Thesis (doctoral)--Universität, Frankfurt (Main), 2000.

No description available.

Noise variance estimator is very important in communication system which could control the transmission power and help system to reach the maximum capacity. But almost every kind of noise variance estimators assume that channel status is perfectly. Unfortunately the estimation error is always existence. This paper discuss that the maximum likelihood noise variance estimator (ML-NVE) in orthogonal frequency division multiplexing (OFDM) system with estimated channel. We take two kinds of the most popular channel estimation (CE) schemes Least Square (LS) and Modified-Least Square (MLS) as our consideration. And we investigate the situation that the ML-NVE suffered by carrier frequency offset (CFO) due to imperfect synchronization. As result we will show off the mean and variance of ML-NVE in those situations.

Indiana University-Purdue University Indianapolis (IUPUI) / The use of carbon-fiber reinforced composite materials is not novel in the field of motorsports industry. Their use in collapsible structures for crashworthiness is however not fully understood and predicted. Due to the complex failure mechanisms occurring within the material, the energy absorbing capacity cannot be easily pre dicted. The need to understand their contributions in crashworthy structures is thus of great importance. Furthermore, failure of carbon-fiber composites is highly depen dent on the geometry of structure. Problems arise in both experimental and numerical modelling of these structures. Although many explicit FEA codes exist, they often include experimental parameters that need to be calibrated through either coupon tests or actual crash tests. As composite structures become more commonly used in automotive industry, it is necessary to set some guidelines to successfully model and simulate composite crashworthy structures. The numerical modelling was done in LS-DYNA Enhanced composite damage MAT54. The material properties were configured using experimental coupon tests. The tests were conducted on square composite tubes. The Specific Energy Absorption (SEA) of the tubes were calculated through several coupons. As SEA is a function of geometry, it was necessary to conduct tests with similar geometry as seen in nosecone. MAT54 was chosen to simulate both crush and crash simulations due to its capability to simulate element level crushing. Furthermore, various modifications within the material model, improve its accuracy to determine composite failure. The research utilizes the characterization of material inputs in MAT54 by con ducting quasi-static compression tests on simpler but similar geometry. By utilizing inputs, a zonal optimization was conducted on the nosecone geometry. The number of layers, layer orientations and ply thicknesses were varied to vary the energy absorbed per zone. The deceleration of the vehicle can thus be controlled, and the weight of the structure could be reduced.

In this thesis, least squares support vector machine for regression (LS-SVR) is applied to image coding. First, we propose five simple algorithms for solving LS-SVR. For linear regression, two simple Widrow-Hoff-like algorithms, in primal form and in dual form, are proposed for LS-SVR problems. The dual form of the algorithm is then generalized to kernel-based nonlinear LS-SVR. The elegant and powerful two-parameter sequential minimization optimization (2PSMO) and three-parameter sequential minimization optimization (3PSMO) algorithms are provided in detail. A predictive function obtained from LS-SVR is utilized to approximate the gray levels of the image. After pruning, only a subset of training data called support vectors is saved. Experimental results on seven image blocks show that the LS-SVR with Gaussian kernel is more appropriate than that with Mahalanobis kernel with a covariance matrix. Two-layer LS-SVR is proposed to choose the machine parameters of the LS-SVR. Before training outer LS-SVR, feature extraction is used to reduce the input dimensionality. Experimental results on three whole images show that the results with two-layer LS-SVR after reducing dimensionality are better than those with two-layer LS-SVR without reducing dimensionality in PSNR for Lena and Baboon images and they are almost the same in PSNR for F16 image.

Computer simulations are widely used in the truck industry in order to provide assistance in the product development. Bolt joints are common in trucks. A bolt fracture usually has a great influence on how a truck structure will behave in a crash. Therefore, when simulating truck crashes it is important to be able to predict when bolt fracture occurs. A material model for 10.9 bolts has been calibrated and validated by using the finite element software LS-DYNA. The material model consists of a failure strain surface, which depends on the triaxiality, Lode parameter and the element size. In this thesis, the calibrated material model is referred to as the bolt model. A good agreement to predict the force at fracture in bolts between simulation model results and physical test results has been obtained. Still, further validation is needed to evaluate the bolt model completely.

