Components of wireless communication systems often do not operate completely within their linear operating range. This applies in particular to power amplifiers in the sub-THz and THz frequency range. While the best possible linear approximation of the transfer characteristic can still be found in the weakly nonlinear case, the resulting nonlinear induced distortion depends on the input signal. Therefore, a behavioral characterization is performed that describes the transmission characteristics for a specific operating mode. Here, we pursue two approaches. The first one is the EVM (error vector magnitude) approach, which reveals the nonlinear induced deviation of the modulation symbols from a reference in terms of magnitude and phase (AM-AM and AM-PM distortion). The other approach is more statistical. It is reminiscent of the NPR (noise power ratio) test, which can be implemented quite effectively for typical communication signals such as OFDM and SC-FDM. The key issue is that the test signal should reproduce the application signal with regard to the relevant statistical characteristics.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Behavioral Characterization of Nonlinear Distortion

  • Dominik Wrana,
  • Simon Haussmann,
  • Jonas Gedschold,
  • Reiner Thomä,
  • Ingmar Kallfass

摘要

Components of wireless communication systems often do not operate completely within their linear operating range. This applies in particular to power amplifiers in the sub-THz and THz frequency range. While the best possible linear approximation of the transfer characteristic can still be found in the weakly nonlinear case, the resulting nonlinear induced distortion depends on the input signal. Therefore, a behavioral characterization is performed that describes the transmission characteristics for a specific operating mode. Here, we pursue two approaches. The first one is the EVM (error vector magnitude) approach, which reveals the nonlinear induced deviation of the modulation symbols from a reference in terms of magnitude and phase (AM-AM and AM-PM distortion). The other approach is more statistical. It is reminiscent of the NPR (noise power ratio) test, which can be implemented quite effectively for typical communication signals such as OFDM and SC-FDM. The key issue is that the test signal should reproduce the application signal with regard to the relevant statistical characteristics.