TWI355808B - - Google Patents

Download PDF

Info

Publication number
TWI355808B
TWI355808B TW96140230A TW96140230A TWI355808B TW I355808 B TWI355808 B TW I355808B TW 96140230 A TW96140230 A TW 96140230A TW 96140230 A TW96140230 A TW 96140230A TW I355808 B TWI355808 B TW I355808B
Authority
TW
Taiwan
Prior art keywords
signal
circuit
bit
feedback
generate
Prior art date
Application number
TW96140230A
Other languages
Chinese (zh)
Other versions
TW200828825A (en
Original Assignee
Realtek Semiconductor Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Realtek Semiconductor Corp filed Critical Realtek Semiconductor Corp
Publication of TW200828825A publication Critical patent/TW200828825A/en
Application granted granted Critical
Publication of TWI355808B publication Critical patent/TWI355808B/zh

Links

Landscapes

  • Compression, Expansion, Code Conversion, And Decoders (AREA)

Description

1355808 100年5月4日修正替換頁1355808 May 4, 100 revised replacement page

-------H 九、發明說明: 【發明所屬之技術頜威】 本發明係關於一種三角積分調變器,尤指一種使用在數位類 比轉換器中的電路誤差之多位元三角積分調變器。 【先前技術】-------H IX, invention description: [Technology of the invention] The present invention relates to a triangular integral modulator, especially a multi-bit triangle of circuit error used in a digital analog converter Integral modulator. [Prior Art]

儘管超取樣三角積分調變器(delta-sigma modulator·,DSMJ 係使用較粗糙的量化器,然其仍被廣泛地使用以達到高解析度 的數位類比轉換。第一圖係為傳統二階三角積分調變器之 功能方塊圖,該調變器係包括有一第一加總電路HQ、一第— ,分器120、-第二加總電路13〇、一第二積分器14〇、一量化 f l50、一第一數位類比轉換器' 17〇以及一第二數位類比轉換 該第-加總電路110係用來將一調變輸入信號灿減 =勺反饋信號(或全域反饋信號)抑);該第-積分器二 路0.5ZV(1/)的轉換函數,用以對該第-加總電 第一積八』m订積分;而該第二加總電路130係用來將 分器‘^有-近第二反11信號/2⑼;該第二積 加總電路130的輸出予以進的轉換函數,用以對該第二 轉換器,就)係用來將笫積为;該量化器150(即類比數位 元調變輸出信號;<Λ)。^;積/^ 140的輪出數位化成一 N位 轉換該Ν位元調變^一數位類比轉換器17〇(或DAC) Z0?);該第二數位類 。。> X77)以成為該第一反饋信號 Κ/7)以成為該第二反、& 160轉換該Ν位元調變輸出信號 貝 15 就 /2(η)。Although the delta-sigma modulator, DSMJ uses a coarser quantizer, it is still widely used to achieve high-resolution digital analog conversion. The first picture is the traditional second-order triangular integral. A functional block diagram of the modulator, the modulator comprising a first summing circuit HQ, a first, a divider 120, a second summing circuit 13A, a second integrator 14A, a quantization f L50, a first digital analog converter '17〇 and a second digital analog to digital conversion circuit is used to subtract a modulation input signal = scoop feedback signal (or global feedback signal); The first integrator two-way 0.5ZV (1/) conversion function is used to set the first-plus total electric first product 八m, and the second totaling circuit 130 is used to divide the ' ^ having - near the second inverse 11 signal / 2 (9); the output of the second product summing circuit 130 is subjected to a transfer function for the second converter, for use in convolving to the quantizer; 150 (ie analog analog bit modulation output signal; <Λ). ^; product / ^ 140 round-out digitization into an N-bit conversion of the Ν-bit modulation ^ a digital analog converter 17 〇 (or DAC) Z0?); the second digit class. . > X77) to become the first feedback signal Κ/7) to become the second inverse, & 160 convert the 调 bit modulation output signal 1515 to /2(η).

