High-resolution compressive channel estimation for broadband wireless communication systems

Guan Gui, Wei Peng, Fumiyuki Adachi

    Research output: Contribution to journalArticlepeer-review

    16 Citations (Scopus)


    Broadband channel is often characterized by a sparse multipath channel where dominant multipath taps are widely separated in time, thereby resulting in a large delay spread. Accurate channel estimation can be done by sampling received signal with analog-to-digital converter (ADC) at Nyquist rate and then estimating all channel taps with high resolution. However, these Nyquist sampling-based methods have two main disadvantages: (i) demand of the high-speed ADC, which already exceeds the capability of current ADC, and (ii) low spectral efficiency. To solve these challenges, compressive channel estimation methods have been proposed. Unfortunately, those channel estimators are vulnerable to low resolution in low-speed ADC sampling systems. In this paper, we propose a high-resolution compressive channel estimation method, which is based on sampling by using multiple low-speed ADCs. Unlike the traditional methods on compressive channel estimation, our proposed method can approximately achieve the performance of lower bound. At the same time, the proposed method can reduce communication cost and improve spectral efficiency. Numerical simulations confirm our proposed method by using low-speed ADC sampling.

    Original languageEnglish
    Pages (from-to)2396-2407
    Number of pages12
    JournalInternational Journal of Communication Systems
    Issue number10
    Publication statusPublished - 2014 Oct 1


    • Compressive channel estimation
    • Compressive sensing
    • High-resolution channel
    • Sub-Nyquist rate sampling

    ASJC Scopus subject areas

    • Computer Networks and Communications
    • Electrical and Electronic Engineering


    Dive into the research topics of 'High-resolution compressive channel estimation for broadband wireless communication systems'. Together they form a unique fingerprint.

    Cite this