A Multi-Box Behavioural Nonlinear Mixer Model


Ozgun A., NESİMOĞLU T., DEMİR Ş.

International Journal of Electronics, cilt.111, sa.8, ss.1384-1402, 2024 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 111 Sayı: 8
  • Basım Tarihi: 2024
  • Doi Numarası: 10.1080/00207217.2023.2233120
  • Dergi Adı: International Journal of Electronics
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Aerospace Database, Applied Science & Technology Source, Business Source Elite, Business Source Premier, Communication Abstracts, Compendex, Computer & Applied Sciences, INSPEC, Metadex, zbMATH, Civil Engineering Abstracts
  • Sayfa Sayıları: ss.1384-1402
  • Anahtar Kelimeler: Behavioural model, intermodulation, linearisation, mixer nonlinearity, Wiener-Hammerstein model
  • Orta Doğu Teknik Üniversitesi Adresli: Evet

Özet

This paper presents a new behavioural mixer model in a multi-box structure. In the proposed model, as well as amplitude and phase nonlinearity, the memory effect is also modelled. The multi-box structure is constructed in line with the Hammerstein model to characterise both fundamental signals, intermodulation distortion (IMD) and harmonic distortion (HD) for mixers in particular and all frequency translating devices in general. A testbed is constructed to measure both amplitude and phase characteristics. A good match between the measured and the calculated results is obtained using curve fitting and optimisation algorithms. In order to model the memory effect, a finite impulse response (FIR) filter is used. The measured IM3 imbalance is successfully modelled with this filter. For fundamental components, amplitude and phase errors are less than 0.1 dB and 1 degree, respectively. For IM3 components, amplitude and phase errors are less than 4 dB and 10 degrees, respectively. For the overall accuracy of the model, measurements are conducted with two-tone and modulated signals, and normalised mean square error (NMSE) values are calculated as −31.5 dB and −29.3 dB, respectively. The results are obtained for a wide range of input power span, including the compression region where nonlinearity is strongly pronounced. The proposed model is capable of accurately representing nonlinearity and memory effect for mixers and frequency translating devices in general and thus may be used for system-level modelling and linearisation applications.