自適應(yīng)濾波器原理（第四版）

第1章 隨機(jī)過程與模型                  
                   
 1.1 離散時(shí)間隨機(jī)過程的部分特性                  
 1.2 平均各態(tài)歷經(jīng)定理                  
 1.3 相關(guān)矩陣                  
 1.4 正弦波加噪聲的相關(guān)矩陣                  
 1.5 隨機(jī)模型                  
 1.6 Wold分解                  
 1.7 回歸過程的漸近平穩(wěn)                  
 1.8 尤爾一沃克方程                  
 1.9 計(jì)算機(jī)實(shí)驗(yàn)：二階自回歸過程                  
 1.10 選擇模型的階數(shù)                  
 1.11 復(fù)值高斯過程                  
 1.12 功率譜密度                  
 1.13 功率譜密度的性質(zhì)                  
 1.14 平穩(wěn)過程通過線性濾波器傳輸                  
 1.15 平穩(wěn)過程的Cramer譜表示                  
 1.16 功率譜估計(jì)                  
 1.17 隨機(jī)過程的其他統(tǒng)計(jì)特征                  
 1.18 多譜                  
 1.19 譜相關(guān)密度                  
 1.20 本章小結(jié)                  
 1.21 習(xí)題                  
                   
 第2章 維納濾波器                  
                   
 2.1 線性最優(yōu)濾波：?jiǎn)栴}綜述                  
 2.2 正交性原理                  
 2.3 最小均方誤差                  
 2.4 維納一霍夫方程                  
 2.5 誤差性能曲面                  
 2.6 多重線性回歸模型                  
 2.7 示例                  
 2.8 線性約束最小方差濾波器                  
 2.9 廣義旁瓣消除器                  
 2.10 本章小結(jié)                  
 2.11 習(xí)題                  
                   
 第3章 線性預(yù)測(cè)                  
                   
 3.1 前向線性預(yù)測(cè)                  
 3.2 后向線性預(yù)測(cè)                  
 3.3 列文森一杜賓算法                  
 3.4 預(yù)測(cè)誤差濾波器的性質(zhì)                  
 3.5 舒爾一科思測(cè)試                  
 3.6 平穩(wěn)隨機(jī)過程的自回歸建模                  
 3.7 Cholesky分解                  
 3.8 格型預(yù)測(cè)器                  
 3.9 全極點(diǎn). 全通格型濾波器                  
 3.10 聯(lián)合過程估計(jì)                  
 3.11 語音預(yù)測(cè)建模                  
 3.12 本章小結(jié)                  
 3.13 習(xí)題                  
                   
 第4章 最速下降算法                  
                   
 4.1 最速下降算法的基本思想                  
 4.2 最速下降算法應(yīng)用于維納濾波器                  
 4.3 最速下降算法的穩(wěn)定性                  
 4.4 示例                  
 4.5 作為確定性搜索法的最速下降算法                  
 4.6 最速下降算法的優(yōu)點(diǎn)與局限性                  
 4.7 本章小結(jié)                  
 4.8 習(xí)題                  
                   
 第5章 最小均方自適應(yīng)濾波器                  
                   
 5.1 最小均方算法的結(jié)構(gòu)與運(yùn)算概述                  
 5.2 最小均方自適應(yīng)算法                  
 5.3 應(yīng)用示例                  
 5.4 統(tǒng)計(jì)LMS理論                  
 5.5 LMS算法與最速下降算法的比較                  
 5.6 自適應(yīng)預(yù)測(cè)的計(jì)算機(jī)實(shí)驗(yàn)                  
 5.7 自適應(yīng)均衡的計(jì)算機(jī)實(shí)驗(yàn)                  
 5.8 最小方差無失真響應(yīng)波束形成器的計(jì)算機(jī)實(shí)驗(yàn)                  
 5.9 非白噪聲輸入時(shí)LMS算法收斂的方向性                  
 5.10 LMS濾波器的魯棒性：H準(zhǔn)則                  
 5.11 不同情況下步長(zhǎng)參數(shù)的上界                  
 5.12 確定性輸入時(shí)的轉(zhuǎn)移函數(shù)方法                  
 5.13 本章小結(jié)                  
 5.14 習(xí)題                  
                   
