نشریه مهندسی برق و مهندسی کامپیوتر ایران
سال چهارم شماره 1 (پیاپی 7، بهار و تابستان 1385)

Data hiding is one of the effective solutions for the exchange of confidential messages in un-secure communication channels such as Internet. In this paper, we propose a new and effective steganography method for robust hiding of information in images with JPEG format. The data is hidden in DCT coefficients of JPEG conversion of a bitmap image using a low quantization rate (for instance, 75% quantization rate) and then convert back to bitmap format. The new bitmap image is converted to JPEG format again, but with a higher quantization rate (90%, for instance). This resulted JPEG image, which carries the hidden data, is robust against the known attacks such as compatibility, F5, outguess and POV attacks. The experimental results show that although the available data capacity using the proposed algorithm is reduced and limited compare to the other similar methods, but it has a more effective robustness under the these attacks.

In this paper, a modified multi band spectral subtraction is proposed for speech denoising. In this method by estimating the non-stationary noise statistics an estimate of the SNR for the speech signal is obtained which is used in the adaptive averaging of the speech signal frames. In addition, by employing such estimate an over-subtraction factor can be obtained which substantially reduces the musical noise. The adjustment of the final algorithm parameters is accomplished based on the obtained SNR, such that the final algorithm has a minimum musical noise and distortion in the speech signal. The obtained results from the Itakura-Saito, global SNR and segmental SNR show that the proposed algorithm has an output signal with a superior quality compared to the multi-band spectral subtraction algorithm and its modified version.

A vast amount of research is going on for design of robust speech recognition in to alleviate speech variability conditions. One of the variability aspects is the difference between telephony speech and direct speech (recorded in noise free conditions). In this paper by using a set of experiments, it is shown that LHCB parameters are superior to traditional MFCCs for speech recognition applications when they are used in a neural network based speech recognition system for both direct and telephony speech. Then by extraction of LHCBs from direct and telephony speech, and training of a MLP based speech recognition model, a direct and telephony speech recognition system is developed. Using a neural network inversion based on gradient descent method, the telephony speech feature vectors are modified toward to the direct speech feature vectors and by training a second network on modified telephony and direct speech feature vectors a 1.4% enhancement on speech recognition was achieved. Later, using general inversion method of neural networks both telephony and direct speech feature vectors are modified in a manner which mainly contains phonetic information and not other speech variations. Then by the training of the second neural network on this dataset, the system achieved 2.98% and 1.68% higher recognition rate for direct and telephony speech, respectively.

In this paper, several methods for optimization of a 5-elements Yagi antenna are proposed using genetic algorithm, genetic algorithm inspired by simulated annealing, genetic algorithm based on fuzzy decision making, and particle swarm method. High speed run time of SuperNEC software, it has been used for analyzing the presented methods. The use of genetic algorithm or genetic algorithm inspired by simulated annealing for antenna optimization in a specific frequency band, needs long run time. Besides, reduction of the number of population and the amount of repetition, causes decrease in optimization precision. So, an optimization system base on fuzzy decision making is proposed. In addition, the particle swarm method which has a good convergence rate and good performance has been proposed to obtain a better optimization. The comparison between the proposed optimization methods shows that the genetic based on fuzzy decision making and the particle swarm methods have the best performance and functionality and the least run time.

A popular method for creating required phase shifts for patches in a microstrip reflectarray uses open ended stubs attached to the patches. Uniform stubs are narrowband and hence limit the bandwidth of the array. In this paper, two novel methods for designing broadband reactive phase shifters are proposed based on the method of least squares. A structure The frequency response of s to be attached to theA popular method for creating required phase shifts for patches in a microstrip reflectarray uses open ended stubs attached to the patches. Uniform stubs are narrowband and hence limit the bandwidth of the array. In this paper, two novel methods for designing broadband reactive phase shifters are proposed based on the method of least squares. A structure consisting of transmission line sections, open-ended stubs, and resistors is considered for the phase shifter. Another one consists of transmission lines and open-ended. A least square error function is constructed for the required phase shift and the characteristic impedances, electrical lengths of transmission lines and resistances are obtained by its minimization. Then, the circuit is realized by microstrip lines to be attached to the microstrip patch. It is shown that the novel phase shifters create the required phase shifts in a broader bandwidth than those of uniform stub phase shifters. The frequency response of proposed microstrip line phase shifters are validated by the results obtained using AWR's Microwave Office software.

The Gate drivers in modern power converters which use the power IGBT must be provide several main operations such as electrical isolation, current amplifying, and protection against overcurrent and overvoltage conditions. This paper describes such a new circuit which is made using SMD devices suitable for driving the high and medium power IGBTs. This driver includes an isolated switching power supply, buffer circuits, and several protection circuits. It can operate by an input signal at TTL level and %50 duty cycle and is able to work up to 6A peak current.