<p>The goal of this project was to implement and compare Invere Discrete Cosine Transform using three methods i.e. by bit parallel, digit serial and bit serial. This application describes a one dimensional Discrete Cosine Transform by bit prallel method and has been implemented by 0.35 ìm technology. When implementing a design, there are several considerations like word length etc. were taken into account. The code was implemented using WHDL and some of the calculations were done in MATLAB. The VHDL code was the synthesized using Design Analyzer of Synopsis; power was calculated and the results were compared.</p>

The goal of this project was to implement and compare Invere Discrete Cosine Transform using three methods i.e. by bit parallel, digit serial and bit serial. This application describes a one dimensional Discrete Cosine Transform by bit prallel method and has been implemented by 0.35 ìm technology. When implementing a design, there are several considerations like word length etc. were taken into account. The code was implemented using WHDL and some of the calculations were done in MATLAB. The VHDL code was the synthesized using Design Analyzer of Synopsis; power was calculated and the results were compared.

In recent years, there has been a significant push to implement Deep Neural Networks (DNNs) on edge devices, which requires power and hardware efficient circuits to carry out the intensive matrix-vector multiplication (MVM) operations. This work presents hardware efficient MVM implementation techniques using bit-serial arithmetic and a novel MSB first computation circuit. The proposed designs take advantage of the pre-trained network weight parameters, which are already known in the design stage. Thus, the partial computation results can be pre-computed and stored into look-up tables. Then the MVM results can be computed in a bit-serial manner without using multipliers. The proposed novel circuit implementation for convolution filters and rectified linear activation function used in deep neural networks conducts computation in an MSB-first bit-serial manner. It can predict earlier if the outcomes of filter computations will be negative and subsequently terminate the remaining computations to save power. The benefits of using the proposed MVM implementations techniques are demonstrated by comparing the proposed design with conventional implementation. The proposed circuit is implemented on an FPGA. It shows significant power and performance improvements compared to the conventional designs implemented on the same FPGA.

The computational abilities of today's parallel supercomputers are often quite impressive, but these machines can be impractical for some researchers due to prohibitive costs and limited availability. These researchers might be better served by a more personal solution such as a "hardware acceleration" peripheral for a PC. FPGAs are the ideal device for the task: their configurability allows a problem to be translated directly into hardware, and their reconfigurability allows the same chip to be reprogrammed for a different problem. Efficient FPGA computation of parallel problems calls for cellular computing, which uses an array of independent, locally connected processing elements, or cells, that compute a problem in parallel. The architecture of the computing cells determines the performance of the FPGA-based computer in terms of the cell density possible and the speedup over conventional single-processor computation. This thesis presents the design and performance results of four computing-cell architectures. MULTIPLE performs all operations in one cycle, which takes the least amount of time but requires the most chip area. BIT performs all operations bit-serially, which takes a long time but allows a large cell density. The two other architectures, SINGLE and BOOTH, lie within these two extremes of the area/time spectrum. The performance results show that MULTIPLE provides the greatest speedup over common calculation software, but its usefulness is limited by its small cell density. Thus, the best architecture for a particular problem depends on the number of computing cells required. The results also show that with further research, next-generation FPGAs can be expected to accelerate single-processor computations as much as 22,000 times. / Master of Science

<p>Bit-serial multiplication with a fixed coefficient is commonly used in integrated circuits, such as digital filters and FFTs. These multiplications can be implemented using basic components such as adders, subtractors and D flip-flops. Multiplication with the same coefficient can be implemented in many ways, using different structures. Other studies in this area have focused on how to minimize the number of adders/subtractors, and often assumed that the cost for D flip-flops is neglectable. That simplification has been proved to be far too great, and further not at all necessary. In digital devices low power consumption is always desirable. How to attain this in bit-serial multipliers is a complex problem. </p><p>The aim of this thesis was to find a strategy on how to implement bit-serial multipliers with as low cost as possible. An important step was achieved by deriving formulas that can be used to calculate the carry switch probability in the adders/subtractors. It has also been established that it is possible to design a power model that can be applied to all possible structures of bit- serial multipliers.</p>