本說明書提及的‘γ θ - A 間系統的信號。當_ ,表樣㈣時間索引或離散時 月’J 5 K/2)的取樣頻率遠高於傳輪資 1355808 •料的頻寬時,三角積分π掛哭合yI--替換頁 抑制其頻帶的作用::周 • iso $ 換器⑽,三角積和第-數位類比轉 訊比(S聊1 10乃有一個高的頻内信號對量化的雜 稽八 Quantization noise ratio, SQMR) 〇 ^ ^ ^ 调:錢用的積分器數量即表示該調變器的階數。3 。兄’ 一個尚階的調變哭 木 不穩定的頻内信號對量。化的雜;^的光,'a成像以及-個較好但 * 二第,_化_、 為⑽三角積分調; =I1位元數據轉換在本質上是線性且較Γ單的計 在咼階調變器中傕用1你-把』]平日7動作然而’ ,範圍小、有條件性地穩ΐ以及產輸 點可猎由使用多位元數據轉換(即Ν>1) *二k二缺 導致來源電路誤差,此誤差同 或_線性)而 三角積分調變器抑制。 η、里化4 ’係不能有效地由 仁仍無法適用於非常高速的應 y上述問碭, 位演算法,藉由三角積分調變器4=^^個簡單的數 特性來有效地抑制電路誤差。 、*轉換ϋ之非線性 】00年5月4日修正替換頁 •【發明内容】 數位電’dl的係f提供—種使用含有-數位類比轉換器之 少非積分輸,树生—反饋信號來減 加總電路月:三角積分調變器’其中包括-第-一量化哭和積$电路、—第二加總電路、—遽波電路、 號和-°第—加總電路係根據由—調變輸入信 中介信組信號來產生一第- -第—積、〜路鱗5切—中介信號進行積分以產生 第二反饋&;二=加總電路係根據由該第—積分信號和- 一〒’丨化5虎。濾波電路包括—筮— 示 介信號以產生-滹波^。刀電路’用來處理第二中 產生- W - r/, 為係將該濾波信號予以量化以 生N位几(或多位元)輸出信號(N>1,_幻。 反饋電路係接收該N位元輪出俨铼廿你够 換器於-第-反__====轉 Ϊ 一 路徑上包括-㈣位元_ : 以例中’其中於該第二反饋路徑上產生 寻二 w虎’用以提供給第-加總電路作為八^ ,亦或,-實施例中,於該第二反饋路徑 饋信號並將其傳給第二加總電路。 生μ弟一反 一實施例中,反饋電路更包括—(Ν〜1:)位 接收該Ν位元調變輸出信號之(㈣個有:用以 significant blts)。㈣位元累加器包括」咖 器和-延遲元件以產生-(Ν-υ位元的累加_ 广勺加法 進位信號。-加法器接議i位元的^n = 變輪出錢之最騎致位w 加總信號,該加總信料=ost Slgmflcant bits)即產生-換器以產生第一反饋^J卢;—反饋路輕上提供給數位類比轉 法器的輸出作為ChG V實施例中,藉由提供(N—1)位元加 位元數位類比轉換哭的輪:=立^比轉換器’的輸入並將該(N-1) 反饋信號給該第一力:‘。二額外的 位兀加法器的延遲輪 —、1中,糟由如供(M) (N-1)位元數位類比轉位元數位類比轉換器並將該 n a姐、專換态的輪出乘以一個縮放來數,淮而吝4 '"弟—反饋信號給該第二加總電路。 乂 產生 可以配第—組信號更包括一第三反饋信號。例如, -置#—數_、比轉換器來接收N 生弟三反饋信號給第 二又勒出域亚產 變器的輸出端和加總電_^:^反饋路徑軸接於調 一二實施例t,攄波電路更包括—第三加總電路,用 一弟二組,紅桓重加來產生—第三中介信號,其中該、 ^虎包括第二積分電路的輸出以及輸人;或者,該=二 包括第二積分電路的輸出以及第二加總電路的^出-、、4 能出現其他繞過-或多個加總電路的前饋補償路徑。 也有可 一貫施例中,錢電路包括至少—額外的加總電路 少-額外的積分電路以增加多㈣三角積分調變器之階數 多位7L二角積分調變H中的積分器可以是連續時間積分哭 (continuous-time integrators)或離散時間積分 ^ (discrete-time integrators)。例如,離散時間積分哭刀= -使用-對輸人電容的差動交換式電容積分器(咖打二咖 switch-capacitor integrator)、一差動運算放大 Da (differential 〇Perational amplifier)、_對反饋電容以2 10 L100年5月4日修正替換頁 複數個辦脈信號控制的 ^^— 在單路差動運算放大哭卜㈣一接1%例中弟—加總電路與 第-對電軸於,電路相結合’其中透過- 類比轉又触第二對電容祕於數位 輸出’以及藉由複數辦脈信號來 _算放大器和複數個開關。同樣地,第-力i ==差動運毅大8上的電容比來權重地加總第二 -實施例中’數位類比轉換器係設置在使料個三對 4(3-to-l multipiexer)的差動電路(diff_tiai circu 。例如’提供2位元的輸入信號給轉接在兩個三 :選擇線上之三元數位觀器的輸⑽, ; ,三:固參考電壓給每個三對-多工器的輸入端。第一 夕工裔產生-絲信號的正向端(pQsitive end),而第二個三 對-多工器產生差分信號的負向端(negative _)。 貝把例中’本發明又揭示—種減少電路誤差的方法,係 在—纽元三純分調變11上的—N位元反饋數位類比 轉換克’其中該多位几三角積分調變器會取代N位元反饋數位 ,比轉換器為-(N-1)位元數位類比轉換器和一數位類比轉換 裔’其執行步驟包括有:對第-組信號執行一第一權重加法以 產生-第-中介化说’其中該第一組信號包含了調變輸入信號 和-第-反饋信號。隨後對第-中介信號作積分以產生一 積分信號。該方法更包括對第二組信號執行一第二權重加法以 產生-第二中介信號’其中該第二組信號包含了第—積分信號 和-第二反饋信號。第二中介信號經過濾後會產生」濟^: 號,該濾波信號隨後被予以量化而產生一 N位元調變輸出信^The signal of the 'γ θ - A system mentioned in this specification. When the sampling frequency of _, the model (4) time index or the discrete time month 'J 5 K/2) is much higher than the bandwidth of the transmission 1355808, the triangular integral π hangs and yI-- replaces the page to suppress its frequency band. The role of: Zhou • iso $ converter (10), triangulation and the first-to-digital analog ratio (S chat 1 10 is a high intra-frequency signal to quantify the Quantization noise ratio, SQMR) 〇 ^ ^ ^ Tune: The number of integrators used for money represents the order of the modulator. 3 . Brother's a gradual change of crying wood unstable intra-frequency signal pairs. The light of the ^; the light of the image, the 'a image and the better one but the second two, the _ _, is the (10) triangular integral tone; the =I1 bit metadata conversion is linear in nature and is more simple. In the modulator, you can use 1]-] 』] weekday 7 action however, the range is small, conditionally stable, and the production and delivery points can be hunted by using multi-bit metadata conversion (ie Ν>1) *2k The second deficiency causes the source circuit error, which is the same as or _ linear) and the delta-sigma modulator suppresses. η, 里化4 ' can not effectively be used by the kernel can not be applied to very high speed y y, the bit algorithm, by the triangular integral modulator 4 = ^ ^ a simple number of characteristics to effectively suppress the circuit error . , *Conversion non-linearity] May 4, 00 revised replacement page • [Invention] Digital power 'dl' f provides a kind of non-integral input using a digital-to-digital analog converter, tree-feedback signal To reduce the total circuit month: the delta-sigma modulator, which includes - the first-quantization of the crying and product $ circuit, the second summing circuit, the chopper circuit, the number, and the - ° first - total circuit based on - modulating the input signal interrogation signal to generate a first - - first product, ~ road scale 5 - the intermediate signal is integrated to generate a second feedback &; = = total circuit is based on the first - integral signal and - One 〒 '丨化5虎. The filter circuit includes - 筮 - the display signal to generate - ^ wave ^. The knife circuit 'is used to process the second generated - W - r / , to quantize the filtered signal to generate an N-bit (or multi-bit) output signal (N > 1, _ illusion. The feedback circuit receives the N-bit rounds out, you have enough to change the -----__==== turn Ϊ a path includes - (four) bits _: in the example 'where the second feedback path produces two The w tiger' is provided to the first-plus-slave circuit as an 八, or, in the embodiment, the signal is fed to the second feedback path and passed to the second summing circuit. In the example, the feedback circuit further includes a - (Ν~1:) bit to receive the 调 bit modulated output signal ((4) has: used for significant blts). (4) The bit accumulator includes a "coffee device" and a delay element Generate - (Ν - 累 的 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Slgmflcant bits) is to generate a first feedback to generate a first feedback. The feedback path is provided to the output of the digital analog converter as a ChG V embodiment by providing (N-1) bits. The bit digit analog analog converts the crying wheel: = the input of the converter = and converts the (N-1) feedback signal to the first force: '. 2 additional bits 延迟 the delay wheel of the adder — 1 , by the (M) (N-1) bit digit analog analog bitwise analog converter and multiply the round of the na sister, the special state by a scale, Huai 吝 4 '" The feedback signal is given to the second summing circuit. The 乂 generation can be matched with the first group signal and further includes a third feedback signal. For example, - set #-number _, ratio converter to receive N sin three feedback signals to Secondly, the output end of the sub-product transformer and the total power _^:^ feedback path axis are connected to the second embodiment t, and the chopper circuit further includes a third total circuit, using a second group, The red dragonfly is re-added to generate a third intermediate signal, wherein the tiger includes the output of the second integrating circuit and the input; or, the =2 includes the output of the second integrating circuit and the output of the second summing circuit - , 4 can have other feed-forward compensation paths that bypass the - or multiple summing circuits. There are also consistent applications where the money circuit includes at least - Less total circuit - additional integration circuit to increase the order of multiple (four) delta-sigma modulators Multi-bit 7L two-angle integral modulation H The integrator can be continuous-time integrators or discrete time Integral ^ (discrete-time integrators). For example, discrete time integral crying knife = - use - differential switching capacitive integrator for input capacitance (switch-capacitor integrator), a differential operation to enlarge Da ( Differential 〇Perational amplifier), _ for the feedback capacitor to 2 10 L100 May 4 correction replacement page multiple pulse signal control ^^ - in a single differential operation to enlarge the crying (four) one after the 1% case - The summing circuit is combined with the first-to-on-axis, the circuit is 'in which the analog-to-analog turns to touch the second pair of capacitors to the digital output' and the multi-signal is used to calculate the amplifier and the plurality of switches. Similarly, the first force i == differential capacitance on the power amplifier 8 is added to the weight ratio second - in the embodiment, the 'digital analog converter system is set in the three pairs of 4 (3-to-l Multipiexer) differential circuit (diff_tiai circu. For example, 'provide a 2-bit input signal to the two-way three-way: select the ternary digitizer on the line (10), ;, three: solid reference voltage for each three The input end of the multiplexer. The first eve of the work produces the pQsitive end of the wire signal, while the second three-pair multiplexer produces the negative side of the differential signal (negative _). In the example, the present invention discloses a method for reducing circuit error, which is based on the -N-ary three-dimensional partial modulation conversion 11 - N-bit feedback digital analog conversion gram, where the multi-digit triangular integral modulator Substituting the N-bit feedback digits, the ratio converter is a -(N-1)-bit digital-to-bit analog converter and a digit-to-digital analog converter. The execution steps include: performing a first weight addition on the first-group signal to generate - The first-mediation says that the first set of signals contains a modulated input signal and a -th-feedback signal. The signal is integrated to generate an integrated signal. The method further includes performing a second weight addition on the second set of signals to generate a second intermediate signal 'where the second set of signals includes a first integrated signal and a second feedback signal After filtering, the second intermediate signal will generate a "me:" number, and the filtered signal is then quantized to generate an N-bit modulated output signal.

