Expected likelihood approach for determining constraints in covariance estimation

Bosung Kang, Vishal Monga, Muralidhar Rangaswamy, Yuri Abramovich

    Research output: Contribution to journalArticle

    2 Citations (Scopus)

    Abstract

    Recent covariance estimation methods for radar space-time adaptive processing exploit practical constraints such as the rank of clutter subspace and the condition number of disturbance covariance to estimate accurate covariance even when training is not generous. While rank and condition number are very effective constraints, often practical nonidealities make it difficult to know them precisely using physical models. Therefore, we propose a method to determine constraints in covariance estimation for radar space-time adaptive processing via an expected likelihood approach. We analyze three cases of constraints: 1) a rank constraint, 2) both rank and noise power constraints, and 3) a condition number constraint. In each case, we formulate precise constraint determination as an optimization problem. For each of the three cases, we derive new analytical results which allow for computationally efficient, practical ways of determining these constraints with formal proofs. Through experimental results from a simulation model and the KASSPER data set, we show that the estimator with optimal constraints obtained by the expected likelihood approach outperforms state-of-the-art alternatives.

    Original languageEnglish (US)
    Article number7812866
    Pages (from-to)2139-2156
    Number of pages18
    JournalIEEE Transactions on Aerospace and Electronic Systems
    Volume52
    Issue number5
    DOIs
    StatePublished - Oct 1 2016

    Fingerprint

    Space time adaptive processing
    Radar

    All Science Journal Classification (ASJC) codes

    • Aerospace Engineering
    • Electrical and Electronic Engineering

    Cite this

    Kang, Bosung ; Monga, Vishal ; Rangaswamy, Muralidhar ; Abramovich, Yuri. / Expected likelihood approach for determining constraints in covariance estimation. In: IEEE Transactions on Aerospace and Electronic Systems. 2016 ; Vol. 52, No. 5. pp. 2139-2156.
    @article{a38341f2418243d88a2da9c32cbed6b6,
    title = "Expected likelihood approach for determining constraints in covariance estimation",
    abstract = "Recent covariance estimation methods for radar space-time adaptive processing exploit practical constraints such as the rank of clutter subspace and the condition number of disturbance covariance to estimate accurate covariance even when training is not generous. While rank and condition number are very effective constraints, often practical nonidealities make it difficult to know them precisely using physical models. Therefore, we propose a method to determine constraints in covariance estimation for radar space-time adaptive processing via an expected likelihood approach. We analyze three cases of constraints: 1) a rank constraint, 2) both rank and noise power constraints, and 3) a condition number constraint. In each case, we formulate precise constraint determination as an optimization problem. For each of the three cases, we derive new analytical results which allow for computationally efficient, practical ways of determining these constraints with formal proofs. Through experimental results from a simulation model and the KASSPER data set, we show that the estimator with optimal constraints obtained by the expected likelihood approach outperforms state-of-the-art alternatives.",
    author = "Bosung Kang and Vishal Monga and Muralidhar Rangaswamy and Yuri Abramovich",
    year = "2016",
    month = "10",
    day = "1",
    doi = "10.1109/TAES.2016.150819",
    language = "English (US)",
    volume = "52",
    pages = "2139--2156",
    journal = "IEEE Transactions on Aerospace and Electronic Systems",
    issn = "0018-9251",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",
    number = "5",

    }

    Expected likelihood approach for determining constraints in covariance estimation. / Kang, Bosung; Monga, Vishal; Rangaswamy, Muralidhar; Abramovich, Yuri.

    In: IEEE Transactions on Aerospace and Electronic Systems, Vol. 52, No. 5, 7812866, 01.10.2016, p. 2139-2156.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Expected likelihood approach for determining constraints in covariance estimation

    AU - Kang, Bosung

    AU - Monga, Vishal

    AU - Rangaswamy, Muralidhar

    AU - Abramovich, Yuri

    PY - 2016/10/1

    Y1 - 2016/10/1

    N2 - Recent covariance estimation methods for radar space-time adaptive processing exploit practical constraints such as the rank of clutter subspace and the condition number of disturbance covariance to estimate accurate covariance even when training is not generous. While rank and condition number are very effective constraints, often practical nonidealities make it difficult to know them precisely using physical models. Therefore, we propose a method to determine constraints in covariance estimation for radar space-time adaptive processing via an expected likelihood approach. We analyze three cases of constraints: 1) a rank constraint, 2) both rank and noise power constraints, and 3) a condition number constraint. In each case, we formulate precise constraint determination as an optimization problem. For each of the three cases, we derive new analytical results which allow for computationally efficient, practical ways of determining these constraints with formal proofs. Through experimental results from a simulation model and the KASSPER data set, we show that the estimator with optimal constraints obtained by the expected likelihood approach outperforms state-of-the-art alternatives.

    AB - Recent covariance estimation methods for radar space-time adaptive processing exploit practical constraints such as the rank of clutter subspace and the condition number of disturbance covariance to estimate accurate covariance even when training is not generous. While rank and condition number are very effective constraints, often practical nonidealities make it difficult to know them precisely using physical models. Therefore, we propose a method to determine constraints in covariance estimation for radar space-time adaptive processing via an expected likelihood approach. We analyze three cases of constraints: 1) a rank constraint, 2) both rank and noise power constraints, and 3) a condition number constraint. In each case, we formulate precise constraint determination as an optimization problem. For each of the three cases, we derive new analytical results which allow for computationally efficient, practical ways of determining these constraints with formal proofs. Through experimental results from a simulation model and the KASSPER data set, we show that the estimator with optimal constraints obtained by the expected likelihood approach outperforms state-of-the-art alternatives.

    UR - http://www.scopus.com/inward/record.url?scp=85010299063&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85010299063&partnerID=8YFLogxK

    U2 - 10.1109/TAES.2016.150819

    DO - 10.1109/TAES.2016.150819

    M3 - Article

    AN - SCOPUS:85010299063

    VL - 52

    SP - 2139

    EP - 2156

    JO - IEEE Transactions on Aerospace and Electronic Systems

    JF - IEEE Transactions on Aerospace and Electronic Systems

    SN - 0018-9251

    IS - 5

    M1 - 7812866

    ER -