軟體基站之可適應性等化器

Abstract

基於現今對於網路傳輸效率的需求，已投入眾多的人力資源於開發 5G 的系統規格上，而在 5G 逐見雛型 的規格下，soft-PHY 成為未來系統架構的主流。此論文提出基於 virtual pilot的通道估測演算法並使用soft-PHY的模擬平台上，由於LTE現行使用於行動通訊上所以模擬時需要遭受時變以及多路徑衰退的通道模型下進行分析，而其平台原來的方法在移動速動提升時估測能力不佳甚至會造成連線上的中斷，因此提出使用virtual pilot的概念在通道估測演算法中而其運算複雜度不會增加太高並能夠增加估測能力，本篇論文著重於在位元錯誤率達到千分之一時和perfect known channel estimation的SNR loss能夠不多於2dB。

Based on demand for network transmission, there has invested lots of resources to develop the specification of 5G. Due to the limitation of hardware, Soft PHY becomes more and more popular to implement PHY layer. In this paper, we peopose a channel estimation based on virtual pilot and simulate on soft-PHY platform. Duo to LTE is used for mobile, we need to simulate and analysis environment under doppler effect and multipath effect. The soft-PHY channel estimation is weak at higher speed. We propose a channel estimation based on virtual pilot, the method is not higher computational complexity and enhance the perofrmance of estimation. This paper focuses on when bits error rate reach 〖10〗^(-3) the SNR loss is not more than 2 dB.