| 研究課題/領域番号 |
15K06072
|
|---|
| 研究機関 | 秋田県立大学 |
|---|
研究代表者 |
徐 粒 秋田県立大学, システム科学技術学部, 教授 (40252324)
|
|---|
| 研究分担者 |
桂 冠 秋田県立大学, システム科学技術学部, 特任助教 (80734904) [辞退]
松下 慎也 秋田県立大学, システム科学技術学部, 准教授 (20435449)
|
|---|
| 研究期間 (年度) |
2015-04-01 – 2019-03-31
|
| キーワード | compressive sensing / sparse representation / 2-D CS algorithm / sparse signal processing / adaptive filtering / block sparsity / MIMO comunication system / pulse Doppler radar |
|---|
| 研究実績の概要 |
The main results obtained in FY2016 are briefly summarized as follows:
A two-dimensional (2-D) compressive sensing based high-resolution identification algorithm,i.e., the 2-D zero-attractive projection algorithm has been proposed for high-resolution sparse targets detection of pulse Doppler radar (PDA) with low-speed ADC, which requires less computational and memory resources and can achieve comparable detection performance to a conventional PDA system with high-speed ADC.
To improve azimuth resolution and capture richer information for synthetic aperture radar (SAR), an adaptive wide-angle SAR imaging algorithm based on the Boltzmann machine model has been proposed, which can achieve better imaging performance than the existing algorithms.
In order to solve, more effectively and efficiently, sparse signal reconstruction problems based on compressive sensing for the so-called “block-structured” or “block sparse” signals with nonzero atoms occurring in clusters,two novel sparse adaptive reconstruction algorithms, i.e., the block zero attracting least mean square (BZA-LMS) algorithm and the block l_0-norm LMS (BL0-LMS) algorithm have been established. Experimental results demonstrate the validity and applicability of these proposed algorithms.
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
2: おおむね順調に進展している
理由
As stated above, significant progress has been made during the FY2016 almost as expected, and some of the obtained results have been published as journal/conference papers, which will provide a fundamental and sound base for further exploration of this research.
|
| 今後の研究の推進方策 |
In the next stage of this research project, the main tasks are to extend the obtained fundamental results to the large-scale MIMO case, to develop new corresponding techniques, particularly super-resolution identification strategies and algorithms for low-speed ADC based large-scale MIMO radar system, and to verify the proposed techniques by theoretical analysis as well as numerical simulation.
|