 第6章 歸一化最小均方自適應(yīng)濾波器                  
                   
 6.1 歸一化LMS濾波器作為約束最優(yōu)化問題的解                  
 6.2歸一化LMS濾波器的穩(wěn)定性                  
 6.3 回聲消除中的步長(zhǎng)控制                  
 6.4 實(shí)數(shù)據(jù)時(shí)收斂過程的幾何考慮                  
 6.5 仿射投影濾波器                  
 6.6 本章小結(jié)                  
 6.7 習(xí)題                  
                   
 第7章 頻域和子帶自適應(yīng)濾波器                  
                   
 7.1 塊自適應(yīng)濾波器                  
 7.2 快速塊LMS算法                  
 7.3 無約束頻域自適應(yīng)濾波器                  
 7.4 自正交化自適應(yīng)濾波器                  
 7.5 自適應(yīng)均衡的計(jì)算機(jī)實(shí)驗(yàn)                  
 7.6 子帶自適應(yīng)濾波器                  
 7.7 自適應(yīng)濾波算法的分類                  
 7.8 本章小結(jié)                  
 7.9 習(xí)題                  
                   
 第8章 最小二乘法                  
                   
 8.1 線性最小二乘估計(jì)問題                  
 8.2 數(shù)據(jù)開窗                  
 8.3 正交性原理的進(jìn)一步討論                  
 8.4 誤差的最小平方和                  
 8.5 正則方程和線性最小二乘濾波器                  
 8.6 時(shí)間平均相關(guān)矩陣                  
 8.7 根據(jù)數(shù)據(jù)矩陣構(gòu)建正則方程                  
 8.8 最小二乘估計(jì)的特性                  
 8.9 MVDR的譜估計(jì)                  
 8.10 WVDR波束形成的正則化                  
 8.11 奇異值分解                  
 8 12 偽逆                  
 8.13 奇異值和奇異向量的解釋                  
 8.14 線性最小H乘問題的最小范數(shù)解                  
 8.15 歸一化最小均方算法看做欠定最小二乘小范數(shù)解                  
 8.16 本章小結(jié)                  
 8.17 習(xí)題                  
                   
 第9章 遞歸最小二來自適應(yīng)濾波器                  
                   
 9.1 預(yù)備知識(shí)                  
 9.2 矩陣求逆引理                  
 9.3 指數(shù)加權(quán)遞歸最小二乘算法                  
 9.4 正則化參數(shù)的選擇                  
 9.5 誤差平方加權(quán)和的更新遞歸                  
 9.6 示例：?jiǎn)蝹€(gè)權(quán)值目適應(yīng)噪聲消除器                  
 9.7 RLS算法的收斂性分析                  
 9.8 自適應(yīng)均衡的計(jì)算機(jī)實(shí)驗(yàn)                  
 9.9 RLS濾波器的魯棒性                  
 9.10 本章小結(jié)                  
 9.11 習(xí)題                  
                   
 第10章 卡爾曼濾波器                  
                   
 10.1 標(biāo)量隨機(jī)變量的遞歸最小均方估計(jì)                  
 10.2 卡爾曼濾波問題                  
 10.3 新息過程                  
 10.4 應(yīng)用新息過程進(jìn)行狀態(tài)估計(jì)                  
 10.5 濾波                  
 10.6 初始條件                  
 10.7 卡爾曼濾波器總結(jié)                  
 10.8 卡爾曼濾波器作為RLS濾波器的統(tǒng)一基礎(chǔ)                  
 10.9 卡爾曼濾波器變形                  
 10.10 廣義卡爾曼濾波器                  
 10.11 本章小結(jié)                  
 10.12 習(xí)題                  
                   
 第11章 平方根自適應(yīng)濾波器                  
                   
 11.1 平方根卡爾曼濾波器                  
 11.2 在卡爾曼濾波器基礎(chǔ)上構(gòu)建平方報(bào)自適應(yīng)濾波器                  
 11.3 QR－RLS算法                  
 11.4 自適應(yīng)波束形成                  
 11.5 逆QR-RLS算法                  
 11.6 本章小結(jié)                  
 11.7 習(xí)題                  
                   
