Cloud computing is an emerging, revenue generating and internet based technology, where computing power, storage and other resources are provided to the stakeholders in a single package. The traditional online banking systems can make use of the cloud framework for providing economical and high-speed online service to the consumers. This paper first describes a systematic Multi-factor bio-metric Fingerprint Authentication (MFA) approach which provides a high-secure identity verification process for validating the legitimacy of the remote users. The significance of this approach is that the authentication credentials of the users are not revealed to the bank and cloud authentication servers, but allows the servers to perform remote users' authentication. We then extend this investigated framework to develop a privacy protection gateway for obscuring and desensitizing the customers' account details using tokenization and data anonymization techniques. This approach retains the original format of data fields at various levels of the database management systems and makes the data worthless to others except the owner. In addition to designing an efficient MFA, through extensive experimental results we illustrate our privacy protection gateway is practical and effective.

Trusted framework for online banking in public cloud using multi-factor authentication and privacy protection gateway

One of the most efficient methods for cracking passwords, which are hashed based on different cryptographic algorithms, is the one based on “Rainbow Tables”. Those lookup tables offer an almost optimal time-memory tradeoff in the process of recovering the plaintext password from a password hash, generated by a cryptographic hash function. In this paper, the first known such generic system is demonstrated. It is implemented in a state-of-the-art reconfigurable device that cracks passwords, which are encrypted with a number of different cryptographic algorithms. The proposed FPGA-based system is up to 1000 times faster than the corresponding software approach. This is achieved by using a highly parallel architecture employing a fine-grained pipeline.

Implementing Rainbow Tables in High-End FPGAs for Super-Fast Password Cracking

Following the repeated reports of high prevalence of functional illiteracy in the countries of the Western Balkans, we go on to study the password composition habits of cultural groups from this region. Ina password cracking experiment, we obtain and analyze a number of passwords composed for use in learning management systems (LMS) by recent high-school graduates from North Macedonia. We estimate the strength of these passwords and their crackability by methods adjusted to local cultural specifics. Our results indicate that high extent of general illiteracy is intertwined with low awareness about good password practices, which in turn may lead to creating various vulnerabilities to deployed systems and confidential data. We conclude that, in the studied region, there is an immediate need for educating the population of good password practices and good system and data security practices in general.

The Password Literacy in North Macedonia: A Case Study

Hash functions are crucial cryptographic primitives and are widely used in security protocols. Up to now SHA-2 is the proposed NIST standard for hash functions, but is going to be substituted by the end of the next year. NIST is holding up a public competition to select a new hash algorithm(s), called SHA-3 (Secure Hash Algorithm Version 3), for the purpose of completely superseding the functions of the SHA-2 family. The four finalists have been selected by NIST, and a year is allocated for the public review of these algorithms. BLAKE hash functions family is one of the most promising finalists. A multi-module hashing system for this hash functions family, and the FPGA integration of it, are proposed in this work. The introduced system supports a multi-module operation in the sense that upon to the users selection performs efficiently for all hash functions of BLAKE family: -28, -32, -48, -64. The proposed multi-module system is compared with the stand alone implementations of each of the hash functions, integrated in separate FPGA devices. Comparisons with previously published related works are also presented, in order to have a fair and detailed evaluation of the proposed design methodology. Compared with them, the proposed multi-module system performs with higher operation frequency and allocates less silicon area resources. Finally, the proposed performs better, compared with other ones, in the term of throughput on a range of 100% to 500% and better at about 50%, compared with area-delay product factor.

Multi-module Hashing System for SHA-3 & FPGA Integration

The quest for developing computer architectures in terms of accuracy, efficiency and cost is unending to tackle a very large-scale system, that have a performance problem, using the parallel processing. The mature technology of the mobile agents is gaining more momentum to be adopted with the parallel processing over a grid computing in contrast to the traditional parallel applications, that base on either a dedicated hardware or software. In this paper, the mobile agents technology was proposed as a new approach to achieve a grid computing-based parallel processing for Windows 7 password recovery problem by using the brute-force procedure and realizing how the mobile agents technology can be used to execute complicated applications from the High performance computation community. In this work a weak migration in intra-platform mobile agent was used. The experimental results demonstrate the computational power of the proposed system.

/pdf/parallel-high-performance-windows-7-password-recovery-using-2sjbzcg3og.pdf

Parallel High-Performance Windows 7 Password Recovery using Weak Migration in Intra-Platform Mobile Agents Technology

One of the most efficient methods for cracking passwords is the one based on ldquorainbow tablesrdquo; those lookup tables are offering an almost optimal time-memory tradeoff in the process of recovering the plaintext password from a password hash generated by a cryptographic hash function. In this paper, we demonstrate the first known system, implemented in a state-of-the-art reconfigurable device that cracks passwords up to 1000 times faster than the software approach. This is achieved by using a highly parallel architecture employing a fine-grained pipeline.

K. Theoharoulis

Papers

HighEnd Reconfigurable Systems for Fast Windows' Password Cracking

K. Theoharoulis

Papers

High­End Reconfigurable Systems for Fast Windows' Password Cracking

HighEnd Reconfigurable Systems for Fast Windows' Password Cracking