Power Spectrum Estimation in Innovation Models by Nuclear Norm Optimization

AKÇAY H., TÜRKAY S.

14th IEEE International Conference on Control and Automation (ICCA), Alaska, Amerika Birleşik Devletleri, 12 - 15 Haziran 2018, ss.662-667 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/icca.2018.8444313
  • Basıldığı Şehir: Alaska
  • Basıldığı Ülke: Amerika Birleşik Devletleri
  • Sayfa Sayıları: ss.662-667
  • Anahtar Kelimeler: system identification, power spectrum, subspace method, regularization, innovation model, missing data, SYSTEM-IDENTIFICATION, RANK MINIMIZATION
  • Anadolu Üniversitesi Adresli: Evet

Özet

In this paper, identification of discrete-time power spectra of multi-input/multi-output models in innovation form from output-only time-domain measurements is studied. Two regularized nuclear norm minimization-based subspace algorithms are proposed. One of the algorithms is capable of handling missing data.