A carregar...
Publicação
Gerador de números aleatórios integrado em tecnologia CMOS
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 4.24 MB | Adobe PDF |
Orientador(es)
Resumo(s)
Desde os primórdios da civilização humana, foram inventadas inúmeras formas de
comunicação, surgindo, assim, a necessidade de tornar essas formas de comunicação privadas.
Desta forma, considera-se que a criptografia existe desde então.
No entanto, com o início da era digital, a quantidade de informação transmitida
aumentou exponencialmente. Consequentemente, a forma como a privacidade das comunicações
é mantida deixa de ser a única questão abordada, levando-nos à seguinte
problemática:
"Como proteger um elevado número de mensagens sensíveis de forma sistemática?"
A solução para esta questão são os Geradores de Números Aleatórios, RNG. Estes
sistemas têm a capacidade de gerar chaves que, ao misturar as mensagens, conseguem
escondê-las de forma rápida e simples.
Existem duas categorias de geradores de números aleatórios: os verdadeiramente
aleatórios e os pseudoaleatórios.
Pretende-se estudar uma fonte de entropia baseada no ruído do oscilador e, para
atingir este objetivo, propôs-se um circuito gerador de números aleatórios que disponha
de um consumo, custo e área reduzidos e uma elevada aleatoriedade. Através do circuito
proposto na presente dissertação, um gerador de números aleatórios híbrido - circuito
composto por osciladores e um circuito caótico - os objetivos relativos à área e ao consumo
de potência foram cumpridos, tendo o circuito 1,19 mW de potência consumida, 34,5 m2
de área de transístores e um throughput de 26 Mbit/s. No entanto, não foram reunidas as
condições necessárias para se testar estatisticamente o circuito quanto à sua aleatoriedade,
sendo que, teoricamente, o sistema apresentado deverá comportar-se como um TRNG.
From the beginning of human civilization, several means of communication were invented and, there was a surge in the need to make the communication private, thus it is considered that cryptography exists since then. Nonetheless, with the beginning of the digital era, the amount of shared information exponentially grew. Consequently, the means of effectively hide the information is not the only concern, due to the amount of information, which brings a very important question: “How can we systematically hide large amounts of information?” The solution to this question is random number generators (RNG). These systems have the capacity to generate cryptographic keys which, when mixed with the information, hide it in an efficient and timely manner. There is two categories of RNG, being truly random (TRNG) or pseudorandom (PRNG). The objective was to study the entropy source based on the noise of an oscillator, and to achieve that, a RNG circuit was designed to have a low power consumption, a high randomness and a low cost and area usage. The chosen architecture for this dissertation is a hybrid RNG, which uses oscillators and a chaotic circuit to generate the random bits. With the simulation of the circuit, it was found to be at the objectives mark, having 1,19mWof power, 34,5 m2 of area of transistors and a throughput of 26 Mbit/s. However, due to limitations with the simulation, it wasn’t possible to run all the statistical tests, although all the run testes were passed.
From the beginning of human civilization, several means of communication were invented and, there was a surge in the need to make the communication private, thus it is considered that cryptography exists since then. Nonetheless, with the beginning of the digital era, the amount of shared information exponentially grew. Consequently, the means of effectively hide the information is not the only concern, due to the amount of information, which brings a very important question: “How can we systematically hide large amounts of information?” The solution to this question is random number generators (RNG). These systems have the capacity to generate cryptographic keys which, when mixed with the information, hide it in an efficient and timely manner. There is two categories of RNG, being truly random (TRNG) or pseudorandom (PRNG). The objective was to study the entropy source based on the noise of an oscillator, and to achieve that, a RNG circuit was designed to have a low power consumption, a high randomness and a low cost and area usage. The chosen architecture for this dissertation is a hybrid RNG, which uses oscillators and a chaotic circuit to generate the random bits. With the simulation of the circuit, it was found to be at the objectives mark, having 1,19mWof power, 34,5 m2 of area of transistors and a throughput of 26 Mbit/s. However, due to limitations with the simulation, it wasn’t possible to run all the statistical tests, although all the run testes were passed.
Descrição
Palavras-chave
Criptografia Aleatoriedade Oscilador Caótico PRNG TRNG