Document Type : Original Article
Authors
1 Dept. of Electronics and Communication Engineering, Faculty of Engineeing, Zagazig University, Zagazig, Egypt
2 Communications and Electronics Department Faculty of Electronic Engineering,Manoufia University: Menouf, Egypt
3 Dept. of Communications, Faculty of Electronic Engineeing, Menofia University, Menouf, Egypt
Abstract
Highlights
In this paper, the image watermarking performances
after transmission over SC-FDMA system have been studied
for different types of equalizers. Simulations results show
that the quality of the received images over the IFDMA
system is better than that over the LFDMA system for both
ZF and MMSE equalizers. Also, the simulation results show
that the clarity of the received image over the IFDMA
system with ZF equalizer is better than that over the IFDMA
system with MMSE equalizer. This indicates that IFDMA
system with ZF equalizer has better performance.
Keywords
Main Subjects
By the quick growth of computer network in the latest years we can process digital contents and access to various digital media easily. Therefore media security became a highly important issue [1].Image watermarking is the process of embedding some useful information inside images for the purpose of security and authentication of the source. Different algorithms of digital image watermarking have been developed. In this paper, two different techniques of digital image watermarking have been used, DCT and DWT.
Single carrier frequency division multiple access (SC-FDMA) is a superior technology for uplink broadband wireless communications, as it does not have the problem of the large Peak-to-Average Power Ratio(PAPR) as opposed to orthogonal frequency division multiple Access (OFDMA)[2]. In image watermarking transmission system, It is highly important to study the images performance with watermarks embedded and extracted through SC-FDMA channel transmission. This paper has studied the performances of two different watermarking algorithms (DCT and DWT) through SC-FDMA channel[1]. The algorithm performance is measured by these important requirements of digital watermarking PSNR, Cr and SSIM .The PSNR in dB to measure the watermarked image imperceptibility, correlation of original mark image and recovering mark image to measure robustness of the mark image and the SSIM [3,11]. For this study a 512×512 gray scale "Lena" image is used as cover image, and a 50×20 copy right image is used as mark image .
2. Image Watermarking over The SC-FDMA System
Figure 1 shows the image watermarking over the SC-FDMA system architecture. The watermark embedding block is used for embedding the watermark into the cover image with the used algorithm either in DCT or DWT digital image watermarking technique. The image formatting converts the image to a binary form that has become suitable to be inserted , processed and transmitted over the SC-FDMA system. The SC-FDMA transmitter starts with an encoder then a modulation of the input signal using binary Quadrature amplitude modulation (QAM). The modulated symbols are grouped into blocks, each block containing N symbols and the FFT is performed. The signal after the FFT is expressed in Eq. (1) [7] :
where N is the input block size, {x(n): n=0,…, N-1}
1 8
represents the modulated symbols. The outputs after the FFT are mapped to M orthogonal subcarriers followed by the M-points IFFT. The resulting signal after the IFFT can be given as in Eq. (2)[7]: where : {:l=0, …, M-1} is the frequency-domain samples after the subcarriers mapping scheme. Cyclic Prefix (CP) which is a group of symbols is also inserted to give a guard time that is very important to forbid Inter-Block Interference (IBI) caused by multipath propagation. CP is a copy of the end part of the block. It is inserted at the start of each block. The transmitted data crossed through the channel [4,5]. At the receiver, the CP is removed, and after that the received signal is converted to the frequency domain by (FFT). The samples are crossed through the Frequency Domain Equalizer (FDE) then the de-mapping operation isolates the frequency-domain samples of each source signal. The Inverse Fast Fourier Transform (IFFT) at the receiver transforms equalized symbols back to the time domain. The demodulation process retrieves the original data, passed through the decoder. The image reconstruction converts the binary form to an image for recovering the original image. The watermark is extracted from the watermarked image by Watermark extraction [4,5].
3. Subcarriers Mapping Methods
The localized and distributed mapping methods are two schemes for mapping the subcarriers between users. The first one is ordinarily called localized scheme LFDMA , while the other one is ordinarily referred to as distributed FDMA transmission scheme. With the LFDMA transmission technique, each user's data is transported with consecutive subcarriers. In the distributed transmission scheme, the user's data is transported with distributed subcarriers. Because of The information symbol spreading through the entire signal band , the distributed FDMA technique is more strong to frequency-selective fading. Therefore, distributed scheme can realize more frequency diversity. For the transmission of localized through a frequency-selective fading channel, both the diversity of the multiuser and the frequency-selective diversity can be realized if each user has got subcarriers with appropriate transmission characteristics. The distributed FDMA technique that has equidistance among occupied subcarriers for the whole of the band is referred to as the interleaved FDMA scheme (IFDMA) [6,7]. Figure 2 illustrates An example for the two Methods for3 users, 12 subcarriers, and 4 subcarriers assigned per user. For this paper, the localized SC-FDMA system is referred to as LFDMA, the interleaved SC-FDMA system is called IFDMA [6,7].