海军工程大学学报
海軍工程大學學報
해군공정대학학보
JOURNAL OF NAVAL UNIVERSITY OF ENGINEERING
2014年
6期
78-83
,共6页
水下电磁探测%直接耦合干扰%可变遗忘因子%QRD-LS 算法
水下電磁探測%直接耦閤榦擾%可變遺忘因子%QRD-LS 算法
수하전자탐측%직접우합간우%가변유망인자%QRD-LS 산법
underwater electromagnetic detection%direct coupling interference%variable forgetting factor%QRD-LS algorithm
针对水下电磁探测系统对直接耦合干扰的补偿精度受限于硬件采样分辨率的问题,研究了基于自适应技术的直接耦合干扰两级补偿方法。采用高速 DSP 芯片实现直接耦合干扰补偿电路,完成直接耦合干扰的粗调补偿;提出了基于可变遗忘因子的 QRD-LS 算法,并采用基于该算法的自适应陷波器实现了直接耦合干扰的微调补偿。实验研究验证了两级补偿方法对直接耦合干扰的补偿性能。结果表明:基于可变遗忘因子的QRD-LS 算法稳定性好,收敛速度快;基于该算法的自适应两级补偿方法对直接耦合干扰的补偿精度达到10-4倍。
針對水下電磁探測繫統對直接耦閤榦擾的補償精度受限于硬件採樣分辨率的問題,研究瞭基于自適應技術的直接耦閤榦擾兩級補償方法。採用高速 DSP 芯片實現直接耦閤榦擾補償電路,完成直接耦閤榦擾的粗調補償;提齣瞭基于可變遺忘因子的 QRD-LS 算法,併採用基于該算法的自適應陷波器實現瞭直接耦閤榦擾的微調補償。實驗研究驗證瞭兩級補償方法對直接耦閤榦擾的補償性能。結果錶明:基于可變遺忘因子的QRD-LS 算法穩定性好,收斂速度快;基于該算法的自適應兩級補償方法對直接耦閤榦擾的補償精度達到10-4倍。
침대수하전자탐측계통대직접우합간우적보상정도수한우경건채양분변솔적문제,연구료기우자괄응기술적직접우합간우량급보상방법。채용고속 DSP 심편실현직접우합간우보상전로,완성직접우합간우적조조보상;제출료기우가변유망인자적 QRD-LS 산법,병채용기우해산법적자괄응함파기실현료직접우합간우적미조보상。실험연구험증료량급보상방법대직접우합간우적보상성능。결과표명:기우가변유망인자적QRD-LS 산법은정성호,수렴속도쾌;기우해산법적자괄응량급보상방법대직접우합간우적보상정도체도10-4배。
As direct coupling interference compensating precision is limited by the hardware sampling resolution of the underwater electromagnetic detection system,this paper analyzes the two-level com-pensating method of direct coupling interference based on the adaptive technology and uses a high-speed DSP chip to compete a direct coupling interference compensating circuit for the coarse compensa-tion.It also proposes a novel QRD-LS algorithm based on the variable forgetting factor.According to the algorithm,an adaptive notch filter is used to make the trimming compensation for the direct coupling interference.Through experiments,the two-level method is proved to be effective direct coupling interference compensation.The results show that the QRD-LS algorithm based on the varia-ble forgetting factor has advantages of high stability and fast convergence,so that the adaptive two-level compensating method based on this algorithm reaches a 10-4 times in precision for the direct cou-pling interfe-rence compensation.