In recent years, transmission of video over the Internet has become an important application. As wireless networks are becoming increasingly popular, it is expected that video will be an important application over wireless networks as well. Unlike wired networks, wireless networks have high data loss rates. Streaming video in the presence of high data loss can be a challenge because it results in errors in the video.<p>Video applications produce large amounts of data that need to be compressed for efficient storage and transmission. Video encoders compress data into dependent frames and independent frames. During transmission, the compressed video may lose some data. Depending on where the packet loss occurs in the video, the error can propagate for a long time. If the error occurs on a reference frame at the beginning of the video, all the frames that depend on the reference frame will not be decoded successfully. <p>This thesis presents the concept of mixed streaming, which reduces the impact of video propagation errors in error prone networks. Mixed streaming delivers a video file using two levels of reliability; reliable and unreliable. This allows sensitive parts of the video to be delivered reliably while less sensitive areas of the video are transmitted unreliably. Experiments are conducted that study the behavior of mixed streaming over error prone wireless networks. Results show that mixed streaming makes it possible to reduce the impact of errors by making sure that errors on reference frames are corrected. Correcting errors on reference frames limits the time for which errors can propagate, thereby improving the video quality. Results also show that the delay cost associated with the mixed streaming approach is reasonable for fairly high packet loss rates.

In recent years, transmission of video over the Internet has become an important application. As wireless networks are becoming increasingly popular, it is expected that video will be an important application over wireless networks as well. Unlike wired networks, wireless networks have high data loss rates. Streaming video in the presence of high data loss can be a challenge because it results in errors in the video.<p>Video applications produce large amounts of data that need to be compressed for efficient storage and transmission. Video encoders compress data into dependent frames and independent frames. During transmission, the compressed video may lose some data. Depending on where the packet loss occurs in the video, the error can propagate for a long time. If the error occurs on a reference frame at the beginning of the video, all the frames that depend on the reference frame will not be decoded successfully. <p>This thesis presents the concept of mixed streaming, which reduces the impact of video propagation errors in error prone networks. Mixed streaming delivers a video file using two levels of reliability; reliable and unreliable. This allows sensitive parts of the video to be delivered reliably while less sensitive areas of the video are transmitted unreliably. Experiments are conducted that study the behavior of mixed streaming over error prone wireless networks. Results show that mixed streaming makes it possible to reduce the impact of errors by making sure that errors on reference frames are corrected. Correcting errors on reference frames limits the time for which errors can propagate, thereby improving the video quality. Results also show that the delay cost associated with the mixed streaming approach is reasonable for fairly high packet loss rates.

In recent years there has been increased research activity in the area of Func- tional Data Analysis. Methodology from finite dimensional multivariate analysis has been extended to the functional data setting giving birth to Functional ANOVA, Functional Principal Components Analysis, etc. In particular, some studies have pro- posed inferential techniques for various functional models that have connections to well known areas such as mixed-effects models or spline smoothing. The methodol- ogy used in these cases is computationally intensive since it involves the estimation of coefficients in linear models, adaptive selection of smoothing parameters, estimation of variances components, etc. This dissertation proposes a wide-ranging modeling framework that includes many functional linear models as special cases. Three widely used tools are con- sidered: mixed-effects models, penalized least squares, and Bayesian prediction. We show that, in certain important cases, the same numerical answer is obtained for these seemingly different techniques. In addition, under certain assumptions, an applica- tion of a Kalman filter algorithm is shown to improve the order of computations, by two orders of magnitude, for point and interval estimates (with n being the sample size). A functional data analysis setting is used to exemplify our results.

Zugl.: Ulm, Univ., Diss., 2007 / Hergestellt on demand

Thesis (Ph. D.)--Lehigh University, 2001. / Includes vita. Includes bibliographical references (leaves 151-160).

Thesis (M.U.D.)--University of Hong Kong, 1998. / Includes bibliographical references (leaves 89-94).

