Predicting Spectrum Occupancies Using a Non-Stationary Hidden Markov Model

Xianfu Chen, H Zhang, A B MacKenzie, Marja Matinmikko

    Research output: Contribution to journalArticleScientificpeer-review

    15 Citations (Scopus)

    Abstract

    One of the critical challenges for secondary use of licensed spectrum is the accurate modeling of primary users' (PUs') stochastic behaviors. However, the conventional hidden Markov models (HMMs) assume stationary state transition probability and fail to adequately describe PUs' dwell time distributions. In this letter, we propose a non-stationary hidden Markov model (NS-HMM), in which the time-varying property of PUs' behaviors is realized. A variant of the Baum-Welch algorithm is developed to estimate the parameters of a NS-HMM. Finally, the performance of the proposed model is evaluated through experiments using real spectrum measurement data. The results show that the NS-HMM outperforms existing HMM-based approaches.
    Original languageEnglish
    Pages (from-to)333-336
    Number of pages3
    JournalIEEE Wireless Communications Letters
    Volume3
    Issue number4
    DOIs
    Publication statusPublished - 2014
    MoE publication typeA1 Journal article-refereed

    Fingerprint

    Hidden Markov models
    Experiments

    Keywords

    • Cognitive radio
    • spectrum measurement
    • non-stationary hidden Markov model
    • Bayes' rule
    • spectrum occupancy
    • spectrum prediction

    Cite this

    Chen, Xianfu ; Zhang, H ; MacKenzie, A B ; Matinmikko, Marja. / Predicting Spectrum Occupancies Using a Non-Stationary Hidden Markov Model. In: IEEE Wireless Communications Letters. 2014 ; Vol. 3, No. 4. pp. 333-336.
    @article{7a6751f7e382480e8bfa134c4b5581f4,
    title = "Predicting Spectrum Occupancies Using a Non-Stationary Hidden Markov Model",
    abstract = "One of the critical challenges for secondary use of licensed spectrum is the accurate modeling of primary users' (PUs') stochastic behaviors. However, the conventional hidden Markov models (HMMs) assume stationary state transition probability and fail to adequately describe PUs' dwell time distributions. In this letter, we propose a non-stationary hidden Markov model (NS-HMM), in which the time-varying property of PUs' behaviors is realized. A variant of the Baum-Welch algorithm is developed to estimate the parameters of a NS-HMM. Finally, the performance of the proposed model is evaluated through experiments using real spectrum measurement data. The results show that the NS-HMM outperforms existing HMM-based approaches.",
    keywords = "Cognitive radio, spectrum measurement, non-stationary hidden Markov model, Bayes' rule, spectrum occupancy, spectrum prediction",
    author = "Xianfu Chen and H Zhang and MacKenzie, {A B} and Marja Matinmikko",
    note = "Project code: 82120",
    year = "2014",
    doi = "10.1109/LWC.2014.2315040",
    language = "English",
    volume = "3",
    pages = "333--336",
    journal = "IEEE Wireless Communications Letters",
    issn = "2162-2337",
    publisher = "IEEE Institute of Electrical and Electronic Engineers",
    number = "4",

    }

    Predicting Spectrum Occupancies Using a Non-Stationary Hidden Markov Model. / Chen, Xianfu; Zhang, H; MacKenzie, A B; Matinmikko, Marja.

    In: IEEE Wireless Communications Letters, Vol. 3, No. 4, 2014, p. 333-336.

    Research output: Contribution to journalArticleScientificpeer-review

    TY - JOUR

    T1 - Predicting Spectrum Occupancies Using a Non-Stationary Hidden Markov Model

    AU - Chen, Xianfu

    AU - Zhang, H

    AU - MacKenzie, A B

    AU - Matinmikko, Marja

    N1 - Project code: 82120

    PY - 2014

    Y1 - 2014

    N2 - One of the critical challenges for secondary use of licensed spectrum is the accurate modeling of primary users' (PUs') stochastic behaviors. However, the conventional hidden Markov models (HMMs) assume stationary state transition probability and fail to adequately describe PUs' dwell time distributions. In this letter, we propose a non-stationary hidden Markov model (NS-HMM), in which the time-varying property of PUs' behaviors is realized. A variant of the Baum-Welch algorithm is developed to estimate the parameters of a NS-HMM. Finally, the performance of the proposed model is evaluated through experiments using real spectrum measurement data. The results show that the NS-HMM outperforms existing HMM-based approaches.

    AB - One of the critical challenges for secondary use of licensed spectrum is the accurate modeling of primary users' (PUs') stochastic behaviors. However, the conventional hidden Markov models (HMMs) assume stationary state transition probability and fail to adequately describe PUs' dwell time distributions. In this letter, we propose a non-stationary hidden Markov model (NS-HMM), in which the time-varying property of PUs' behaviors is realized. A variant of the Baum-Welch algorithm is developed to estimate the parameters of a NS-HMM. Finally, the performance of the proposed model is evaluated through experiments using real spectrum measurement data. The results show that the NS-HMM outperforms existing HMM-based approaches.

    KW - Cognitive radio

    KW - spectrum measurement

    KW - non-stationary hidden Markov model

    KW - Bayes' rule

    KW - spectrum occupancy

    KW - spectrum prediction

    U2 - 10.1109/LWC.2014.2315040

    DO - 10.1109/LWC.2014.2315040

    M3 - Article

    VL - 3

    SP - 333

    EP - 336

    JO - IEEE Wireless Communications Letters

    JF - IEEE Wireless Communications Letters

    SN - 2162-2337

    IS - 4

    ER -