本文選題：數(shù)字濾波器 + 結(jié)構(gòu)進(jìn)化�。� 參考：《河南大學(xué)》2015年碩士論文

【摘要】：目前,濾波器在電子系統(tǒng)中已經(jīng)成為一種不可或缺的組成部分,廣泛應(yīng)用于通信技術(shù)、圖像識(shí)別、軍事雷達(dá)、航空領(lǐng)域、醫(yī)療設(shè)備和語音等眾多領(lǐng)域。卓越的信號(hào)處理能力使其具有越來越大的市場應(yīng)用價(jià)值。模擬濾波器若要滿足較高的精度或者多個(gè)技術(shù)指標(biāo),不僅設(shè)計(jì)過程復(fù)雜、元器件個(gè)數(shù)多,結(jié)構(gòu)龐大,而且不一定能達(dá)到目標(biāo)要求。隨著電子計(jì)算機(jī)技術(shù)和大規(guī)模集成電路的不斷發(fā)展,數(shù)字濾波器可用計(jì)算機(jī)軟件或者大規(guī)模集成數(shù)字硬件實(shí)時(shí)實(shí)現(xiàn)。計(jì)算機(jī)技術(shù)的快速發(fā)展為數(shù)字濾波器的設(shè)計(jì)與實(shí)現(xiàn)創(chuàng)造了條件。但是,數(shù)字濾波器的功能要求不斷提高,數(shù)字濾波器的結(jié)構(gòu)越來越復(fù)雜,普通的設(shè)計(jì)方法很難滿足要求。因此,借助計(jì)算機(jī)設(shè)計(jì)數(shù)字濾波器起到越來越重要的作用。近些年,許多研究者提出各種各樣的算法應(yīng)用于數(shù)字濾波器的設(shè)計(jì)。包括遺傳算法(Genetic Algorithm,GA),模型退火算法(Simulate Anneal Arithmetic,SAA),禁忌搜索(Taboo Search,TS),蟻群最優(yōu)化算法(Ant Colony Optimization,ACO),神經(jīng)網(wǎng)絡(luò)算法(Neural Network Algorithm,NEA)和人工免疫算法(Immune Clonal Selection Algorithm,ICSA)等等。但是這些方法都是先根據(jù)目標(biāo)特性要求標(biāo)定傳輸函數(shù)的系數(shù),然后再考慮數(shù)字濾波器的結(jié)構(gòu),這種設(shè)計(jì)只保證了傳輸函數(shù)在系數(shù)標(biāo)定階段是最優(yōu)的,但是在整個(gè)數(shù)字濾波器的設(shè)計(jì)過程中可能不是最優(yōu)。本文主要提出一種基于演化算法的數(shù)字濾波器優(yōu)化設(shè)計(jì)方法,利用遺傳算法優(yōu)化數(shù)字濾波器結(jié)構(gòu),在得到最優(yōu)濾波器結(jié)構(gòu)之后,再利用差分算法和步長變化算法優(yōu)化濾波器系數(shù),最終得到數(shù)字濾波器最優(yōu)解。該方法能夠根據(jù)目標(biāo)特性要求直接設(shè)計(jì)數(shù)字濾波器結(jié)構(gòu),無需標(biāo)定傳輸函數(shù)系數(shù)。本文主要進(jìn)行以下幾個(gè)方面的研究工作:(1)綜述了數(shù)字濾波器設(shè)計(jì)方法,并著重闡述了數(shù)字濾波器的演化設(shè)計(jì)方法。(2)利用遺傳算法設(shè)計(jì)數(shù)字濾波器結(jié)構(gòu)。因?yàn)檫z傳算法的性能和效率主要由交叉率和突變率的取值決定,所以本文對(duì)這兩個(gè)參數(shù)進(jìn)行了深入的分析,并得出該算法最優(yōu)的輸入?yún)?shù)集。(3)利用差分算法和步長變化算法繼續(xù)優(yōu)化數(shù)字濾波器系數(shù)。與遺傳算法優(yōu)化結(jié)構(gòu)所得實(shí)驗(yàn)結(jié)果進(jìn)行比較,得出系數(shù)的優(yōu)化對(duì)阻帶最小衰減有明顯的提高。(4)全文最后對(duì)研究內(nèi)容進(jìn)行總結(jié)并做出展望。通過本文的實(shí)驗(yàn)結(jié)果可以證明,基于演化算法的數(shù)字濾波器優(yōu)化設(shè)計(jì)方法能夠獲得較好的實(shí)驗(yàn)效果。根據(jù)目標(biāo)特性直接設(shè)計(jì)濾波器結(jié)構(gòu),無需濾波器階數(shù)、傳遞函數(shù)等先驗(yàn)知識(shí),為濾波器的設(shè)計(jì)提供了一種有效的設(shè)計(jì)方法,同時(shí)該方法也有廣泛的適用性,可以解決其他類似的優(yōu)化問題。
[Abstract]:At present, filters have become an indispensable part of electronic systems, widely used in communications technology, image recognition, military radar, aviation, medical equipment and voice and many other fields. The outstanding signal processing ability makes it has more and more market application value. If analogue filter is to satisfy higher precision or more technical indexes, not only the design process is complicated, the number of components is large, the structure is huge, but also the target requirement is not always met. With the development of computer technology and large-scale integrated circuits, digital filters can be realized in real time by computer software or large-scale integrated digital hardware. The rapid development of computer technology creates conditions for the design and implementation of digital filters. However, the functional requirements of digital filters are increasing, the structure of digital filters is becoming more and more complex, and the common design methods are difficult to meet the requirements. Therefore, the design of digital filters by computer plays an increasingly important role. In recent years, many researchers have proposed a variety of algorithms for digital filter design. It includes genetic algorithm, simulated Anneal algorithm, Tabu search algorithm, Ant Colony Optimization algorithm, Neural Network algorithm and artificial immune algorithm, and so on. However, these methods first calibrate the coefficients of the transfer function according to the target characteristics and then consider the structure of the digital filter. This design only ensures that the transmission function is optimal in the calibration stage of the coefficients. But it may not be optimal in the whole design process of digital filter. In this paper, a digital filter optimization design method based on evolutionary algorithm is proposed. Genetic algorithm is used to optimize the digital filter structure, and the optimal filter structure is obtained. Then the difference algorithm and step size algorithm are used to optimize the filter coefficients, and the optimal solution of the digital filter is obtained. This method can design the digital filter structure directly according to the target characteristics, without calibrating the transfer function coefficient. In this paper, the following aspects of the research work: 1) the design method of digital filter is summarized, and the evolutionary design method of digital filter is described emphatically. 2) the structure of digital filter is designed by genetic algorithm. Because the performance and efficiency of genetic algorithm are mainly determined by the values of crossover rate and mutation rate, the two parameters are deeply analyzed in this paper. The optimal input parameter set of the algorithm is obtained. The difference algorithm and step size algorithm are used to continue to optimize the digital filter coefficients. Compared with the experimental results obtained by genetic algorithm optimization, it is concluded that the optimization of the coefficient can significantly improve the minimum attenuation of the stopband. 4) at the end of this paper, the research content is summarized and the prospect is made. The experimental results in this paper show that the optimal design method of digital filter based on evolutionary algorithm can obtain better experimental results. The filter structure is designed directly according to the target characteristics without prior knowledge such as filter order, transfer function and so on. It provides an effective design method for filter design, and it also has wide applicability. Other similar optimization problems can be solved.
【學(xué)位授予單位】：河南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類號(hào)】：TN713.7