丄JJJOUO 位元來產生_(N ^ N位元調變輸出信號_—l)較低有效 位信號加上N位元^^信號和—丨紅的進健號,將該進 信號,透過數位:ΐί。出信號的最高有效位元以產生一加總 另外—^專、為轉換該加總信號成為第一反饋信號。 (N-i)位元餘和域會猶過—包含有 號。例如,使㈣^裔之—反饋路徑來轉換成第二反饋信 ㈣較低有效位元與變=信號的 並提供_-nn 遲以產生0^1)位元總和信號, 出。(N-1)位元數二=成為(N—1)位元數位類比轉換器的輸 )位疋數位類比轉換器的輸出會被縮放 =I數位類比轉換器以及對(Ν_υ位元 輸,以進行微分來產生,卜的反饋信號,其中== 第—組信號合併以進而產生第—中介信號。其它反 饋U虎可域行Ν位元觀_錄龍_轉換 ^ = 70三角積分調變器也可使用前饋補償方式來處ς 務j入就。 以上之概述與接下來的詳細說明及附圖,皆是為了铲、一 步說明本發料達成就目的所採取之方式、手段及功=進而 有關本發明的其他目的及優點,將在後續的說明及圖式中加以 闡述。 【實施方式】 本發明所揭示之-種減少電賴差的方法及裝置,例如適 用於在一多位元三角積分調變器中的反饋數位類比轉換器。然 12丄JJJOUO bit to generate _ (N ^ N bit modulation output signal _ - l) lower significant bit signal plus N bit ^ ^ signal and - 丨 red health key, the incoming signal, through the digit :ΐί. The most significant bit of the signal is output to generate a total of additional signals, and the converted signal is converted into a first feedback signal. (N-i) Bits and domains will still pass - contain numbers. For example, the (four)^--the feedback path is converted into the second feedback signal. (4) The lower-effective bit and the variable=signal are provided and the _-nn is delayed to generate the 0^1 bit sum signal. (N-1) bit number two = become the (N-1) bit analog converter output) bit 疋 digital analog converter output will be scaled = I digital analog converter and pair (Ν _ υ bit input, The differential signal is used to generate the feedback signal, wherein == the first group signal is combined to generate the first intermediate signal. Other feedback U tiger can be used to perform the bit position view_recording dragon_conversion ^=70 triangular integral modulation The feedforward compensation method can also be used to enter the service. The above summary and the following detailed description and the accompanying drawings are all for the purpose of shovel and one step to explain the purpose, means and merits of the purpose of the release. Further objects and advantages of the present invention will be described in the following description and drawings. [Embodiment] The present invention discloses a method and apparatus for reducing electric power difference, for example, in a plurality of positions. Feedback digital analog converter in the meta-triangular integral modulator.