 第12章 階遞歸自適應(yīng)濾波器                  
                   
 12.1 梯度自適應(yīng)格型濾波器                  
 12.2 采用最小二乘估計(jì)的階遞歸自適應(yīng)濾彼器：概述                  
 12.3 自適應(yīng)前向線性預(yù)測(cè)                  
 12.4 自適應(yīng)后向線性預(yù)測(cè)                  
 12.5 變換因子                  
 12.6 最小二乘格型預(yù)測(cè)器                  
 12.7 角度歸一化估計(jì)誤差                  
 12.8 格型濾波的一階狀態(tài)空間模型                  
 12.9 基于QR分解的最小二乘格型濾波器                  
 12.10 QRD－LSL濾波器基本特性                  
 12.11 自適應(yīng)均衡的計(jì)算機(jī)試驗(yàn)                  
 12.12 采用后驗(yàn)估計(jì)誤差的遞歸最小二乘格型濾波器                  
 12.13 采用帶誤差反饋先驗(yàn)估計(jì)誤差的遞歸LSL濾波器                  
 12.14 遞歸LSL濾波器和RLS濾波器的關(guān)系                  
 12.15 本章小結(jié)                  
 12.16 習(xí)題                  
                   
 第13章 有限精度效應(yīng)                  
                   
 13.1 量化誤差                  
 13.2 最小均方算法                  
 13.3 遞歸最小二乘算法                  
 13.4 平方根自適應(yīng)濾波器                  
 13.5 階遞歸自適應(yīng)濾波器                  
 13.6 快速橫向?yàn)V波器                  
 13.7 本章小結(jié)                  
 13.8 習(xí)題                  
                   
 第14章 時(shí)變系統(tǒng)的跟蹤                  
                   
 14.1 系統(tǒng)辨識(shí)用馬爾可夫模型                  
 14.2 非平穩(wěn)度                  
 14.3 跟蹤性能評(píng)價(jià)準(zhǔn)則                  
 14.4 LMS算法的跟蹤性能                  
 14.5 RLS算法的跟蹤性能                  
 14.6 LMS算法和RLS算法的跟蹤性能比較                  
 14.7 如何改進(jìn)RLS算法的跟蹤性能                  
 14.8 系統(tǒng)辨識(shí)的計(jì)算機(jī)實(shí)驗(yàn)                  
 14.9 自適應(yīng)常數(shù)的自動(dòng)調(diào)節(jié)                  
 14.10 本章小結(jié)                  
 14.11 習(xí)題                  
                   
 第15章 無限脈沖響應(yīng)自適應(yīng)濾波器                  
                   
 15.1 IIR自適應(yīng)濾波器：輸出誤差法                  
 15.2 IIR自適應(yīng)濾波器：方程誤差法                  
 15.3 某些實(shí)際考慮                  
 15.4 Laguerre橫向?yàn)V波器                  
 15.5 自適應(yīng)Laguerre格型濾波器                  
 15.6 本章小結(jié)                  
 15.7 習(xí)題                  
                   
 第16章 盲反卷積                  
                   
 16.1 盲反卷積問題概述                  
 16.2 利用循環(huán)平穩(wěn)統(tǒng)計(jì)量的信道辨識(shí)                  
 16.3 分?jǐn)?shù)間隔盲辨識(shí)用子空間分解                  
 16.4 Bussgang盲均衡算法                  
 16.5 將Bussgang算法推廣到復(fù)基帶信道                  
 16.6 Bussgang算法的特例                  
 16.7 分?jǐn)?shù)間隔Bussgang均衡器                  
 16.8 本章小結(jié)                  
 16.9 習(xí)題                  
                   
 第17章 反向傳播學(xué)習(xí)                  
                   
 17.1 S形神經(jīng)元模型                  
 17.2 多層感知器                  
 17.3 復(fù)反向傳播算法                  
 17.4 萬能逼近定理                  
 17.5 網(wǎng)絡(luò)復(fù)雜性                  
 17.6 瞬態(tài)處理：如何考慮時(shí)間                  
 17.7 反向傳播學(xué)習(xí)的優(yōu)點(diǎn)和局限性                  
 17.8 本章小結(jié)                  
 17.9 習(xí)題                  
 后記                  
                   
 附錄 A 復(fù)變量                  
 附錄 B 對(duì)向量微分                  
 附錄 C 拉格朗日乘子法                  
 附錄 D 估計(jì)理論                  
 附錄 E 特征分析                  
 附錄 F 旋轉(zhuǎn)和映射                  
 附錄 G 復(fù)數(shù)Wishart分布                  
                   
 術(shù)語                  
                   
 參考文獻(xiàn)