Abstract: Over the last decade, the development of IEEE 802.11a which referred to as Local Area Network (LAN) standard is rapidly advancing. This is due to the limitation and what Orthogonal Frequency Division Multiplexing (OFDM), which is used in the heart of LAN standard, susceptible to like carrier frequency offset and hence higher Bits Error Rate (BER). In this paper, the performance of IEEE 802.11a based on a sub-frequency band algorithm was investigated and tested over Additive White Gaussian Noise and Rician channels and under different Doppler frequency effects. The proposed system was investigated over the physical layer of IEEE 802.11a. Within the transmitter, the data vector of 64-bits is divided into three equal vectors of 21 bits each, with the last bit discarded. Each vector was then processed by fast Fourier transform block, with the output of each block processed by a proposed algorithm and then the total vector of length 64-bits converted to a time domain signal via the inverse fast Fourier transform. Within the receiver, the inverse procedure is used to recover the transmitted signals. The results show that the proposed algorithm outperforms the standard OFDM model in terms of Bit Error Rate (BER) versus Signal to Noise Ratio (SNR).

Abstract: Interferenceand noise has been a major area of concern in any communication system. It is unwanted signal which affect the transmitted signal during transmission, signal may be voice, image or video depending upon the system requirement. However for any type, the noise affects are always present. In some system interference or noise reduction has more importance like biometric system. Most of the biometric systems use images, where the quality is a prime factor. Hence a proper technique is required to improve the image quality by reduction of interference. Here we presented an algorithm for improving the quality of fingerprint images of biometric system. The technique adopted here for interference or noise reduction and improving the quality of image is on mean power calculation based algorithm, in which the mean power will be calculated for each frame, and certain frame having mean power above the threshold is accepted and enhanced by a specific value. Finally the results are compared for structure similarity using mean structural similarity index.

Abstract: Thispaper proposes a novel approach for handwritten digit recognition system. The present paper extracts digit image features based on distance measure and derives an algorithm to classify the digit images. The distance measure can be performed on the thinned image. Thinning is the one of the preprocessing technique in image processing.The present paper mainly concentrated on an extraction of features from digit image for effective recognition of the numeral. The main objective of the paper is exact features are extracted for good classification results. To find the effectiveness of the proposed method, tested on MNIST database, CENPARMI, CEDAR, and newly collected data. The proposed method is implemented on more than one lakh digit images and got good comparative recognition results.The percentage of the recognition is achieved about 97.32%.