100年5月4日修正替換頁 且述本發明的數較_表明本剌翻在許多方面, 且不會破特洙條件限制其功用。 藉八2閱第二細,此實施例係為本發明所揭示之多位元三角 =_器魏之魏方顧。多㈣三_分調變器飄包 弟加總(summation)電路210Α,其係根據一第 一組信 f之域值來產生一第一中介㈤赚diate)信號ri⑶。例 ',二域電賴_機“錢心)減M —反饋信號 f和弟一反饋信勤,心)以產生第一中介信號咖,並藉由 =積分電路咖對該第-中介信動⑶予以進行積分來產生 弟-和分信號/,ω。舉例來說,第—積分電糊有一約為 予5ζ /(11 )的轉移函數,提供第—積分信號ζ⑼給第二加總 ,路23Q—A,剛二組信號的權重加總值(we神d _) 二產生第:t AK/?) ’例如,藉由第二加總電路2継來將 弟-積分信號/心)減去第三反饋信號a心)(或稱區域反饋信號 )以產生第二中介信號r2〇7)。 多位元三角積分調變器雇更包括一渡波電路24〇,係用以 處理第二中介信號乃㈤以產生一滤波信號/心)。如第二A圖所 不’濾波電路240包括-第二積分器,其適用於二階三角積分調 變器,該濾波電職0也可包括額外的積分器和加總電路以適用 於更多階的三角積分調變器。此外,第—積分電路娜和遽波電 路240的積分a麵離散時間積分n,也可為連續時間積分器。 舉例來說。]I波電路240有-約松-νπγ)的轉移函數,提供 濾波信號(或第二積分信號)/2(/7)給N位元的量化器25〇,該量化 器250會將級信號Λ(/7)抑進行量化以產生—N位元調變輸出 信號y(/?),其中,N係為一大於1的整數。 13 100年5月4曰修正替換頁 多位元三角積分調變器200A包括多個反饋電路。第一反饋 電路295A產生第-反饋信號ai⑼和第二反饋信號2⑼並將 其傳給第-加總電路2l〇A。第二反馈電路26〇產生第三反饋信號 出〇7)並提供給第二加總電路23〇A。透過額外的反饋電路來產生 額外的反饋信號給第-加總電路21〇A、第二加總電路襲或其 他位在濾波電路240上的加總電路。一實施例中,第二反 260和任何其他⑽電路係制位元触職器,該轉換器 的輸入端搞接於N位元調變輸出信號^(^)。 -實施例中’第—反饋電路m包括―㈣)位元加法哭 280、-延遲電路265 (如:一單位樣本(-a卿⑹延遲^ 件)、:(N—D位元數位類比轉換器270、-微分器275A、一丄位 tl加法器293和-三元數位類比轉換·Q。财延遲電路挪的 0H)位元加法器280累加N位元調變輸出信號之⑹)較低有效 位元(即咖))’進而產生一⑹)位元的加法信號s㈤和以元 的進位信號c〇7)。提供(μ])位元的加法信號办)給延遲電路 265以產生-延遲加法信號⑽)來作為㈣位元加法職 的輪入值。另外又提供㈣位福加法信號办)給㈣位元 數位類比賴湖以產生—類比輸出咖,鋪比 ,:步地透過微分纖鐵分後成為第二反齡勤,心)。) =提健號心位元晴輸出信號的最高有 效位兀(即咖)給丨位元的加法器293以產生一2位元的三 轉換崎處理該三元信觸以般 户明比圖的多位元三角積分調變器2_和第-圖的 多位元三角積分調變器刚,其中的差別就是使用第一反饋=Amendment of the replacement page on May 4, 100, and the description of the present invention _ indicates that the 剌 剌 在 在 在 在 在 在 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 By means of the second and second, the embodiment is a multi-bit triangle disclosed by the present invention. The multi (four) three_segment transformer wrapper summation circuit 210Α generates a first intermediaries (five) earning dire) signal ri(3) according to the domain value of a first set of letters f. Example ', the second domain electric _ _ machine "Qian Xin" minus M - feedback signal f and the younger one feedback letter, heart) to generate the first intermediary signal coffee, and by the = integral circuit coffee to the first - intervening (3) Integrate to generate the di-and sub-signal /, ω. For example, the first-integral electric paste has a transfer function of about 5 ζ / (11), providing the first-integral signal 9 (9) to the second summation, 23Q-A, the weight of the two sets of signals plus the total value (we god d _) two produces the first: t AK /?) 'For example, by the second summing circuit 2 将 to reduce the brother-integrated signal / heart) Going to the third feedback signal a heart) (or region feedback signal) to generate a second intermediate signal r2 〇 7). The multi-element triangular integral modulator employs a wave circuit 24 〇 for processing the second intermediary The signal is (5) to generate a filtered signal/heart. As in the second Figure A, the filter circuit 240 includes a second integrator, which is suitable for a second-order delta-sigma modulator, which may also include additional The integrator and the summing circuit are applied to a more order triangular integral modulator. In addition, the product of the first-integral circuit and the chopper circuit 240 The a-plane discrete time integral n can also be a continuous time integrator. For example, the I-wave circuit 240 has a transfer function of -Jorsong-νπγ, providing a filtered signal (or a second integrated signal)/2 (/ 7) to the N-bit quantizer 25A, the quantizer 250 quantizes the level signal Λ(/7) to generate a -N-bit modulated output signal y(/?), where N is a An integer greater than 1. 13 Multi-bit triangular integral modulator 200A includes a plurality of feedback circuits. The first feedback circuit 295A generates a first feedback signal ai (9) and a second feedback signal 2 (9) and The second feedback circuit 26 generates a third feedback signal output )7) and supplies it to the second summing circuit 23A. An additional feedback circuit is used to generate an additional feedback signal. a first summing circuit 21A, a second summing circuit or other summing circuit on the filter circuit 240. In one embodiment, the second counter 260 and any other (10) circuit system bit contacts, The input end of the converter is connected to the N-bit modulation output signal ^(^). In the embodiment, the 'first-feedback circuit m includes (4) Bit addition 295, - delay circuit 265 (eg: one unit sample (-a qing (6) delay ^), : (N-D bit digital analog converter 270, - differentiator 275A, a tl bit tl Adder 293 and - ternary digit analog conversion · Q. The delay delay circuit shifts 0H) The bit adder 280 accumulates the N bit modulated output signal (6)) the lower significant bit (ie, coffee)) to generate a (6)) the addition signal s (f) of the bit and the carry signal c 〇 7) of the element. The (μ) bit addition signal is supplied to the delay circuit 265 to generate the -delay addition signal (10) as the (four) bit plus function. The turn-in value. In addition, it provides (four) Fu Jiafa signal to give (four) bits to the digital analogy of Laihu to produce - analog output coffee, shop ratio, step by step through the differential fiber and iron to become the second anti-age, heart). = = the most significant bit of the heart-level output signal (ie, coffee) is added to the adder 293 of the bit to generate a 2-bit three-conversion, and the three-dimensional letter is compared to the map. The multi-bit triangular integral modulator 2_ and the multi-bit triangular integral modulator of the first graph, the difference is to use the first feedback =

