基于Laplace先驗的Bayes壓縮感知波達(dá)方向估計
doi: 10.11999/JEIT140937
基金項目:
山東省自然科學(xué)基金(ZR2014FQ003)和國家自然科學(xué)基金(61371181)資助課題
Direction-of-arrival Estimation Using Laplace Prior Based on Bayes Compressive Sensing
-
摘要: 基于多任務(wù)貝葉斯壓縮感知(BCS)理論，該文提出一種使用Laplace先驗的目標(biāo)到達(dá)角(DOA)估計算法。該算法利用陣元輸出為觀測值，將DOA估計轉(zhuǎn)化為Laplace先驗約束下的BCS求解稀疏信號問題，使用Laplace先驗獲得比傳統(tǒng)BCS更好的稀疏性。該算法不需要信源個數(shù)的先驗信息和進(jìn)行特征值分解，能夠適應(yīng)相干信源場景，仿真結(jié)果表明該算法具有比傳統(tǒng)BCS方法和經(jīng)典MUSIC算法更好的DOA估計性能。
-
關(guān)鍵詞:
- 目標(biāo)到達(dá)角估計 /
- 多任務(wù) /
- Bayes壓縮感知 /
- Laplace先驗
Abstract: Based on the multi-task Bayes Compressive Sensing (BCS), a Direction-Of-Arrival (DOA) estimation strategy using Laplace prior is proposed. The DOA estimation is formulated as the reconstruction of sparse signal constrained by the Laplace prior through the BCS framework. The outputs of array sensors are directly employed as the observations, and the exploiting of Laplace prior leads to better spare property than the conventional BCS method. The proposed method needs not the prior information of the number of sources, needs not the eigenvalue decomposition and can work in the coherent signal scenario. The numerical experiments show that the proposed method has the better performance than the conventional BCS and MUSIC algorithm on the DOA estimation. -
計量
- 文章訪問數(shù): 2156
- HTML全文瀏覽量: 127
- PDF下載量: 691
- 被引次數(shù): 0
下載:
