正交分頻多工為基礎的IEEE 802.11a無線區域網路之性能研究

Abstract

無線區域網路IEEE 802.11a根據不同的通道品質提供6 Mb/s ~ 54 Mb/s的資料傳輸速率供使用者選擇﹔然而，在文獻中去探討各種資料傳輸速率的涵蓋範圍之研究並不多見。於是，本論文針對IEEE 802.11a 的標準，進行實體層(Physical layer)部分之模擬，得到各種資料速率在雷利(Rayleigh)以及萊斯(Rician)的衰減通道下，位元錯誤率(BER)對訊號雜訊比(SNR)的關係圖，再利用這些模擬結果，進行連線預算分析(Link budget Analysis)，求得有效的傳輸距離。結果發現，在室內NLOS環境下，傳送功率為23 dBm，IEEE 802.11a傳輸距離為7公尺(54 Mb/s)~62公尺(6 Mb/s)。另外我們亦發現加入非線性放大器之後，IEEE 802.11a的系統涵蓋範圍大幅縮小成2.8 公尺~ 39.5公尺。另外，考慮非線性放大器的影響，且改用soft-decision Viterbi Decoder，系統涵蓋範圍小幅增加至3.3 公尺~ 47.7公尺。 另外，利用TDD(Time Division Duplex)系統通道可逆性(Channel reciprocity)的優點，吾人提出一種”Adaptive Channel Pre-compensation”方法，專門應用在TDD模式下，以相位調變為主的QPSK-OFDM系統，比起傳統的接收端補償方式，我們提出的方法不需要複雜的通道估測演算法就能得到不錯的效能增益。另外，我們提出的技術也比現存的通道預補償方法優異，因為不會有error floor出現。在一個簡單的例子裡，用我們的方法可達到 的BER，然而用Witrisal [28]的方法，則BER會因非線性放大器的影響而限制在 。

The IEEE 802.11a Wireless Local Area Network (WLAN) provides several data rates from 6 Mb/s to 54 Mb/s according to different channel quality. However, few studies about the coverage area for various data rates have appeared in the literature. The aim of this thesis is to simulate the physical layer performance of the IEEE 802.11a system and to obtain the BER performance for various data rates in Rayleigh and Rician fading channel. By incorporating the physical layer simulation results into the link budget analysis, we then calculate the effective coverage area of the IEEE 802.11a system for different service rates. Based upon our simulation results, for transmission power of 23 dBm and under a non-light-of sight channel, the indoor coverage range for 54 Mb/s is 7 meters and that for 6 Mb/s is 62 meters. With the nonlinear effects of power amplifier, our simulation results show that the transmitter has a range of 2.8 meters and 39.5 meters for data rates 54 Mb/s and 6 Mb/s, respectively. Soft decision Viterbi decoder can improve the range area to 3.3 meters for 54 Mb/s and 47.7 meters for 6 Mb/s when taking the nonlinearity of power amplifier into account. Secondly, we have proposed an “Adaptive Channel Pre-compensation Method” for a TDD QPSK-OFDM system. By taking advantage of channel reciprocity of the TDD system, we avoid using complex channel estimation algorithms in the receiver. We demonstrate that the proposed pre-compensation method can still outperform the conventional post-compensation method by 1 dB of SNR. Furthermore, we show that the proposed pre-compensation method does not exhibit error floor phenomenon as other approaches. In a typical example, our approach can achieve a BER less than , while the conventional technique in [28] will be bounded by an error floor of due to the nonlinear effects of power amplifier.