1355808 1 I 100年5月4日修正替換頁 295A取代第一N位元數位類比轉換器17〇。以下說明擁有三元數 位類比轉換器290和(N-1)位元數位類比轉換器27〇功能的第一 反饋電路295A是如何達到第,立元數位類比轉換器17〇的功能 〇 例^1,N位元調變輸出信號y(/?)有個2〃可能的數值:{〇,L 2’ ,2 },這邊假設第—圖和第二八圖中所有的數位類比轉換 β都是理想的且較低有效位元的權重都是近似於△,第一圖中 的調變器100之第一反饋信號力⑼由以下等式定義: f\ (^) = y(ri) · Δ (1) 在第二A圖的調變器申,係藉由對從(N-1)位元數位類比轉 換器傳來的類比輪出决(刀)做(1 -夕)的微分運算進而得到第二反 饋信號<9’ 2(/7)。第二A圖中的第二反饋信號係由以下等式定義: a\ {n)^a2{n)-a2{n-\) (2) Ν位元調變輪出信號的ΟΜ)較低有效位元和Ν位元調變輸 出信號的最高有效位元可由以下數學算式表示: 八⑻=mod(y⑻,2"-丨) ⑶ yM («) = \y(rt) - yL (n)]/2N-] ⑷ 附加單位樣本延遲(unit_sampledelay)265的(N—1)位元加 法器280對N位元調變輸出信號的(N—丨)較低有效位元進行累加 以產生(N-1)位元的加法信號义乃)和1位元的進位信號^(刀), (N_1 )位兀的加法信號〆乃)和1位元的進位信號C(/7)係由以下數 學异式表示: 5(^) = mod^^ (n) + s(n -1),2 ) ⑸ ♦) = [^⑻ + 咖-1) - φ)]/2ΛΜ (6) 由以下算式可看出2位元的三元信號甙77)係由C(;7)和j^/7) 加總求得: 15 100年5月4曰 d{n) = c{n) + yM{n) 修正替換頁1355808 1 I May 4th, 100th revised replacement page 295A replaces the first N-bit digital analog converter 17〇. The following describes how the first feedback circuit 295A having the function of the ternary digit analog converter 290 and the (N-1) bit digital analog converter 27 is up to the first function of the dynamometer analog converter 17 ^ The N-bit modulation output signal y(/?) has a possible value of 2〃: {〇, L 2' , 2 }, which assumes that all the digital analog conversions in the first and second eight graphs are It is ideal and the weight of the lower significant bit is approximately Δ. The first feedback signal force (9) of the modulator 100 in the first figure is defined by the following equation: f\ (^) = y(ri) Δ (1) In the second A diagram, the differential operation is performed by the analog wheel (knife) from the (N-1)-bit digital analog converter. Further, a second feedback signal <9' 2(/7) is obtained. The second feedback signal in Figure 2A is defined by the following equation: a\ {n)^a2{n)-a2{n-\) (2) Ν 调 调 轮 轮 轮 ΟΜ) The most significant bits of the effective bit and the 调 bit modulated output signal can be represented by the following mathematical formula: 八(8)=mod(y(8),2"-丨) (3) yM («) = \y(rt) - yL (n) ]/2N-] (4) The (N-1) bit adder 280 of the additional unit sample delay (unit_sampledelay) 265 accumulates the (N-丨) lower significant bits of the N-bit modulated output signal (N) -1) The addition signal of the bit is the sum signal) and the 1-bit carry signal ^ (knife), the (N_1)-bit addition signal 〆 is) and the 1-bit carry signal C (/7) is the following mathematics The expression is: 5(^) = mod^^ (n) + s(n -1), 2 ) (5) ♦) = [^(8) + coffee-1) - φ)]/2ΛΜ (6) The following formula can be used It can be seen that the 2-bit ternary signal 甙77) is obtained by summing C(;7) and j^/7): 15 100 May 4曰d{n) = c{n) + yM{n ) Correct replacement page