A year-long case study of nutrient cycling within an agro-pastoral community of southern Mali revealed substantial differences in nutrient balances between households and between settlement sub regions of the study area. Inter-household differences should, in great part, bb attributed to differential access to livestock products such as manure and animal traction (both ox-ploughs and donkey carts). Exchanges of manure, carts, and ploughs between owners and non-owners of livestock account for the different levels of access. Exchanges within the settlement sub-regions were more numerous than those between them, had exchanges between e pastoral Fulani and the other communities were limited. The farming systems of the sub-regions presented three patterns of agricultural intensification: 1) cotton-centred rotations sing ploughs and carts to increase yields on perm - anently cultivated sites at the core of the village and in the adjacent hamlets,2 ) cotton-centred rotations on sandiers oils of the village periphery supported largely by intensive hand cultivation with the borrowing of carts to transport nutrients,and3 ) cereal-centred cultivation of small fields heavy manure by large cattle herds owned and managed by Fulani, with plough-ownerships seeming to ensure that household labour could remain available for herding. None of these models of intensification correspond fully with the owner/manager operation of a "mixed farm" and suggest that the potential for better crop-livestock integration in the sub-humidz one may be under-estimated and mis-represented by the "mixed farming" model. Although the overall nitrogen balance of the region after the 1996 cropping year was -8.2 kg N/ha, the different systems described above are associated with significantly different balances:the villa.g1 >. e core and hamlet cores were- 3.2 and- 4.7 kg N/ha, the village periphery- 21.4 kg N/ha, and the Fulani system+ 23.3k g N/ha. Phosphorus and potassium were in positive balance throughout the study area. 0 The different degrees of crop-livestock integration associated with each sub-region suggest that the importance of soil fertility problems must be understood with reference to local algro-pastoral practice. The growing livestock population and the presence of transhumant Fulani pastoralists in the sub-humid zone are key components of these practices and must not be excluded. The access of different actors to key resources such as manure, animal traction, pasture and crop-land deten-nine the ability to increase agricultural production and the degree to which agricultural intensification will mine the soil of nutrients. b ateg eneralisations about the extent of Regional or national studies that make aggregate nutrient loss misrepresent the soil fertility of the region, and risk feeding into an unrealistic" crisis narrative".

Regression models with a scalar response and a functional predictor have been extensively studied. One approach is to approximate the functional predictor using basis function or eigenfunction expansions. In the expansion, the coefficient vector can either be fixed or random. The random coefficient vector is also known as random effects and thus the regression models are in a mixed effects framework. The random effects provide a model for the within individual covariance of the observations. But it also introduces an additional parameter into the model, the covariance matrix of the random effects. This additional parameter complicates the covariance matrix of the observations. Possibly, the covariance parameters of the model are not identifiable. We study identifiability in normal linear mixed effects models. We derive necessary and sufficient conditions of identifiability, particularly, conditions of identifiability for the regression models with a scalar response and a functional predictor using random effects. We study the regression model using the eigenfunction expansion approach with random effects. We assume the random effects have a general covariance matrix and the observed values of the predictor are contaminated with measurement error. We propose methods of inference for the regression model's functional coefficient. As an application of the model, we analyze a biological data set to investigate the dependence of a mouse's wheel running distance on its body mass trajectory.

No description available.

There is little information comparing the impacts of bison and cattle grazing in the northern mixed-grass prairies; understanding which grazer to use for management is important. I collected data in sites grazed by bison or cattle using point counts to survey the songbird community, and habitat structure measurements. I used generalized- and non-linear mixed models to determine the relationships between grazers and habitat or songbird communities. Bison and cattle grazing caused changes in habitat structure, including decreased vegetation height and litter depth. Sprague’s pipits declined more steeply with bison grazing than with cattle grazing. 0.4 AUM/ha was an important threshold, because chestnut-collared longspurs increased in abundance at this stocking rate, and grasshopper sparrows decreased in abundance at this stocking rate. Either grazer is appropriate for managing for grassland birds. There is also an opportunity to encourage surrounding landowners to manage their lands for the benefit of wildlife.