標題: 以DSP 為基礎的IS-2000 接收機架構
A DSP-Based IS2000 Receiver Structure
作者: 許芳聖
Fang-Sheng Hsu
蘇育德
Yu T. Su
電信工程研究所
關鍵字: DSP;IS-2000;犁耙式接收機;DSP;IS-2000;RAKE
公開日期: 2001
摘要: 本篇論文著重於IS-2000 上行鏈路接收機架構的DSP 實現,我們不僅使用電腦程式來模擬IS-2000 在上行鏈路下接收機性能的表現,亦使用德州儀器公司的C6201 DSP 晶片來達成硬體實現。IS-2000 傳送端使用混合式相位鍵移 (HPSK) 來降低鋒均比(peak-to-average ratio)並增進前端功率放大器的效率。IS-2000 系統同時也使用多重碼分碼多工 (Multi-Code CDMA) 機制以符合IMT-2000 (International Mobile Telecommunications Systems-2000) 高速資料傳輸的要求。在傳送過程中,我們使用 SRRC (Square-Root Raise-Cosine) 波形來傳送資料,因此導入了符元間干擾(Inter-symbol interference)的效應。 在IS-2000 接收機部分,我們使用最大比率結合(Maximal Ratio Combining)犁耙式接收機(RAKE receiver) 來減輕因為通道衰退效應(fading channel effect)所導致的系統性能降低。我們使用二階的Butterworth 低通濾波器來估計因為通道效應所引起的相位失真,並作為最大比率結合犁耙式接收機的加權值。 為了實現整個接收機架構,我們使用德州儀器公司的C6201 定點(fixed-point)DSP 晶片來實現IS-2000 接收機架構。最後我們並比較電腦軟體模擬的結果與DSP 硬體實現的性能表現,來證實兩者非常接近。
This thesis addresses issues on implementing the receiver of IS-2000 reverse link. We estimate the performance of an IS-2000 reverse link receiver by both computer simulations and DSP-based realization. The transmitter of the IS-2000 system uses Hybrid Phase Shift Keying (HPSK) to reduce the peak-to-average e?ect and enhance the e?ciency of the front-end power ampli?er. IS-2000 also uses a multi-code code division multiple access (Multi-Code CDMA) scheme. This scheme makes high data rate transmission required for International Mobile Telecommunications Systems-2000 (IMT-2000) possible. Square-Root Raise-Cosine (SRRC) chip pulse is used to produce compact signal spectrum but at the same time introduces inter-symbol interference (ISI) in a multipath fading environment. The RAKE receiver we investigate uses the maximum ratio combining (MRC) to mitigate the performance degradation caused by the channel fading e?ect. A crucial part of an MRC RAKE receiver is the channel estimator. We use a pair of low-pass Butterworth ?lters at each ?nger to estimate the associated complex channel distortion. To realize our design, we employ TI's ?xed-point DSPs (TMS320C6201). Detailed implementation issues such as bu?er management, scheduling and C code optimization are discussed and performance examples are given. We compare the performance predicted by computer simulation and that by hardware (DSP) and ?nd both yield almost the same result.
URI: http://140.113.39.130/cdrfb3/record/nctu/#NT900435028
http://hdl.handle.net/11536/68903
顯示於類別:畢業論文