(N-1)位元加法信號汉/7)的最低效位元之權重也近似於厶 ,但1位元的進位信號^(/7)大約是2"1倍高,因為其係從(Ν_υ位 元加去态280進位的。从/?)係因Ν位元調變輪出信號^乃)的最苎 有效位元而使得其權重為f倍高。因此,從(Ν])位元數位類= 轉換β 270傳來的類比輸出信號敌㈠)和第二a圖中的第一回饋作 號5丨(/?)可由以下算式求得: 、。 a2(n) = s(n)-A (8) 4 ⑻哨《). (2ΛΜ △) = [c⑻ + & ⑻].(2"-ι △) ⑼ 根據等式⑵和⑻,第二八圖中的第二反饋信號占,細可改 由以下算式表示: a'2 («) = α2 (η) - α2 (η -1) = s(n)d, - s(n - 1)Δ (J〇) 根據等式(10)、(9)、⑹和⑷,第二Α圖中的第二反饋信 號2(/2)可由以下算式簡化表示: 2 (扣,⑻=s⑻△-咖-ΐμ + 咖)_(2λμ△)+〜⑻(2叫△) (11) -々)△ φ - !)△ + h ⑻ + 咖 _ ” _ 咖)]△ + ♦) △⑻△ =y(n) · Δ —從等式(1)和(11)得知,第一圖中的第一反饋信號/ι(/7)等同 於第二Ail中的第-反饋信號a(/?)和第二反饋信號? 2〇?),因 此’若所有的數位類比轉換器皆是理想的,則擁有三元數位類 比轉換益290和(N-1)位元數位類比轉換器27〇的第一反饋電路 295A在功能上係等同於第—_元數賴比轉換器】7〇。 /⑻二心⑻+巧⑻ (12) u當數鋪比轉換器並非為理想時,第二A圖中具備第一反饋 電路295A的多位元三角積分調變器2_會比第-圖中的多位元 1355808 】〇〇年5月4日修正替換頁 三角積分調變器⑽的效能更好 ^ 調變器100中,從第一·μ圖中的夕位元三角積分 差將在第-加總電路11〇之輸入端弓= 奐器17,0所產生的電路誤 雜訊就像調變輸入H槐"Λ “、外的雜訊。這些額外的 訊益、去㈣f 虎一樣’χ傳送特性的影響,因而這此雜 Λ…'法於§周芰态的輸出端被抑制。 二才准 轉換多位凡三角積分調變器2_可抑制數位類比 。、差a在斂分姦275A的輸入端引 =結:第-加總電路驗的調變輸入信:前= 變器輸出端:::位元數位類比轉換器⑽之雜訊可於調 提供三元數位類比轉換器29〇的輸出值給第一加 210A,因而任何由二开盍f /作东g μ姑 〜毛路 —數位通比轉換克290產生的電路誤差都會 在第一加總電路210Α的輸入端引起額外的雜訊 訊就像調變輸人信號—樣,受傳送特性的影響,_這1= 二=數偏趾轉換〇之雜訊無法於調變器的輸出端被抑制 。心而,貫作於全差動電路拓撲(fully伽价 _〇1嗯)上的三元數鋪比賴最小麵性相關電路 ΐ合時1本紅仍然可達到線性的功效。所以,第二A圖中的調 k:輸出#號可藉由在反饋數位類比轉換器的電路誤差來相對地 避免掉雜訊。 第二。B。圖說明另-個多位元三角積分調變器2_的實施例 ’此調變器200B更佳地省略了微分器275A。第二A圖中的微分器 275A是設置於第-積分電路2默後,微分器若置於積分器之後 則表示除了會產生一延遲和一縮放信號外,實際上對信號並不 17 100年5月4曰修正替換頁 會起任何作用。例如,結合了第二A圖中的微分器275A以及第一 積分電路220對類比輸出32(77)的作用是0. 5/ (即對單位取樣延 遲之值做一個參數為0. 5的縮放動作)。因此,可繞過第一積分 電路220來移除微分器275A以產生一個適當的縮放和延遲值給 (N-1)位元數位類比轉換器27〇的類比輸出激⑼。 第二B圖中的第一加總電路210B會根據調變輸入信號X/7) 和第一反饋信號《si (/?)的總和值來產生第一中介信號〆,(/3)。隨 ,就由第一積分電路220對第一中介信號r’,(/?)作積分來產生 第一積分信號尸,0)。之後提供第一積分信號厂1(/?)給第二加 總電路23 0B以根據第二組信號的權重加總值來產生第二中介信 竣1(/?)。例如,第二加總電路230B將從第一積分信號/,,(…減 去第一反饋彳§號〇. 5a” 2(/?-1)和第三反饋信號故!^)來產生第二 中介信號r2(/?h之後再由濾波電路24〇處理第二中介信號厂2(^) 以產生濾波信號/2(Λ) ’該濾波信號,(或第二積分信號)會 被提供給Ν位元量化器250來將其數位化,進而產生糾立元調變^ 出信號jK/?)。 多位元二角積分調變器2〇〇b包括一第一反饋電路295B,該 第-反饋電路2舰侧以產生第-反饋錢ai㈤給第一加總電 路210B以及產生第一反饋號〇· 2(Λ_ι)給第二加總電路 230Β。第二反饋電路26〇(即ν位元數位類比轉換器)也會產生第 二反鎖信號激(/?)給第二加總電路23〇β。 一實施例中’第一反饋電路295Β包括-oh)位元加法器 280、延遲電路265、-(Ν-1)位元數位類比轉換器27〇、一縮 放參數器獅、-1位元加法·3以及—三元數_比轉換器 290。附有延遲電路265的㈤)位元加法器累加N位元調變輸 18 1355808 • < 100年5月4日修正替換頁 出t號之(N-1)較低有效位元(即只(乃)),進而產生一(Ν_ι)位元 的加法彳§唬式/?)和1位元的進位信號叭⑺。提供⑺—丨)位元的加 法#號5(/7)給延遲電路265以產生一延遲加法信號办_乃,來作 為(N-1)位元加法器28〇的輸入值,隨後將該延遲加法信號汉刀一乃 傳至(N-1)位元數位類比轉換器27〇以產生一延遲類比輸出 5 ,更進一步地透過縮放參數器275B縮放該延遲類比輸 出5” 2(/?-1)以成為第二反饋信號〇. 5a” “刀^。之後提供1位元 的進位信號K/7)和N位元調變輸出信號的最高有效位元(即乃<;?)) 給1位元的加法态293以產生一加總信號(如一 2位元的三元信號 成/?))。二元數位類比轉換器290會處理該三元信號成^)以產生 第一反饋信號51(77)。 第二B圖中的多位元三角積分調變器2〇〇B在功能上等同於 第二A圖中的多位元三角積分調變器2〇〇A。第一反饋電路29诏更 精簡地刪去微分器275B以達到在第二A圖中之第一反饋電路. 295A的功能,其係藉由提供延遲加法信號$& 給(n)位元數 位通比轉換為270 ’並適當地縮放(N-i)位元數位類比轉換器wo 的類比輸出值,以及將縮放後之值作為第二反饋信號給第二加 總電路230B ’而不是給第一加總電路2i〇b。 第二C圖中揭露多位元三角積分調變器200C之又一實施例 。第一C圖中的多位元二角積分調變器2〇〇c本質上近似於第二β 圖中的多位元三角積分調變器2〇〇Β,差別僅在於減少了第二反 饋電路260以及增加了 一前饋路徑。本發明係適用於使用前饋補 償,區域前饋補償或其兩者所結合之多位元三角積分調變器。 舉例來說,如第二c圖中所示,多位元三角積分調變器2〇〇c 具備了繞過濾波電路240的前饋路徑,並包括至少一積分電路。 19 1355808 100年5月4日修正替換頁 上述前饋路徑包括-增益器(gain bl。⑻,其用以接收第二加 總電路23QC的輸出值(或濾波電路24Q的輪入值)並產生一前饋 #號,而s亥兩饋#號隨後會在第三加總電路ms中結合濾波電路 240的輸出值。前徑在功能上等同於第二㈣巾^元^立 類比轉換器之區域反饋路經。前饋路徑也可為繞過第—積分带 路220或其他積分電路(圖中未示)。 包 第三圖揭露了實作於-全差動電路招墣上的三元數位類比 轉換器300之-實施例。三元數位類比轉換器3〇〇係接收一2位元 的三元信號(或㈣社DATA)並輸出—對應的差分信號,立 中該三從财三個可缝值_,丨,2),而該差分信號包括一 正向端㈣肋e end, V〇UTP)和—負向端(卿⑽_ vo_。三元數位類比轉換器300使用三種參考電壓:v_、’ VCM和麵,差分錢的正向端透料—“細祕於三種 =轉壓,而差分信號的負向端透過第二多工器32〇搞接 ° ^131°' 控制,且三讀鶴比轉換器3_輪出是正向端和負向端之間 的電壓差(即多工器31〇 ' 32〇的各別輸出) 位類比轉換()的可故輸纽,而三域蝴 一^ 的三種可能輸出㈣等間隔產生的,因此n ^ 何,三元數位類比轉換器300本質上都屬線性的卿/ 為The weight of the least significant bit of the (N-1) bit addition signal han/7) is also similar to 厶, but the 1-bit carry signal ^(/7) is about 2"1 times higher because it is from ( Ν _ υ 加 加 加 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 Therefore, the analogy output signal (1) from the (Ν)) bit number class = conversion β 270 and the first feedback number 5丨 (/?) in the second a picture can be obtained by the following formula: . A2(n) = s(n)-A (8) 4 (8) Whistle "). (2ΛΜ △) = [c(8) + &(8)].(2"-ι △) (9) According to equations (2) and (8), second The second feedback signal in the eight diagrams can be changed by the following formula: a'2 («) = α2 (η) - α2 (η -1) = s(n)d, - s(n - 1) Δ (J〇) According to equations (10), (9), (6) and (4), the second feedback signal 2(/2) in the second graph can be simplified by the following formula: 2 (deduction, (8) = s(8) Δ-咖-ΐμ + 咖)_(2λμ△)+~(8)(2called △) (11) -々)△ φ - !)△ + h (8) + coffee _ _ _ 咖)]△ + ♦) △(8)△ = y(n) · Δ - From equations (1) and (11), the first feedback signal /1(/7) in the first graph is equivalent to the first feedback signal a in the second Ail (/? And the second feedback signal? 2〇?), so 'if all digital analog converters are ideal, then have a ternary digit analog conversion 290 and (N-1) bit analog converter 27〇 The first feedback circuit 295A is functionally equivalent to the first-to-digital ratio converter [7]. / (8) two-core (8) + coincidence (8) (12) u when the number-pitch converter is not ideal, the second A The figure has the first anti The multi-element delta-sigma-integrator 2_ of the feed circuit 295A will be better than the multi-element 1355808 in the first picture. The modified page delta-integral modulator (10) is better on May 4th of the following year. ^ Modulator In 100, the delta-level delta integration difference from the first·μ map will be at the input end of the first-to-high total circuit 11〇, and the circuit error generated by the circuit is like a modulation input H槐. "Λ ", outside the noise. These extra signals, the effect of the (four)f tiger's transmission characteristics, and thus the miscellaneous...'s method is suppressed at the output of the § week. Secondly, the conversion of multiple bits of the triangular integral modulator 2_ can suppress the digital analogy. , the difference a in the convergence of the 275A input terminal = knot: the first - plus total circuit test modulation input letter: before = transformer output::: bit digital analog converter (10) noise can be provided The output value of the ternary analog converter 29〇 is given to the first addition 210A, so any circuit error generated by the second opening f / the east g μ 〜 毛 数 数 数 数 数 290 290 290 290 will be in the first total The input of the circuit 210Α causes additional noise to be transmitted like a modulated input signal, which is affected by the transmission characteristics. The noise of the 1=2=number-to-toe conversion is not available at the output of the modulator. inhibition. At the same time, the ternary number on the fully differential circuit topology (fully gamma _〇1) is compared to the minimum surface correlation circuit. When the combination is 1 red, the linear effect can still be achieved. Therefore, the adjustment k: output # in the second A diagram can relatively avoid the noise by feeding back the circuit error of the digital analog converter. second. B. The figure illustrates an embodiment of another multi-bit delta-sigma modulator 2_. This modulator 200B more preferably omits the differentiator 275A. The differentiator 275A in the second A picture is set after the first-integration circuit 2 is silent. If the differentiator is placed after the integrator, it means that the signal is not 17 100 years except that a delay and a scaling signal are generated. Correcting the replacement page on May 4 will have any effect. 