Return
Enhancing OFDM system performance through discrete contourlet transform
M
Mohamed H. M. NermaA
Adam Mohamed Ahmed AbdoA
Aamer Mohamed Huroon Mohamed DOI:10.23919/jcn.2025.000107.png)
Abstract
En 中文
The orthogonal frequency-division multiplexing (OFDM) system is foundational for current fifth-generation (5G) wireless communication networks and a strong candidate for future sixth-generation (6G) advancements. However, conventional fast Fourier transform (FFT)-based OFDM suffers from critical limitations, including a high peak-to-average power ratio (PAPR), spectral inefficiency due to slow out-of-band emission (OOBE) decay, and sensitivity to synchronization errors. This paper introduces a novel multicarrier waveform, contourlet transform-OFDM (ConT-OFDM), which replaces the conventional FFT/IFFT pair with the inverse discrete contourlet transform (IDConT) and discrete contourlet transform (DConT) of the proposed method successfully adapts the DConT's multi-resolution and directional filter bank structure, which was originally designed for 2D image processing, to 1D communication signals to enhance time-frequency localization and spectral containment. A rigorous comparative analysis was conducted across key performance metrics, focusing on the highly demanding Rayleigh fading channel. The results, supported by theoretical analysis, demonstrate that DConT-OFDM offers: superior performance in terms of bit error rate (BER) and symbol error rate (SER) due to its inherent interference suppression capabilities, a significant reduction in PAPR, and enhanced power spectral density (PSD) characteristics. Crucially, by substituting FFT's sinc-based windowing with the sophisticated, compactly supported filters of the DConT, the system achieves these gains with a substantially lower linear computational complexity of O(N) compared to the O(N log N) complexity of FFT-OFDM, thus positioning DConT-OFDM as an efficient and robust filter-bank multicarrier (FBMC) alternative for future broadband wireless systems.
Keywords:
BER
computational complexity
contourlet transform
OFDM
PAPR
PSD
SER
Journal
J
IF:
3.2
Papers:
47
Citations:
0

