Nuclear norm minimization for blind subspace identification (N2BSID)

Dexter Scobee, Lillian Ratliff, Roy Dong, Henrik Ohlsson, Michel Verhaegen, S. Shankar Sastry

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

5 Citations (Scopus)


In many practical applications of system identification, it is not feasible to measure both the inputs applied to the system as well as the output. In such situations, it is desirable to estimate both the inputs and the dynamics of the system simultaneously; this is known as the blind identification problem. In this paper, we provide a novel extension of subspace methods to the blind identification of multiple-input multiple-output linear systems. We assume that our inputs lie in a known subspace, and we are able to formulate the identification problem as rank constrained optimization, which admits a convex relaxation. We show the efficacy of this formulation with a numerical example.

Original languageEnglish
Title of host publicationProceedings of the 54th IEEE Conference on Decision and Control (CDC 2015)
EditorsYoshito Ohta, Mitsuji Sampei, Shigemasa Takai
Place of PublicationPiscataway, NJ, USA
ISBN (Electronic)978-1-4799-7886-1
Publication statusPublished - 2015
Event54th IEEE Conference on Decision and Control, CDC 2015 - Osaka, Japan
Duration: 15 Dec 201518 Dec 2015
Conference number: 54


Conference54th IEEE Conference on Decision and Control, CDC 2015
Abbreviated titleCDC 2015


  • Optimization
  • Mathematical model
  • Finite impulse response filters
  • Minimization
  • Maximum likelihood estimation
  • MIMO
  • Numerical models


Dive into the research topics of 'Nuclear norm minimization for blind subspace identification (N2BSID)'. Together they form a unique fingerprint.

Cite this