support generation algorithm improve considerably the support technics efficiency and rapid prototyping efficiency of the technics software.

As smart mobile devices are widely used, mobile threats are more serious, so security in mobile becomes more and more important. However, the performance of these devices are not powerful enough to use the same security algorithms as PCs. Public key cryptosystems such as RSA need big primes to enhance the security, however, generating a big prime takes substantial time even on a PC. In this paper, we study two prime generation algorithms for smart mobile devices. First, we analyze a previous prime generation algorithm using a GCD test, named PGCD-MR, and show it sometimes performs inferior to the traditional TD-MR test. Second, we propose a new GCD test, named m-bit GCD-MR, for fast prime generation in both PCs and smart mobile devices. We compare the running times of PGCD-MR, m-bit GCD-MR, and TD-MR combinations on PCs and Samsung Galaxy Tab 10.1. The experimental results show our running time analysis is accurate (only 2% error) and m-bit GCD-MR test is the fastest among three prime generation algorithms. More exactly, m-bit GCD-MR test is about 20% faster than the TD-MR combination.

ML models are increasingly being used to increase the test coverage and decrease the overall testing time. This field is still in its nascent stage and up till now there were no algorithms that could match or outperform commercial tools in terms of speed and accuracy for large circuits. We propose an ATPG algorithm HybMT in this paper that finally breaks this barrier Like sister methods, we augment the classical PODEM algorithm that uses recursive backtracking. We design a custom 2-level predictor that predicts the input net of a logic gate whose value needs to be set to ensure that the output is a given value (0 or 1). Our predictor chooses the output from among two first-level predictors, where the most effective one is a bespoke neural network and the other is an SVM regressor. As compared to a popular, state-of-the-art commercial ATPG tool, HybMT shows an overall reduction of 56.6% in the CPU time without compromising on the fault coverage for the EPFL benchmark circuits. HybMT also shows a speedup of 126.4% over the best ML-based algorithm while obtaining an equal or better fault coverage for the EPFL benchmark circuits.

Malicious modification of integrated circuits (ICs) in untrusted foundry, referred to as “Hardware Trojan”, has emerged as a serious security threat. While side-channel analysis has been reported as an effective approach to detect hardware Trojans, increasing process variations in nanoscale technologies pose a major challenge, since process noise can easily mask the Trojan effect on a measured side-channel parameter, such as supply current. Besides, existing side-channel approaches suffer from reduced Trojan detection sensitivity with increasing design size. In this paper, we propose a novel scalable side-channel approach, named self-referencing, along with associated vector generation algorithm to improve the Hardware Trojan detection sensitivity under large process variations. It compares transient current signature of one region of an IC with that of another, thereby nullifying the effect of process noise by exploiting spatial correlation across regions in terms of process variations. To amplify the Trojan effect on supply current, we propose a region-based vector generation approach, which divides a circuit-under-test (CUT) into several regions and for each region, finds the test vectors which induce maximum activity in that region, while minimizing the activity in other regions. We show that the proposed side-channel approach is scalable with respect to both amount of process variations and design size. The approach is validated with both simulation and measurement results using an FPGA-based test setup for large designs including a 32-bit DLX processor core (~105 transistors). Results shows that our approach can find ultra-small (<0.01% area) Trojans under large process variations of up to ± 20% shift in transistor threshold voltage.