Ad hoc networks have emerged recently as an important trend of future wireless systems. The evolving wireless networks are seriously challenging the traditional OSI layered design. In order to provide high capacity wireless access and support new multimedia network, the various OSI layers and network functions should be considered together while designing the network. In this thesis, we briefly discuss performance optimization challenges of ad hoc networks and cross-layer design. Ad hoc wireless networks were implemented by using Network Simulator NS-2 and the wireless physical, data link, dynamic source routing (DSR) routing protocol models have been included in the simulation. Simulations show that the performance begins to drop at the moderate offered load due to congestion. In addition, the mobility and fading cause the route failures and packet loss in wireless environment. To improve the performance for wireless networks, we implemented a congestion-aware cross-layer design in NS-2. The MAC layer adaptively selects a transmission data rate based on the channel signal strength information from physical layer and congestion information from network layer. The MAC layer utilization gathered at MAC layer is sent to DSR as a congestion aware routing metric for optimal route discovery. We modified the source codes of 802.11 MAC layer and DSR protocol. The simulations show that rate adaptation in MAC layer improves the network performance in terms of throughput, delivery ratio, and end-to-end delay; using congestion information from MAC layer in routing discovery improves the performance of the network benefited from overall network load balance.

The Master’s Thesis comprises a summary of the differences existing design of MNS switchgear series against AGOMIN project innovative proposal. The introduction deals with the introduction to the CAD program SolidWorks and its extension for enterprise data management SolidWorks Enterprise PDM. The body of the paper deals with a new range of air circuit breakers ABB SACE Emax 2 and their implementation into the product line of low voltage switchgears. Included are other improvements that together with implementation of circuit breakers AGOMIN project brings.

Made available in DSpace on 2014-07-29T15:22:59Z (GMT). No. of bitstreams: 1 Giselle - Dissertacao FINAL.pdf: 462400 bytes, checksum: d49d65a0e8eee6016f109e6efdb53aa9 (MD5) Previous issue date: 2010-04-30 / The processing of mineral water is characterized by a system composed of step sequences, input variables, parameters and output variables. Thus, this work was conducted with the objective of implementing a computer model to simulate the steps that comprise the operational flowchart of a mineral water industry, using the software Stella 8.0. The model defined as dynamic, stochastic and discrete consisted of eleven interconnected blocks. It was built following the steps of characterizing the real system, creating the conceptual model, structuring and verification of computational model, data collection, and model validation. From data collected in the real system we obtained fits of the distributions used in assembling and verifying the model with the help of the @ risk 5.5 program. Validation of the program was conducted by comparing the data (number of bottles to be processed and total processing time) of the real system with simulated data, using regression analysis and the program Statistic 6.0 (Statsoft). The values collected in the real system were within the upper and lower limits on the graph obtained by regression analysis, with a confidence level of 95%, ensuring that the computer model adequately represents the real system. Fails in the process were identified by analyzing of collected values needed more processing time than the simulated values, for the same amount of bottles. However, the computational model developed is appliable to simulate the dynamics of processing lines of Mineral Water, being a tool of production control, which allows the visualization of the behavior of the real system, facilitates the identification of errors, improve production flow, and reduces manufacturing costs by controlling the total processing time. / O processamento de água mineral é caracterizado por um sistema composto por etapas seqüenciais, variáveis de entrada, parâmetros e variáveis de saída. Deste modo, este trabalho foi conduzido com o objetivo de implementar um modelo computacional para simular as etapas que integram o fluxograma operacional de uma indústria de água mineral, através do software Stella 8.0. O modelo definido como dinâmico, estocástico e discreto constituiu-se de onze blocos interligados. O mesmo foi construído seguindo as etapas de caracterização do sistema real; criação do modelo conceitual; estruturação e verificação do modelo computacional; obtenção de dados; e validação do modelo. Partindo dos dados coletados no sistema real obteve-se os ajustes das distribuições utilizadas na montagem e verificação do modelo, com o auxílio do programa @risk 5.5. A validação do programa foi efetuada através da comparação dos dados (quantidade de galões a serem processados e tempo total de processamento) do sistema real com os dados simulados, por meio da análise de regressão e do programa Statistica 6.0 (statsoft). Os valores coletados no sistema real apresentaram-se dentro dos limites superior e inferior no gráfico obtido pela análise de regressão, com nível de confiança de 95%, garantindo que o modelo computacional representa de forma adequada o sistema real. Falhas no processo foram identificadas analisando que alguns valores coletados para a mesma quantidade de galões, necessitaram de maior tempo de processamento do que os valores simulados. Contudo, o modelo computacional desenvolvido é aplicável para simular a dinâmica operacional de linhas processadoras de Água Mineral, por ser uma ferramenta de controle da produção, que permite a visualização do comportamento do sistema real, facilita a identificação de erros, melhora o fluxo produtivo e reduz custos industriais por controlar o tempo total de processamento.