Bit-serial multiplication with a fixed coefficient is commonly used in integrated circuits, such as digital filters and FFTs. These multiplications can be implemented using basic components such as adders, subtractors and D flip-flops. Multiplication with the same coefficient can be implemented in many ways, using different structures. Other studies in this area have focused on how to minimize the number of adders/subtractors, and often assumed that the cost for D flip-flops is neglectable. That simplification has been proved to be far too great, and further not at all necessary. In digital devices low power consumption is always desirable. How to attain this in bit-serial multipliers is a complex problem. The aim of this thesis was to find a strategy on how to implement bit-serial multipliers with as low cost as possible. An important step was achieved by deriving formulas that can be used to calculate the carry switch probability in the adders/subtractors. It has also been established that it is possible to design a power model that can be applied to all possible structures of bit- serial multipliers.

Submitted by Johnny Rodrigues (johnnyrodrigues@ufcg.edu.br) on 2017-09-13T17:59:11Z No. of bitstreams: 1 Utilizacao de Aritmetica Bit-serial para Reducao de Consumo de Energia-Roberto Medeiros de Faria.pdf: 1661698 bytes, checksum: c7ef8816ca92eeeed7c8d271bc93933a (MD5) / Made available in DSpace on 2017-09-13T17:59:11Z (GMT). No. of bitstreams: 1 Utilizacao de Aritmetica Bit-serial para Reducao de Consumo de Energia-Roberto Medeiros de Faria.pdf: 1661698 bytes, checksum: c7ef8816ca92eeeed7c8d271bc93933a (MD5) Previous issue date: 2014-12 / Hoje, uma das maiores preocupações, senão a maior, da indústria de semicondutores é o desenvolvimento de chips com baixo consumo de energia. Existem vários fenômenos físicos causadores de consumo de energia em circuitos CMOS e várias técnicas que reduzem o consumo de energia de um chip. O objetivo principal desta pesquisa de mestrado foi investigar o quanto o consumo de energia estática em circuitos CMOS pode ser reduzido por meio do emprego de aritmética bit-serial em substituição à aritmética bit-paralela. A pesquisa está focada em circuitos construídos a partir de standard cells (células padrão), com aplicação em processamento de sinais, e para os quais o principal requisito não é o alto desempenho computacional, mas o baixo consumo de energia. A metodologia foi aplicada em um estudo de caso, utilizando-se para isto, simulações com o IP core SPVR. O SPVR é um verificador de identidade vocal implementado em um circuito dedicado capaz de ter desempenho suficiente para funcionar em tempo real, mesmo empregando um sinal de clock lento. Foi constatado na pesquisa, que o uso de aritmética bit-serial, em termos de diminuição de consumo estático, é vantajoso para somadores e circuitos de pequena complexidade. Porém, para sistemas de maior complexidade, esta substituição só é vantajosa em situações específicas de grande número de operações aritméticas e baixo uso de armazenamento em registradores paralelos. No caso inverso, as vantagens se perdem, porque embora haja diminuição de consumo estático, há um crescimento muito grande de consumo dinâmico. / Today, one of the biggest concerns, if not the largest, for the semiconductor industry is the development of chips with low power consumption. There are several physical phenomena that cause power consumption in CMOS circuits and various techniques that reduce the energy consumption of a chip. The main objective of this masters research was to investigate how the static power consumption in CMOS circuits can be reduced through the use of bit-serial arithmetic in place of bit-parallel arithmetic. The research is focused on circuits built from standard cells, with application to signal processing, and for which the main requirement is not the high computing performance, but the low power consumption. The methodology was applied in a case study, using for this, simulations with the SPVR IP core. The SPVR is a vocal identity checker implemented in a dedicated circuit able to have enough performance to run in real time, even employing a slow clock signal. It has been found in research that the use of bit-serial arithmetic, in terms of reduction of static consumption, is advantageous to adders and small circuit complexity. However, for more complex systems, this substitution is only advantageous in specific situations of large number of arithmetic operations and low storage usage in parallel registers. In the reverse case, the advantages are lost, because although there are static consumption decrease, there is a very large dynamic consumption growth.