In this thesis, I characterized the association of Arf4GDP with ER-Golgi intermediate compartment membranes. We confirmed that GDP-arrested Arf4 mutants associated with membranes irrespective of nature of tag or mutation. Recruitment appeared specific since loss of N-terminal myristoylation abolished binding. Surprisingly, mutations of residues unique to class II Arfs did not prevent recruitment of Arf4 to peripheral puncta. We then examined the failure of the GDP-arrested Arf4 mutant to disrupt Golgi structure. We identified residues R79 and E113 (likely involved in salt bridge interaction) only present in Arf1 and Arf5 as critical to the ability of their GDP-arrested mutants to disrupt Golgi structure. As predicted, introduction of these residues transformed Arf4GDP into a dominant negative mutant. Interestingly, overexpression of the putative ArfGDP receptor membrin prevented the effects of dominant negative Arf1 but not dominant negative Arf4. These results will facilitate identification of a novel Arf target critical to protein trafficking.

Diss. (sammanfattning) Linköping : Linköpings universitet, 2009. / Härtill 4 uppsatser.

碩士 / 國立清華大學 / 電機工程學系 / 88 / In this thesis, we propose a hardware implementation for a 512-bit RSA cryptosystem core using systolic array structures. Based on modified Montgomery's algorithm, the iteration required is half of the original Montgomery's algorithm and thus has some speedup. Besides, our circuit is also designed for the Chinese Remainder Theorem (CRT) technique. This can further improve the throughput with a maximum factor of 4 in the best case. The processing unit of the systolic array has 100% utilization because of using Block Interleaving ofmultiplication and square operations in the modular exponentiation algorithm. The number of clock cycles needed for a modular exponentiation is only 0.13M in the best case, and 0.24M in the worst case, assuming that we are dealing with 512-bit number. The critical path delay is only 6.13ns, so our design can achieve decryption rate of 578Kb/s and 328Kb/s in the best and worst cases, respectively. This design is suitable for decryption and digital signature.

碩士 / 國立交通大學 / 資訊科學系 / 87 / Identifying outliers and remainder clusters which are used to designate few patterns that much different from other clusters is a fundamental step in many application domain. However, current outliers diagnostics are often inadequate when in a large amount of data. In this thesis, we propose a two-phase clustering algorithm for outliers. In Phase 1 we modified k-means algorithm by using the heuristic "if one new input pattern is far enough away from all clusters'' centers, then assign it as a new cluster center". So that the number of clusters found in this phase is more than that originally set in k-means algorithm. And then we propose a clusters merging process in the second phase to merge the resulting clusters obtained in Phase 1 into the same number of clusters originally set by the user. The results of three experiments show that the outliers or remainder clusters can be easily identified by our method.

碩士 / 逢甲大學 / 資訊工程學系 / 103 / Residue Number System improves the delay of operation. It can be used to solve many problems (e.g., digital signal processor (DSP) [1][2], finite impulse response filter (FIR)[3][4][5] , image and cryptography ). The conversion process from residue number to binary/decimal number system is very demanding. There are three convert methods. One is Look-up-Table (LUT), another is Mix-Radix-Convert (MRC) [9] and the other is Chinese-Residue-Theorem (CRT)[10][11]. LUT requires lots of memory. But its conversion time is the fastest. Mix-Radix-Convert is a linear operation. So its conversion time is the slowest. MRC hardware cost is the smallest. CRT conversion time is better than MRC, but slower than LUT. We select CRT for converting residue number system to binary/decimal number system in this paper. Conversion time will be affected for two reasons. One is number of moduli set, the other is the feature of moduli set. The moduli set we select is {2^n – 1, 2^n , 2^n + 1}[12][13]. Because this moduli set reduces computational complexity. The Altera DE4 supports many devices. It is very suitable to simulate resource-constrained device. We use FPGA on the board to implement and verify the Chinese-Residue-theorem and Residue Number System.

碩士 / 國立中興大學 / 應用數學研究所 / 75 /

碩士 / 國立清華大學 / 資訊系統與應用研究所 / 103 / Due to the fast growth of information technology, the transmission of sensitive information such as military images, commercial reports etc. over insecure communication channels causes a challenging security issue. The image sharing is a useful method to protect our secret images. Based on (k,n) threshold scheme, this method distributes the secret image into n shadow images which are preserved by n participants instead of only a single carrier. If we collect at least k shadow images, we can reveal the secret image. Fewer than k shadow images cannot reveal the secret image. In this thesis, we implement an image sharing and revealing system by the mechanism of Chinese remainder theorem. We encrypt the pixel values using the simple exclusive-or (XOR) cipher, so anyone who wants to reveal the secret image must own the key in encryption. In our proposed method, we can only reveal a distortion image with the least significant bit loss. However, naked eyes cannot distinguish the difference between the secret image and distortion image with only the difference of the least significant bit.

碩士 / 國立清華大學 / 資訊工程學系所 / 106 / For the reason that the rapid growth of information in the era of Big Data nowadays, information security plays an important role. To avoid secret information carried by single carrier, this thesis studied image sharing based on (k,n)-threshold scheme, which distributes a secret image to n shadow images preserved by n participants. Collect at least k of them, we can recover the secret image, but fewer than k of them could not. We proposed a method for the extension of Chuang’s scheme [Chua2016] based on Chinese Remainder Theorem in this thesis. Compared to other image sharing methods, our method is much simpler and efficient. Although the recovered image in our method has the difference of the least significant bit with the secret image, the experimental result shows that human beings can hardly distinguish two pictures with PSNR value over 50db. Moreover, we can additionally process the least significant bit of each pixel during the sharing and recovering parts if we want to recover the image completely.

碩士 / 國立中山大學 / 資訊工程學系研究所 / 107 / In recent years, We store many sensitive datas in databases. But using traditional block ciphers encrypt data in databases, it will change data’s length and format. To store this ciphertexts, it is necessary to change existing database schema, and need a lot of cost. There is no need to change in Format-Preserving Encryption. This technique can make the cipertext has the same length and format as the plaintext. But too short input’s bits is a problem for security. So, we propose a method, it increases the length of bits, and using Chinse Remiander Theorem to query the specific data. It will provide high security.

This paper is a study of Vadstena monastery's book acquisitions and book collection 1374-1595. Studying catalogue descriptions of manuscripts and incunabula in Uppsala University library, the author has aimed to examine the development over a period of time, concentrating on way and place of acquisition, and desired types of texts.A short description and history of the monastery and its library is given. The study opens with the author stating there must have existed a plan for the book acquisitions and collection. A survey investigation shows, that the usable material is too small and somewhat nonrepresenta­tive of the monastery library to form a basis for a fair analysis. Further investigations give material to form a basis for some careful statements, saying that book production and also acquisitions concentrated on sermons literature, and that this type dominated the library collection throughout the time of development, followed by the philosophical-theological university literature.