5的放大。 For example, the value of the unit sampling delay is 0.25. action). Thus, the differentiator circuit 275 can be bypassed to remove the differentiator 275A to produce an appropriate scaling and delay value to the analog output of the (N-1) bit digital analog converter 27A (9). The first summing circuit 210B in the second B diagram generates a first intermediate signal 〆, (/3) according to the summed value of the modulated input signal X/7) and the first feedback signal "si (/?). Then, the first integration circuit 220 integrates the first intermediate signal r', (/?) to generate a first integrated signal corpse, 0). A first integrated signal factory 1 (/?) is then provided to the second summing circuit 23 0B to generate a second intermediate signal 竣1 (/?) based on the weighted total value of the second set of signals. For example, the second summing circuit 230B will generate the first from the first integral signal /,, (... minus the first feedback 彳§ 〇. 5a" 2 (/?-1) and the third feedback signal !!! The second intermediate signal r2 (/?h is then processed by the filter circuit 24〇 to process the second intermediate signal factory 2(^) to generate a filtered signal/2(Λ) 'the filtered signal, (or the second integrated signal) is provided The bit quantizer 250 digitizes it to generate the erecting element modulation signal jK/?). The multi-bit binaural integral modulator 2〇〇b includes a first feedback circuit 295B, which is - feedback circuit 2 ship side to generate a first feedback money ai (five) to the first summing circuit 210B and generate a first feedback number 〇 2 (Λ _ι) to the second summing circuit 230 Β. The second feedback circuit 26 〇 (ie ν bit) The digital bit analog converter) also generates a second anti-lock signal (/?) to the second summing circuit 23 〇 β. In an embodiment, the 'first feedback circuit 295 Β includes -oh) the bit adder 280, the delay circuit 265, - (Ν-1) bit digital analog converter 27, a scaling parametric lion, -1 bit addition · 3 and - ternary - ratio converter 290. Delay circuit 265 is attached. (5)) The bit adder accumulates the N bit modulation transfer 18 1355808 • < On May 4, 100, the replacement page is replaced by the (N-1) lower effective bit (ie only ()), and then An addition 一 唬 / ? ? ? ? ? ? ? ? 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 产生 7 7 7 7 7 7 7 7 The addition signal is used as an input value of the (N-1)-bit adder 28〇, and then the delayed addition signal is transmitted to the (N-1)-bit digital analog converter 27 to generate A delay analog output 5, further scaling the delay analog output 5" 2 (/?-1) through the scaling parameter 275B to become the second feedback signal 〇. 5a" "knife ^. The 1-bit carry signal K/7) and the most significant bit of the N-bit modulated output signal (ie, <;?)) are then provided to the 1-bit adder 293 to generate a summed signal (eg, The 2-bit ternary signal is /?)). The binary digital analog converter 290 processes the ternary signal to generate a first feedback signal 51 (77). The multi-bit delta-sigma modulator 2 〇〇 B in the second B diagram is functionally equivalent to the multi-bit delta-sigma modulator 2 〇〇 A in the second A diagram. The first feedback circuit 29 删 more simplifies deleting the differentiator 275B to achieve the function of the first feedback circuit 295A in Figure 2A by providing a delayed addition signal $& to (n) bit digits The ratio is converted to 270' and the analog output value of the (Ni) bit digital analog converter wo is appropriately scaled, and the scaled value is used as the second feedback signal to the second summing circuit 230B' instead of the first plus The total circuit 2i〇b. A further embodiment of a multi-bit delta-sigma delta transformer 200C is disclosed in Figure 2C. The multi-bit two-angle integral modulator 2〇〇c in the first C picture is essentially similar to the multi-element triangular integral modulator 2〇〇Β in the second β-picture, the only difference being that the second feedback is reduced. Circuit 260 also adds a feedforward path. The present invention is applicable to a multi-element triangular integral modulator using feedforward compensation, regional feedforward compensation, or a combination thereof. For example, as shown in the second c-picture, the multi-bit delta-sigma modulator 2〇〇c has a feedforward path around the filter circuit 240 and includes at least one integration circuit. 19 1355808 May 4, 2014 Correction Replacement Page The feedforward path described above includes a gainer (gain bl. (8) for receiving the output value of the second summing circuit 23QC (or the rounding value of the filter circuit 24Q) and generating A feed forward ##, and shai two feed # number will then combine the output value of the filter circuit 240 in the third total circuit ms. The front diameter is functionally equivalent to the second (four) towel ^ yuan ^ analog converter The regional feedback path. The feedforward path can also bypass the first-integral band 220 or other integration circuit (not shown). The third figure reveals the ternary digits on the full-differential circuit. Analog converter 300 - embodiment. The ternary analog converter 3 receives a 2-bit ternary signal (or (4) DATA) and outputs - the corresponding differential signal, the three in the three The seam value _, 丨, 2), and the differential signal includes a forward end (four) rib e end, V 〇 UTP) and a negative end (clear (10) _ vo_. The ternary digit analog converter 300 uses three reference voltages: V_, 'VCM and face, the positive end of the differential money - "fine three kinds of = pressure, and the negative of the differential signal The end is connected to the ^^131°' control through the second multiplexer 32, and the three-reader ratio converter 3_wheeling is the voltage difference between the forward end and the negative end (ie, the multiplexer 31〇' 32 The individual outputs of the ) are analogous to the transition of (), and the three possible outputs of the three domain are generated at equal intervals. Therefore, the ternary analog converter 300 is linear in nature. Qing / for

