Massey Documents by Type

Permanent URI for this communityhttps://mro.massey.ac.nz/handle/10179/294

Browse

Subject: search.filters.subject.Data encryption (Computer science)

Search Results

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    A platform for practical homomorphic encryption in neural network classification : a thesis submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy (Ph.D.) in Information Technology, Massey University
    (Massey University, 2021) Baryalai, Mehmood
    Convolutional neural networks (CNN) have become remarkably better in correctly identifying and classifying objects. By using CNN, numerous online services now exist that processes our data to provide meaningful insight and value-added services. Not all services are reliable and trustworthy due to which privacy concerns exist. To address the issue, the work presented in this research develops and optimise new techniques to use Homomorphic Encryption (HE) as a solution. Researchers have proposed solutions like the CryptoNets, Gazelle, and CryptoDL. However, homomorphic encryption is yet to see the limelight for real-world adoption, especially in neural networks. These proposed solutions are seen as a solution only for a particular CNN model and lack generality to be extended to a different CNN model. Moreover, the solutions for HE-CNN integration are seen as unprepared for adoption in a practical and real-world environment. Additionally, the complex integration of hybrid approaches limits their utilization with privacy-preserving based CNN models. For that reason, this research develops the mathematical and practical knowledge required to adopt HE within a CNN. This knowledge of performing encrypted classification for a CNN model is based on a careful selection of appropriate encryption parameters. Furthermore, this study succeeds in developing a dual-cloud system to mitigate many of the technical hurdles for evaluating an encrypted neural network without compromising privacy. Moreover, in the case of a single cloud, this study develops methods for overcoming technical issues in selecting encryption parameters for, and evaluating, a convolutional neural network. In the same context, the novel method of selecting and optimizing encryption parameters based on probability is given. The proposals and the knowledge from this research can aid and advance the strategies of HE-CNN integrations in an efficient and easy way.
  • Thumbnail Image
    Item
    Towards implementing RSA-based CP-ABE algorithm on Android system : a thesis presented in partial fulfilment of the requirements for the degree of Master of Information Sciences at Massey University, Auckland, New Zealand
    (Massey University, 2019) Xing, Jiaxin
    Cipher-text-Policy Attribute-Based Encryption (CP-ABE) algorithm has been proposed to encrypt and decrypt data based on the matching between attributes and an access policy placed over cipher-text. Using CP-ABE, data owner can encrypt data along with an access policy to enforce a fine-grained access control. To improve the efficiency of performance, this study chose a RSA-based CP-ABE algorithm with an access-tree structure while most existing CP-ABE has been implemented using ECC. This new RSA-based CP-ABE algorithm was implemented in the Linux system in another study while this thesis addresses an implementation strategy on an Android system. To achieve this goal, a simple encryption application was designed for users who want to encrypt and decrypt messages through their mobile devices. This study used Android Studio to create the encryption application. In this cipher program, users input the message they want to encrypt and get the encrypted data through the function button named “CIPHER”, and they also can decrypt the cipher-text in the same way. There are four main algorithms involved in a CP-ABE scheme. They respectively are setup, key generation, encryption and decryption. During the setup process, the CP-ABE scheme uses the RSA algorithm to choose two prime numbers. These prime numbers are used to a master public key and a master private key. In the key generation algorithm, a secret key is generated for a set of attributes using the master private key. In the encryption step, it creates a cipher-text with an access tree. In the decryption algorithm, if and only if the attributes for the user’s decryption key satisfies this access policy is able to decode the encrypted data. This algorithm uses the construction of lightweight no-paring crypto-system based on RSA, and the construction supports an expressive monotone tree access structure to implement the complex access control as a more generic system. By using this algorithm, the encryption and decryption processes are more efficient and secure.
  • Thumbnail Image
    Item
    Novel lightweight ciphertext-policy attribute-based encryption for IoT applications : a thesis presented in partial fulfilment of the requirements for the degree of Master of Information Science at Massey University, Auckland, New Zealand
    (Massey University, 2018) Li, Ping
    As more sensitive data are frequently shared over the Internet of Things (IoT) network, the confidentiality and security of IoT should be given special consideration. In addition, the property of the resources-constraint nodes raises a rigid lightweight requirement for IoT security system. Currently, the Attribute-Based Encryption (ABE) for fine-grained access control is the state-of-the-art technique to enable the secure data transmission and storage in the distributed case such as IoT. However, most existing ABE schemes are based on expensive bilinear pairing with linear size keys and ciphertexts. This results in the increase of the memory and computational requirement on the devices, which is not suitable for the resource-limited IoT applications. Leveraging on the advantages offered by the Ciphertext-Policy ABE (CP-ABE), this thesis proposes two constructions of lightweight no-paring cryptosystems based on Rivest–Shamir–Adleman (RSA). One realized work is a construction of AND-gate CP-ABE to achieve both constant-size keys and ciphertexts. The result of the evaluation shows that it reduces the storage and computational overhead. The other construction supports an expressive monotone tree access structure to implement the complex access control as a more generic system. Both have respective advantages in different contexts and are provably secure to guarantee the sharing of data, as well as more applicable and efficient than the previous scheme. In this thesis, practical issues are also described about implementations and evaluations of both proposals.