High-precision SAW bandpass filtering at 1747.5 MHz for LTE applications using wavelet transform techniques

TL;DR

This paper introduces a high-precision SAW bandpass filter for LTE applications using wavelet transforms to improve performance and reduce complexity.

Contribution

The novelty lies in integrating wavelet transforms with SAW filters to enhance spectral resolution and frequency stability at high carrier frequencies.

Findings

The filter achieves a passband width of 17.736 MHz with side-lobe attenuation below 140 dB.

Wavelet integration enables efficient spectral decomposition and reduces computational complexity.

Simulation results confirm strong frequency selectivity and stable center frequency across substrate variations.

Abstract

Recent developments in mobile communication technologies have intensified the demand for higher data rates, necessitating the use of elevated carrier frequencies and compact, high-performance, cost-effective radio frequency (RF) filters. Surface acoustic wave (SAW) filters offer notable advantages in mobile phone applications due to their low insertion loss, compact footprint, and scalable manufacturing. These attributes make them particularly well-suited to meet the growing need for affordable filtering solutions at increasingly higher operating frequencies. This paper presents the design and simulation of a wavelet transform-based surface acoustic wave (SAW) bandpass filter centered at 1747.5 MHz, optimized for GSM/LTE applications. The filter employs a multi-stage configuration enhanced by window functions such as Gaussian, Kaiser, Hanning, and Hamming to achieve precise null bandwidth control and effective side-lobe suppression. Wavelet transform integration streamlines spectral decomposition, enabling efficient frequency-domain analysis and reducing computational complexity. Finite element modeling is performed using COMSOL Multiphysics to simulate the electromechanical behavior of piezoelectric substrates, specifically quartz types. MATLAB is utilized for wavelet-domain signal processing and graphical analysis. Simulation results reveal a passband width of approximately 17.736 MHz, side-lobe attenuation below 140 dB, and stable center frequency alignment across substrate variations. FFT plots confirm strong frequency selectivity, while displacement profiles illustrate substrate-dependent acoustic wave propagation and energy confinement. The proposed wavelet-integrated SAW filter demonstrates high spectral resolution, robust frequency stability, and low-loss transmission, validating its potential for next-generation RF front-end systems.

Figures