Agus Bejo

AES S-box construction using different irreducible polynomial and constant 8-bit vector

S-box plays a major role in the AES algorithm. The strength of S-box depends on the design and algebraic constructions. In this paper, the construction of S-box will be built using a basic polynomial equation and the addition of a constant 8-bit vector different from the standard AES. The quality of the created S-box is evaluated by measuring several standard criteria such as nonlinearity, strict avalanche criterion (SAC), and bit independence criterion-nonlinearity (BIC-Nonlinearity). The evaluation shows that the values of nonlinearity, SAC, and BIC-Nonlinearity are 112, 0.4995, and 112 respectively. This research also found that the proposed S-box construction method outperforms other existing S-boxes construction methods.

GAMA32 processor modeling using SystemC

Nowadays, the demand of embedded processors has been increasing rapidly. However, the design time required for producing a new processor can not catch up the time-to-market demand. In order to solve this problem, a modeling tool which allows designer to crate new hardware and estimate its performace without having the physical products is very important. SystemC is one among existing modeling tools which is widely used for designing hardware. Using SystemC, a processor architecture can be modelled and simulated quickly to speed up the architecture exploration time. In this paper, a new processor architecture called GAMA32 is designed and modelled using SystemC. GAMA32 is a 32-bit RISC processor architecture which is modelled in instruction accurate. It contains 45 instructions consisting of standard GPP instruction-set. For simulation, two memories RAM and ROM are also modelled. These memories are necessary to store program code and data code respectively. Both RAM and ROM are directly connected to the GAMA32 processor. Simulation results show that all instructions can be executed properly.

An improvement of minutiae-based fingerprint matching: Two Level of Scoring System

A number of researchers have proposed numerous methods in fingerprint matching aspect, especially, in term of speed and memory consumption for decades. However, the problem of the matching accuracy remains until recent days, which is obviously caused by distortions. In this paper, the Two Level of Scoring System is an extension modified from original Hough transform (HT) approach, which is proposed to improve the matching accuracy. However, according to our experiment result that yields higher accuracy than the conventional method, it succeeds in conducting EER 0.98 against the conventional approach which achieves EER 3.94. As the result of the experiment, it reveals that the proposed approach offers higher accuracy than the conventional technique in general fingerprint applications. In addition, time consumption of proposed method is slightly different from conventional technique.

A review of chaff point generation methods for fuzzy vault scheme

Bio-encryption is a concept that aims to improve the traditional cryptographic security by combining biometrics and cryptography. Fuzzy Vault scheme is a bio-encryption method which is very famous for its ability to handle fuzzy biometric data while at the same time combines them with cryptographic mechanism. Moreover, Fuzzy Vault scheme is claimed as a bio-encryption method that can be deployed in System-on-Chip (SoC) based security devices. One of the requirements that must be fulfilled by a system that can be run on SoC based devices is less resource consuming. In other words, the system must have sufficient resource efficiency or must use the fairly efficient algorithms. One of the important phases in the Fuzzy Vault scheme is chaff point generation. Chaff points are analogized as a noise used for covering valuable information inside of Fuzzy Vault. Although the chaff point generation phase is a quite important phase, it would be the phase that has the highest computation level. Hence, research on chaff point generation method to find efficient methods while maintaining the security aspects of Fuzzy Vault Scheme became an interesting topic in bio-encryption field for some researchers. This paper will discuss and also will do some comparisons toward three current chaff point generation methods that have been proposed by some researchers. Thus, this paper is expected to contribute knowledge about the characteristics of these three methods.

Optimal resource allocation scheme in femtocell networks using Bacterial Foraging Optimization

The femtocell network has currently been considered as one of the most promising networking technologies. Nevertheless, due to dense deployment, interference generated by nearby femto base station (FBS) is expected to significantly degrade the signal to interference plus noise ratio (SINR) of femto users (FUEs). Ineffective resource block combination is one of the culprits behind the reduction in performance of each FUE. Therefore, appropriate resource block allocation becomes critical in reducing interference while improving SINR, resulting in better system performance. Bacteria Foraging Optimization (BFO) is proposed to select the most appropriate combination of resource block for FUE, leading to the decrease in the amount of interference in femtocell network. The simulation results illustrate that the proposed algorithm is capable of enhancing the SINR in comparison with random allocation scheme.