China
Africa
Explore chapters and articles related to this topic
Time series analysis and forecasting
Published in Amithirigala Widhanelage Jayawardena, Fluid Mechanics, Hydraulics, Hydrology and Water Resources for Civil Engineers, 2021
Amithirigala Widhanelage Jayawardena
Random sequences can be generated using the Monte Carlo technique, or tossing a coin. Box and Muller (1958) proposed a method of generating Standard Normal random numbers ε1 and ε2: ε1=(ln(1u1))1/2cos(2πu2)ε2=(ln(1u1))1/2sin(2πu2)
Modelling risk effect using Monte Carlo Technique
Published in Stephen O. Ogunlana, Prasanta Kumar Dey, Risk Management in Engineering and Construction, 2019
Random Numbers can be generated using a computer source of pseudo-random numbers. Many off-shelf applications (such as MS Excel) have built-in functions to generate RNs. Pseudo-random numbers generation is an algorithm to produce a fixed and deterministic sequence numbers that can at best be called ‘Pseudo-Random’ which behaves, according to statistical tests, like a truly random sequence. Pseudo-Random numbers are uniformly distributed within the unit interval (0,1) with equal likelihood. Uniformly distribution random number provides a basis for generating the random varieties required in a wide variety of realistic simulation problems. It is not correct to use the same random number to sample all distributions on a specific pass. The reason for this is that using the same random number would automatically imply fixed values for all variables (all values will be near their upper or lower limits).
Time series analysis and forecasting
Published in A. W. Jayawardena, Environmental and Hydrological Systems Modelling, 2013
Random sequences can be generated using the Monte Carlo technique, or tossing a coin. Box and Muller (1958) proposed a method of generating standard normal random numbers ε1 and ε2: () ε1=(ln(1u1))1/2cos(2πu2) () ε2=(ln(1u1))1/2sin(2πu2)
A mathematically-based study of the random wheel-rail contact irregularity by wheel out-of-roundness
Published in Vehicle System Dynamics, 2022
Random number, also called as random sequence, is usually classified into two types – true random number and pseudo random number, according to the generation method [36]. The true random number is generated by physical methods, which is unpredictable before the generation, and has a lower generation rate and a very complicated implement process. Thus, it is limited to some specific research fields, although it meets the requirements of various randomness indexes. The pseudo random number is usually generated mathematically, and it is predictable in nature according to a given algorithm. Obviously, the pseudo random number is impossible to be a true random number; however, a random sequence with good statistical properties can be generated by choosing a fine computing method and reasonable parameters. On the other hand, the pseudo random number has advantages of the faster generation and simpler implementation, compared to the true random number. Therefore, the pseudo random number is employed in the present study.
Sensitivity analysis of a three-invariant plasticity model with different sampling algorithms
Published in International Journal of Geotechnical Engineering, 2023
Xuejun Li, Sheng-Wei Chi, Craig D. Foster
The pseudo-random sampling method is typical for creating a random sampling sequence. It is a convenient replacement for generating a truly random sequence by some physical techniques. The sequence generated by the pseudo-random technique has the advantage of being low discrepancy; however, it is not truly random and is dependent on the initial seed value, thus requiring sufficiently large sampling points (Wen and Yu 2019).
A 0.7 pJ/bit, 1.5 Gbps Energy-Efficient Image-Based True Random Number Generator
Published in IETE Journal of Research, 2023
Dhirendra Kumar, Lakshmi Likhitha Mankali, Prasanna Kumar Misra, Manish Goswami
The highlights of the proposed paper are therefore 72.97%, 13.04%, and 85.71% power saving over reported works [2,10,23], respectively.More than 1.01×, 82.41×, 42.86×, 1388.88×, and 5555.55× speed as compared to other state-of–the-art works [2,3,10,13,21], respectively.3.57×, 47.14×, 14.28×, and 4.14× higher energy-efficiency than previously designed TRNGs [2,10,13,23], respectively.Monte Carlo simulation and Corner analysis results revealed the robustness of work and least sensitivity to process variation. This parametric analysis also demonstrates independency of the design with temperature and supply perturbation, respectively.The generated random bit streams passed all NIST test successfully with 0.45 average p-value (essential pass threshold >0.01) and Shannon entropy as 0.99999, while the calculated autocorrelation factor (ACF) that is approximately 0 (ideally 0) within 96% confidence bounds of a Gaussian distribution confirmed the validity of proposed work.The proposed TRNG is designed to use as a key stream generator for cryptographic application. For such an application, the random sequence is required to have the high level of uniformity, independency and unpredictability. The statistical test analysis results of Chi-Square Test-Level-2, Kolmogorov–Smirnov tests and NIST 800.22 [25] have justified the uniformity, independency and unpredictability behavior in the proposed design.The image-based entropy sources are mainly used to provide the requirements of high throughput and low energy efficiency. However, the digital random source-based TRNG demands huge power dissipation to provide the high throughput and energy efficiency [2,10,23]. Moreover, digital random sources often require extreme circuit complexity.