Nisha Mary Joseph, Amrutha V Nair
Abstract: Orthogonal Frequency Division Multiplexing (OFDM) is a multicarrier modulation technique. In this scheme, there are subcarriers that are orthogonal to each other and closely spaced.These subcarriers will be converted to parallel streams. There are channel impairments like Clock Frequency Offset (CFO), Chromatic Dispersion (CD), and polarization Mode Dispersion (PMD) that affect the performance of the OFDM system. This can cause tilting of OFDM signal i.e, Common Phase Error (CPE) on the OFDM symbol. This CPE is compared with the skewed image in image processing domain and algorithm like Minimum Bounding Box (MBB) and Best fit Bounding Box (BBB) can be used for its correction. Traditional methods for CPE correction is also simulated and algorithms implemented are Least Mean Square (LMS) and Constant Modulus Algorithm (CMA). Traditional methods for channel equalization like Least Mean Square (LMS) and Constant Modulus Algorithm (CMA) are simulated in this work. LMS is a pilot aided method and CMA is a blind equalization method. The convergence rate of LMS is greater than CMA but the spectral efficiency is reduced due to the requirement of training sequence. CPE can be compared with the skewed image in image processing and algorithms like Minimum Bounding Box (MBB) and Best-Fit Bounding Box can be used for its correction. MBB and BBB algorithms are implemented and its performances are compared.The performance comparison is based on Bit Error Rate and BBB is found to have lower BER than MBB. BBB algorithm has less computational complexity and better spectral efficiency than MBB.
Keywords: Skewed image, CPE, Image processing, MBB, LMS, CMA