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 蔡毓,劉貴忠,侯興松,王海龍;基于多分辨框架下實(shí)現(xiàn)數(shù)字濾波器[J];廣西科學(xué);2001年03期

2 蘭琴;;微機(jī)保護(hù)中數(shù)字濾波器講授方法探討[J];重慶電力高等�？茖W(xué)校學(xué)報(bào);2001年03期

3 楊益新,孫超;一種改進(jìn)的FIR數(shù)字濾波器自適應(yīng)設(shè)計(jì)方法[J];西北工業(yè)大學(xué)學(xué)報(bào);2002年04期

4 Ron Pearson;;以比例不變非線性數(shù)字濾波器有效凈化數(shù)據(jù)[J];電子設(shè)計(jì)技術(shù);2002年10期

5 陳佳敏 ,胡俊;FIR數(shù)字濾波器在數(shù)字化仿型數(shù)據(jù)處理中的應(yīng)用[J];精密制造與自動(dòng)化;2005年03期

6 馬潔;董金明;;IIR數(shù)字濾波器在定點(diǎn)DSP上的實(shí)現(xiàn)[J];聲學(xué)與電子工程;2006年02期

7 楊永昌;李晨輝;王凱;;FIR數(shù)字濾波器的設(shè)計(jì)方法[J];桂林航天工業(yè)高等�？茖W(xué)校學(xué)報(bào);2006年03期

8 雷能芳;;FIR數(shù)字濾波器的一種快速算法[J];現(xiàn)代電子技術(shù);2006年21期

9 馬寶山;朱義勝;;一種用于基因預(yù)測(cè)的FIR數(shù)字濾波器[J];電子學(xué)報(bào);2007年09期

10 楊志飛;;基于軟件Measurement Studio的數(shù)字濾波器的設(shè)計(jì)與實(shí)現(xiàn)[J];電氣傳動(dòng)自動(dòng)化;2007年05期

相關(guān)會(huì)議論文 前10條

1 劉煥淋;何方白;張?jiān)器?莊陵;;FIR數(shù)字濾波器的設(shè)計(jì)與實(shí)現(xiàn)[A];第九屆全國青年通信學(xué)術(shù)會(huì)議論文集[C];2004年

2 賈再一;賴曉平;陳仕兵;;設(shè)計(jì)2D-FIR數(shù)字濾波器的隨機(jī)抽樣遞推最小二乘法[A];第九屆全國信號(hào)處理學(xué)術(shù)年會(huì)（CCSP-99）論文集[C];1999年

3 黃建軍;謝維信;;一種直角可分的二維最佳數(shù)字濾波器[A];第十屆全國信號(hào)處理學(xué)術(shù)年會(huì)（CCSP-2001）論文集[C];2001年

4 張培珍;;FIR數(shù)字濾波器特性分析與組合窗設(shè)計(jì)法教學(xué)方案[A];2011高等職業(yè)教育電子信息類專業(yè)學(xué)術(shù)暨教學(xué)研討會(huì)論文集[C];2011年

5 夏驚雷;范建華;王庭昌;;高階FIR數(shù)字濾波器的優(yōu)化設(shè)計(jì)與實(shí)現(xiàn)[A];2005通信理論與技術(shù)新進(jìn)展——第十屆全國青年通信學(xué)術(shù)會(huì)議論文集[C];2005年

6 黃凱明;王兆明;林發(fā)炳;;連續(xù)平衡數(shù)字濾波器參數(shù)研究[A];2006中國控制與決策學(xué)術(shù)年會(huì)論文集[C];2006年

7 張濤;蔣靜坪;張項(xiàng)安;;應(yīng)用于電力系統(tǒng)保護(hù)的FIR數(shù)字濾波器的設(shè)計(jì)研究[A];2006中國電力系統(tǒng)保護(hù)與控制學(xué)術(shù)研討會(huì)論文集[C];2006年

8 陳s，

本文編號(hào)：2023450