This study was conducted to evaluate the acoustic performance of low-height noise barriers installed adjacent to rails. First, the acoustic performance of the low-height noise barriers was measured in an anechoic chamber using scaled-down models, and an easy-to-use approximation formula was suggested to predict insertion loss instead of using the boundary element method. Then, this prediction program was validated through a comparison of the actual measurement results in the anechoic chamber and the prediction results. Also, acoustic characteristics and methods to improve the acoustic performance of sound-absorbing systems equipped with double-layered polyester sound-absorbing materials were investigated. To improve the acoustic performance of the low-height noise barriers, a simulation program was used to analyze performance according to changes in the shape of the top of the noise barriers and the absorption coefficient of the sound-absorbing materials used. Finally, using the prediction program mentioned, an approximation formula for insertion loss was suggested for the low-height noise absorption barriers.

· According to the measurement results found using a scaled-down model in an anechoic chamber without a train model, the insertion loss of low-height noise barriers with an ‘I’ and ‘ㄱ’ shape was similar. The insertion loss due to the installation of sound-absorbing materials was also analyzed and found to be insignificant.

· According to the measurement results found in the experiments with a train model, the insertion loss of ‘I’ shaped noise barriers resulted in superior acoustic performance over ‘ㄱ’ shaped barriers due to reflected noise from the top of the barriers. So, when ‘ㄱ’ shaped noise barriers are designed to be installed, sound-absorbing materials should be added to both the inside and top.

· Other analytical approaches instead of the simple empirical formula used, such as the boundary element method, would be required to calculate the actual insertion loss of noise barriers because this changes due to reflected noise in cases with train model.

· According to the measurement results from the scaled-down model in an anechoic chamber, the noise level at a passenger location increased by 1 to 4 dB(A) in cases with and without sound-absorbing materials, respectively. So, if absorptive noise barriers are installed, the noise level increase at passenger locations is expected to be insignificant.

· After a simulation using a 2D boundary element method, a relatively accurate prediction of insertion loss was possible, regardless of the presence of sound-absorbing materials.

· Methods to improve the acoustic performance of a sound-absorbing system equipped with double-layered polyester sound-absorbing materials were investigated. Numerical models were set up and the results obtained from the models were compared with the actual measurement data. Strategies to improve acoustic performance based on different configurations were considered. To improve performance in limited spaces, a configuration that combines polyester with a density over 100 kg/m3 at a thickness of 10 mm in front of normal density polyester with a density of 24 to 64 kg/m3 has been proposed.

· Acoustic performance was analyzed by using a simulation program while changing the shape of the top of the noise barriers and the absorption coefficient of the sound-absorbing materials. As results of the simulations, the insertion loss of ‘T’ and ‘U’ shaped noise barriers was improved by 0.1 dB and 1.8 dB in comparison with that of ‘ㄱ’ shaped noise barriers, respectively. Also, the insertion loss of absorptive soundproofing panel with infinite impedance boundary conditions was improved by 2.8 dB when compared with that of reflective soundproofing panels.

· The easy-to-use approximation formula used to evaluate insertion loss relied on measurement results from scaled-down models in an anechoic chamber, in contrast to the boundary element method. This approximation formula for insertion loss was suggested for designing the low-height noise absorption barriers.