Summary: | 碩士 === 國立高雄第一科技大學 === 電子與資訊工程研究所 === 96 === Orthogonal Frequency Division Multiplexing technology (OFDM) is a high-spectrum efficiency of the multi-carrier transmission technology. However, the OFDM transmission technology has shortcoming of high peak power of the average ratio (PAPR). Therefore it requires larger scope of linearity for the system of high-power amplifier and Analog-to-Digital Converter. So will reduce system performance, increase
the difficulty of achieving hardware and system costs.
To reduce the PAPR in MCM systems, various methods have been proposed in the literature. They can be roughly divided into two categories. One is as such clipping、μ-law、companding techniques. In terms of complexity, these systems are more simples. They do not need to increase the complexity and time delay. The other is a system with no signal distortion but need to join redundant coding such as Selected Mapping(SLM), Partial Transmit Sequences(PTS)…etc. The system''s complexity requires
additional external information at the expense of bandwidth.
This paper presents a combination of pressure and expansion of high-power amplifier nonlinear uniform distribution of the conversion law companding (joint uniform companding). It will reduce PAPR and at the same time solve the problem of non-linear PA. The OFDM system implemented by the method of Inverse Fast Fourier Transform (IFFT) of symbol can be approximated as Gaussian distribution. The signals after the treatment of compression and expansion of solid-state power amplifier (SSPA) showed uniform distribution. However, non-linear role interval BER poor performance, adding convolution code and turbo code to improve role of non-linear range, is able to effectively improve BER performance. This method not only can reduce the PAPR and at the same time solve the problem of non-linear power amplifier. Simulations had been mode to these three methods Joint Uniform, Uniform and μ-law compression technology. The results showed that use of Joint Uniform technology to OFDM signal can effectively reduce the PAPR and BER performance is superior to other methods.
|