The first topic of this thesis is to carry out a multi-symbol codec (encoder-decoder) design for interfacing variable-length and fixed-length data conversion of H.263. The poor memory efficient of the variable-length can be avoided while its advantages can be reserved. The proposed codec converts variable-length symbols to fixed-length packets which can be decoded parallelly. The basic idea is to encode extra symbols in the redundant bits of the fixed-length packets. This encoding scheme relaxes the intrinsic poor compression rate of the prior fixed-length data codec. The second topic is a synthesizable Verilog code generator for the mentioned multi-symbol codec. According to different requirements and constraints of encoding bit rate, the generator can provide several different kinds of encoding modes by selecting proper parameters. Each codec generated by the generator is synthesizable by thorough simulations.

The domain of computer music is an interesting area which combines computer science and music art. We propose a music accompaniment system using a variable-length genetic algorithm. Via the system one can make the music corresponded to his demands. In the style music accompaniment we analyze some important characteristic of pop music, and propose a new chromosome representation scheme to include the concept of rhyme, chord and melody. Chord progression is used as one of the evaluation criterions in this thesis. The system allows a user to input melody, to select emotion and rhyme, and the system will automatically generate the appropriate accompaniment based on the database compiled from some music theory relating to the chord progression. In addition, the system allows a user to select his favorite accompaniment that generated by the system. Based on the user selected accompaniment the system will generate similar accompaniments for the user.

No description available.

The first topic of this thesis proposes a digital video decoder for NTSC. The new fully digital design employs a DDFS (digital direct frequency synthesizer) and an adaptive digital PLL to track and lock the demodulation carrier. The complexity of the digital video decoder, hence, is drastically reduced. The overall cost of the proposed design is 6.0 mm2 (39K gates). The maximum power dissipation is 86 mW at the hightest clock rate which is 21.48 MHz. The second topic is to carry out a codec (encoder-decoder) design for interfacing variable-length and fixed-length data compression. The poor memory efficiency caused by the variable-length words converting into a fixed-length packet such that the compression can be hardwaredly and parallelly processing is significantly improved. The proposed codec is to encode more symbols in the redundant bits of the padding bits of the fixed-length packets. This novel encoding scheme relaxes the intrinsic poor bit rate of the traditional fixed-length data compression.

In this thesis we have proposed a variable-length multiobjective genetic algorithm to solve heterogeneous wireless transmitter placement with multiple constraints. Among many factors that may affect the result of placement, we focus on four major requirements, coverage, cost, data rate demand, and overlap. In the proposed algorithm we release the need for the upper bound number of transmitters that is a major constraint in the existing methods and achieve better wireless transmitter placement while considering the transmitter position and design requirement simultaneously. In experiments, we use the free space propagation model, the large scale propagation model which considers the shadowing effect, and the extended Hata-Okumura model to predict the path loss in a real two dimensional indoor environment, and an outdoor environment and even a real three dimensional outdoor environment. Experimental results show that the proposed algorithm can find many feasible solutions for all test cases under four objectives.

In this work, we show how a variable-length genetic algorithm naturally evolves populations whose mean chromosome length grows shorter over time. A reduction in chromosome length occurs when selection is absent from the GA. Specifically, we divide the mating space into five distinct areas and provide a probabilistic and empirical analysis of the ability of matings in each area to produce children whose size is shorter than the parent generation's average size. Diversity of size within a GA's population is shown to be a necessary condition for a reduction in mean chromosome length to take place. We show how a finite variable-length GA under random selection pressure uses 1) diversity of size within the population, 2) over-production of shorter than average individuals, and 3) the imperfect nature of random sampling during selection to naturally reduce the average size of individuals within a population from one generation to the next. In addition to our findings, this work provides GA researchers and practitioners with 1) a number of mathematical tools for analyzing possible size reductions for various matings and 2) new ideas to explore in the area of bloat control.