Formula One race cars have to pass rigorous safety tests before they are allowed on track. This type of testing has been in place for years but the requirements for testing are continually increasing in order to reduce the amount of risk to the drivers’ safety during a race. The number of structures that need to be made and tested can quickly make this process an expensive one. Additionally, it is necessary to pass the mandated tests within a reasonable amount of time so as not to have an impact on the development on the rest of the car. There is a desire to reduce the number of structures needed for testing through finite element analysis (FEA), and as such, to reduce the time needed to pass the safety tests. FEA of laminated composites can be complex and is a balance between accuracy and the time it takes to find a solution. The current project looks into increasing understanding of the requirements for material characterisation, experimental impact testing, and explicit simulation of a carbon fibre fabric pre-impregnated with epoxy resin. Mercedes-Benz Grand Prix (MGP) Formula One Team has provided a pre-preg material for evaluation. Material experiments were performed per the American Society for Materials and Testing (ASTM) in order to find the tensile modulus, tensile strength, Poisson’s ratio, compressive strength, shear modulus, and shear strength of the material. Nine tubes were manufactured at MGP and tested in the drop tower at the Cranfield Impact Centre (CIC) ... [cont.].

The design of new forming tools get more problemtic as the geometries get more complicated and the materials less formable. The idea with this project is to evaluate if an implementation of a simulation software in the designing process, to simulate the forming process before actually building the tools, could help Duroc Tooling avoid expensive mistakes. To evaluate this, the commercial FEM simulation software LS-DYNA was used in a complicated project, where the design of the forming tools for forming a girder was considered. The main objective was to avoid cracking and severe wrinkling which may result in the forming process. With help of simulations a stable forming process which did not yield cracks or severe wrinkling, was eventually found. The girder was almost impossible to form without cracking, but the breakthrough came when we tried to simulate a preforming step which solved the problem. Without a simulation software this would never have been tested since it would be to risky and expensive to try an idea which could turn out to be of no use. The simulations also showed that the springback - shape deformation occuring after pressing - was large and hard to predict without simulations. Therefore, the tools were also finally springback compensated. We concluded that simulations are very effective to quickly test new ideas which may be necessary when designing the tools for forming complicated parts. Simulation also provided detailed quantitative information about the expected cracks, wrinkles, and weaknesses of the resulting pieces. Even though there is cost associated with simulations, it is obvious from this project that a simulation software is a must if Duroc Tooling wants to be a leading company in sheet metal forming tools, and stand ready for the higher demands on the products in the future.

In this thesis, we propose a modified MT-CDMA system, which can improve channel estimation accuracy by using transmit diversity of pilot signals. We not only expound the principles and structures of the system we proposed, but analyze its performance in slow Rayleigh fading channel environment. According to different ways to assign data symbols of transmitted signals, we have two different bit error rate results. At the same transmit power, the simulation results show that when we combine comb-type pilot signals of two parallel channels to estimate channel gains, we can recover the drawbacks of comb-type pilot arrangement, which can not perform well in frequency selective fading channel. Finally, the numerical results will be also shown.

The technique of direct-sequence (DS) code division multiple access (CDMA) cellular systems has been the focus of increased attention. In this thesis, the problem of narrow-band interference (NBI) cancellation for the DS-CDMA communication systems is considered. It has been shown that the performance of single NBI cancellation for CDMA systems by using the non-linear filtering approach, the so-called DDK filter or the MDK filter, is superior to the one using the linear filtering approach. The main concern of this thesis is to deal with the multiple NBI cancellation. This may occur in some practical application, for instance, in the 2.4GHz CDMA system, the bluetooth and wireless LAN may exist in the same frequency band with different power ratio. In this thesis, the nonlinear filtering with fast convergence least square (LS) algorithms, viz., the modified inverse QRD-RLS (IQRD-RLS) and the interior point (IP) LS algorithms, are devised for multiple NBI cancellation in the multi-user CDMA system. In fact, the IQRD-RLS and the IP LS algorithms are known to have better numerical stability and convergence property in the RLS family. Since in the non-linear MDK filter with the IQRD-RLS algorithm, the prediction error £`k,k-1 used in the conventional IQRD-RLS is replaced by the nonlinear function of £l(£`k,k-1), and is defined to as the modified IQRD-RLS algorithm. The merits of the proposed algorithms are verified via computer simulation. We showed that the performance of our proposed algorithms outperformed the one using the conventional nonlinear filtering approach with LMS algorithm, in terms of convergence property and the signal-to-noise ratio improvement (SNRI).