No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

The interconnect mechanisms (shared bus or crossbar) used in current chip-multiprocessors (CMPs) are expected to become a bottleneck that prevents these architectures from scaling to a larger number of cores. Tiled CMPs offer better scalability by integrating relatively simple cores with a lightweight point-to-point interconnect. However, such interconnects make snooping impractical and, thus, require alternative solutions to cache coherence. This thesis proposes a novel, cost-effective hardware mechanism to support shared-memory parallel applications that forgoes hardware maintained cache coherence. The proposed mech- anism is based on the key ideas that mapping of lines to physical caches is done at the page level with OS support and that hardware supports remote cache accesses. It allows only some controlled migration and replication of data and provides a sufficient degree of flexibility in the mapping through an extra level of indirection between virtual pages and physical tiles. The proposed tiled CMP architecture is evaluated on the SPLASH-2 scientific benchmarks and ALPBench multimedia benchmarks against one with private caches and a distributed direc- tory cache coherence mechanism. Experimental results show that the performance degradation is as little as 0%, and 16% on average, compared to the cache coherent architecture across all benchmarks for 16 and 32 processors.

The conditional independence assumption imposed by the hidden Markov models (HMMs) makes it difficult to model temporal correlation patterns in human speech. Traditionally, this limitation is circumvented by appending the first and second-order regression coefficients to the observation feature vectors. Although this leads to improved performance in recognition tasks, we argue that a straightforward use of dynamic features in HMMs will result in an inferior model, due to the incorrect handling of dynamic constraints. In this thesis I will show that an HMM can be transformed into a Trajectory-HMM capable of generating smoothed output mean trajectories, by performing a per-utterance normalisation. The resulting model can be trained by either maximisingmodel log-likelihood or minimisingmean generation errors on the training data. To combat the exponential growth of paths in searching, the idea of delayed path merging is proposed and a new time-synchronous decoding algorithm built on the concept of token-passing is designed for use in the recognition task. The Trajectory-HMM brings a new way of sharing knowledge between speech recognition and synthesis components, by tackling both problems in a coherent statistical framework. I evaluated the Trajectory-HMM on two different speech tasks using the speaker-dependent MOCHA-TIMIT database. First as a generative model to recover articulatory features from speech signal, where the Trajectory-HMM was used in a complementary way to the conventional HMM modelling techniques, within a joint Acoustic-Articulatory framework. Experiments indicate that the jointly trained acoustic-articulatory models are more accurate (having a lower Root Mean Square error) than the separately trained ones, and that Trajectory-HMM training results in greater accuracy compared with conventional Baum-Welch parameter updating. In addition, the Root Mean Square (RMS) training objective proves to be consistently better than the Maximum Likelihood objective. However, experiment of the phone recognition task shows that the MLE trained Trajectory-HMM, while retaining attractive properties of being a proper generative model, tends to favour over-smoothed trajectories among competing hypothesises, and does not perform better than a conventional HMM. We use this to build an argument that models giving a better fit on training data may suffer a reduction of discrimination by being too faithful to the training data. Finally, experiments on using triphone models show that increasing modelling detail is an effective way to leverage modelling performance with little added complexity in training.

This thesis investigates the classification of the behaviour of multiple persons when viewed from a video camera. Work upon a constrained case of multiple person interaction in the form of team games is investigated. A comparison between attempting to model individual features using a (hierarchical dynamic model) and modelling the team as a whole (using a support vector machine) is given. It is shown that for team games such as handball it is preferable to model the whole team. In such instances correct classification performance of over 80% are attained. A more general case of interaction is then considered. Classification of interacting people in a surveillance situation over several datasets is then investigated. We introduce a new feature set and compare several methods with the previous best published method (Oliver 2000) and demonstrate an improvement in performance. Classification rates of over 95% on real video data sequences are demonstrated. An investigation into how the length of time a sequence is observed is then performed. This results in an improved classifier (of over 2%) which uses a class dependent window size. The question of detecting pre/post and actual fighting situations is then addressed. A hierarchical AdaBoost classifier is used to demonstrate the ability to classify such situations. It is demonstrated that such an approach can classify 91% of fighting situations correctly.