Structure is one of the fundamentals of music, yet the complexity arising from the vast number of possible variations of musical elements such as rhythm, melody, harmony, key, texture and form, along with their combinations, makes music modelling a particularly challenging task for machine learning. The research presented in this thesis focuses on the problem of learning a generative model for melody directly from musical sequences belonging to the same genre. Our goal is to develop probabilistic models that can automatically capture the complex statistical dependencies evident in music without the need to incorporate significant domain-specifc knowledge. At all stages we avoid making assumptions explicit to music and consider models that can can be readily applied in different music genres and can easily be adapted for other sequential data domains. We develop the Dirichlet Variable-Length Markov Model (Dirichlet-VMM), a Bayesian formulation of the Variable-Length Markov Model (VMM), where smoothing is performed in a systematic probabilistic manner. The model is a general-purpose, dictionary-based predictor with a formal smoothing technique and is shown to perform significantly better than the standard VMM in melody modelling. Motivated by the ability of the Restricted Boltzmann Machine (RBM) to extract high quality latent features in an unsupervised manner, we next develop the Time-Convolutional Restricted Boltzmann Machine (TC-RBM), a novel adaptation of the Convolutional RBM for modelling sequential data. We show that the TC-RBM learns descriptive musical features such as chords, octaves and typical melody movement patterns. To deal with the non-stationarity of music, we develop the Variable-gram Topic model, which employs the Dirichlet-VMM for the parametrisation of the topic distributions. The Dirichlet-VMM models the local temporal structure, while the latent topics represent di erent music regimes. The model does not make any assumptions explicit to music, but it is particularly suitable in this context, as it couples the latent topic formalism with an expressive model of contextual information.

O objetivo central desta tese é demonstrar a existência de uma estrutura regenerativa visível para cadeias de alcance variável não limitadas. Também apresentamos um algoritmo de identificação de seqüências de instantes de regeneração que converge quase certamente quando o tamanho da amostra diverge. / Our main aim is prove the existence of a regeneration scheme in unbounded variable length chains. We present an algorithm to identify sequences of regeneration times which converges almost surely as the sample length.

Packet switches are used in the Internet to forward information between a sender and receiver and are the critical bottleneck in the Internet. Without faster packet switch designs, the Internet cannot continue to scale-up to higher data rates. Packet switches must be able to achieve high throughput and low delay. In addition, they must be stable for all traffic loads, must efficiently support variable length packets, and must be scalable to higher link data rates and greater numbers of ports. This dissertation investigates a new combined input and crossbar queued (CICQ) switch architecture. Some unbalanced traffic loads result in instability for input queued (IQ) and CICQ switches. This instability region was modeled, and the cause of the instability was found to be a lack of work conservation at one port. A new burst stabilization protocol was investigated that was shown to stabilize both IQ and CICQ switches. As an added benefit, this new protocol did not require a costly internal switch speed-up. Switching variable length packets in IQ switches requires the segmentation of packets into cells. The process also requires an internal switch speed-up which can be costly. A new method of cell-merging in IQ switches reduced this speed-up. To improve fairness for CICQ switches, a block and transfer method was proposed and evaluated. Implementation feasibility of the CICQ switch was also investigated via a field programmable gate array (FPGA) implementation of key components. Two new designs for round robin arbiters were developed and evaluated. The first of these, a proposed priority-encoder-based round robin arbiter that uses feedback masking, has a lower delay than any known design for an FPGA implementation. The second, an overlapped round robin arbiter design that fully overlaps round robin polling and scheduling, was proposed and shown to be scalable, work conserving, and fair. To allow for multi-cabinet implementation and minimization of the size of the cross point buffers, a distributed input port queue scheduler was investigated. This new scheduler minimizes the amount of buffering needed within the crossbar. The two primary contributions of this dissertation are 1) a complete understanding of the performance characteristics of the CICQ switch, and 2) new methods for improving the performance, stability,and scalability of the CICQ switch. This work has shown that the CICQ switch can be the switch architecture of the future.

Wireless network placement of transmitters, such as base stations for 2G and 3G, access points for WLAN, is a NP-hard problem, since many factors have to be considered, like QoS, coverage, cost, etc. In wireless network placement problem, the goal is to find a set of transmitters which achieves the widest coverage on a given map and spends the minimal cost. In this thesis, we propose a novel variable-length genetic algorithm for solving this problem. Most of existing methods for solving wireless network placement problem, to our best knowledge, users must assign an upper bound or a total number of transmitters for placement. Unlike these existing methods, the proposed algorithm can search the optimal number of transmitters automatically. In addition, the proposed algorithm can find near optimal solutions even in heterogeneous transmitters placement problem, i.e., transmitters with different power radius or cost. The results on several benchmarks are very close to the optimal solutions, which validate the capability of the proposed method in finding the numbers, the types, are the positions of transmitters in heterogeneous wireless network environment.