![](https://digilib.itb.ac.id/assets/icon/pdf.gif)
![](https://digilib.itb.ac.id/assets/icon/pdf.gif)
![](https://digilib.itb.ac.id/assets/icon/pdf.gif)
![](https://digilib.itb.ac.id/assets/icon/pdf.gif)
![](https://digilib.itb.ac.id/assets/icon/pdf.gif)
![](https://digilib.itb.ac.id/assets/icon/pdf.gif)
![](https://digilib.itb.ac.id/assets/icon/pdf.gif)
![](https://digilib.itb.ac.id/assets/icon/pdf.gif)
![](https://digilib.itb.ac.id/assets/icon/pdf.gif)
Channel modeling for High Speed Rail (HSR) is fundamental to design a reliable wireless communication system. A common practice is to conduct a tremendous amount of channel measurement data by deriving statistical method in order to achieve the appropriate channel models. For HSR communications, channel estimation in channel modeling enables high bandwidth physical layer transmission in state-of-the-art mobile communications. The recent development of HSR as an emerging of a high mobility transportation system introduces new challenges to create a reliable channel modeling communications due to its complexity and inaccuracy. Accurate and efficient channel models considering both large-scale and small-scale fading characteristics are crucial for the design, performance evaluation, and parameter optimization of HSR wireless communication systems. Fading in the HSR wireless communications can be categorized into large-scale and small-scale fading. The large-scale fading explains the main characteristics of the channel such as path loss and shadowing. Large-scale fading cases examine the relationship between path loss and separated distance between the transceiver in multiple environments such as in HSR scenarios. While small-scale fading cases go over to the relation between the Doppler spread and time coherence. Both factors are very meaningful in conducting a reliable channel modeling for HSR. The primary concern in this study is the investigation of path loss for HSR which contributes to large-scale fading. A new approach of conducting path loss prediction for HSR is introduced by proposing new empirical path loss models for HSR to be utilized in free space and urban/suburban propagation. The state-ofthe- art path loss models for HSR are validated using metaheuristic optimization techniques which have never been investigated in any literatures. The initial procedure in this study addresses a validation procedure of path loss measurement of HSR from Beijing Jiaotong University (BJTU) by utilizing Naïve Bayes classification method. The validation procedure is conducted priorly in this study because the proposed new empirical path loss models are underlied upon the measurement. path loss models for HSR are constructed. The first model is based from the free space propagation characteristic, assuming the transmitter and receiver are located in an otherwise empty environment. However, when this free space propagation is not sufficient enough to mitigate the maximum los e characteristics, another path loss model is also introduced. The two ray ground reflection path lossmodel for HSR is proposed as a more accurate model assuming the combination of direct signal path and reflected path from the earth. This model is meant to mitigate the losses in urban and suburban envinronment with a minimum antenna height of 50 meter. Both of these path loss models are validated by utilizing the metaheuristic optimization technique. The utilization of metaheuristic optimization technique model is performed on the proposed path models because of its applicability in obtaining the optimal path loss parameter values where physical phenomenon must be taken into account in a very dynamic system such as HSR. As both proposed path models comprise the corresponded los characteristics, the special MIMO antenna for HSR is specially designed in which its gain becomes the constant in the constructed path loss model. The final stage of this research reports the power link budgets for HSR to be utilized in low and high frequency spectrum. The dataset is once again optimized using the combination of Particle Swarm Optimization (PSO) and coeffiecient correlation in order to constraint the technical parameters in the power link budget so the obtained path loss values are in the range of observation. However, in a low frequency spectrum, correction factors are calculated by utilizing a point-to-point path loss model. Satisfying results of Received Signal Level (RSL) in power link budgets for both allocated frequency spectrum and propagations are achieved. Finally, the designed power link budgets are readily to be utilized by network designers to deploy the frequency planning and optimal base stations for HSR to its best performances.