VOUTP VREFNVOUTP VREFN

DAC OUTPUTDAC OUTPUT

VOUTN VREFN-VREFPVOUTN VREFN-VREFP

2020

100年5月4日修正替換頁 2 VREFP ———-— VREFN ------—_ VREFP-VREFN 表格一 1355808 、第四圖中揭露—離散時間積分器之實施例,該離散時間積 的可以全差較赋電容積分料實作。_缝人電容Correction Replacement Page 2 May 4, 100 VREFP —————— VREFN ------—VREFP-VREFN Table 1135808, Rev. 4, Example of Discrete Time Integrator, Discrete Time Product The difference can be compared with the capacitance integral material. _Sewing capacitor

Cm2)係雛於全差動運算放大器侧的輸人端,而—對反饋電 容(Cfbl,Cfb2)_於全差動運算放大器棚的輸出端,由複數 個時脈信號來控概數個關(switeh)以達_散時間積分之 功能。 第五圖係揭露具備-加總電路功能之積分電路的實施例, 其中該積分電路係使用-離散時間積分器實作。如圖五中所示 ,位在積分電路之前的加總電路係配絲從—數位類比轉換器 接收一反饋信號,該加總電路可有效率地與積分器結合至一單 路全差動父換式電容電路。例如,第—加總電職⑽和第一積 分器220可互相結合於一單路全差動交換式電容電路上,其中該 單路全差動交換式電容電路藉由一第一對輸入電容(即cini, (:1112)來接_變輸人信號以及藉由—第二對輸人電容(即⑽, Cin4)來接收由數位類比轉換器產生的第一反饋信號。第二加總’ 電路230B也可與在濾波電路240上的第二積分器結合於一單路 全差動交換式電容電路上,其中該單路全差動交換式電容電路 藉由一第一對輸入電容來接收第一積分信號、藉由一第二對輸 入電谷末接收第一反饋彳§號以及藉由一第三對輸入電容來接收 第三反饋信號。一實施例中,上述交換式電容電路可藉由設定 各對輸入電容之電容比來將縮放參數器275B包括在内。 21Cm2) is the input end of the fully differential op amp side, and the feedback capacitor (Cfbl, Cfb2) is at the output of the fully differential op amp shed, and is controlled by a plurality of clock signals. (switeh) to achieve the function of time-integration. The fifth figure discloses an embodiment of an integrating circuit having a summing circuit function, wherein the integrating circuit is implemented using a discrete time integrator. As shown in FIG. 5, the summing circuit in front of the integrating circuit receives a feedback signal from the digital-to-digital converter, and the summing circuit can be efficiently combined with the integrator to a single-channel fully differential parent. Switching capacitor circuit. For example, the first-plus-electricity (10) and the first integrator 220 can be combined with each other on a single-channel fully differential switched capacitor circuit, wherein the single-channel fully differential switched capacitor circuit has a first pair of input capacitors (ie cini, (:1112) to receive the _ change input signal and by the second pair of input capacitors (ie (10), Cin4) to receive the first feedback signal generated by the digital analog converter. The second summation' The circuit 230B can also be combined with a second integrator on the filter circuit 240 on a single fully differential switched capacitor circuit, wherein the single full differential switched capacitor circuit is received by a first pair of input capacitors The first integrated signal receives the first feedback 彳§ by a second pair of input valleys and receives the third feedback signal by a third pair of input capacitors. In an embodiment, the switched capacitor circuit can be borrowed The scaling parameter 275B is included by setting the capacitance ratio of each pair of input capacitors.

t而Γ ’以t所述,僅為本發明的具體實施例之詳細說明及圖 = ,本發明之所有範圍應以下述之 ^月,乾圍為準,任何熟悉該項技藝者在本發明之領域内, 範 麵易思及之變化或修飾皆可涵蓋在以下本案所界定之專利 【圖式簡單說明】 :^係為自知—卩衫位兀三角積分調變器之功能方塊圖; 的一實施例;^為本4明之多位7°三角積分調變器之功能方塊圖 的又^實^^料本發明之多位元三_分調變ϋ之功能方塊圖 的再1實施^料本發明之錄几三角積分調變器之功能方塊圖 例示意圖明之電龜式二7°數位類比轉換器的一實施 及第四_為本發明之離觸_分料1施例示意圖;以 圖。第五_4树明之絲賴式電容麵的—實施例示意 【主要元件符號說明】 習知: 一階三角積分調變器100 第一加總電路U〇 22 1355808 100年5月4日修正替換頁 第一積分器120 第二加總電路130 第二積分器140 量化器150 第一數位類比轉換器170 第二數位類比轉換器160 調變輸入信號X/7) 第一反饋信號/;(/?) 第二反饋信號/2(/?) N位元調變輸出信號X/?) 本發明: 200A、200B、200C :多位元三角積分調變器 21.0A、210B :第一加總電路 220 :第一積分電路 230A、230B、230C :第二加總電路 240 :濾波電路 245 :第三加總電路 250 :量化器 260 :第二反饋電路 265 :延遲電路 270 : (N-1)位元數位類比轉換器 275A :微分器 275B :縮放參數器 280 : (N-1)位元加法器 290、300 :三元數位類比轉換器 1355808 100年5月4曰修正替換頁 293 : 1位元加法器 295A、295B :第一反饋電路 /(/?):調變輸入信號 7*1(77)、尸1(77):第一中介信號 7*2(77):第二中介信號 /ι(/7)、尸1(77):第一積分信號 /2 ( η):濾波信號 y(77) : Ν位元調變輸出信號 aO?):第一反饋信號 a、(/7) ' 0.53” 2(77-1):第二反饋信號 a” 2(77-1):延遲類比輸出 <33(7?):第三反饋信號 S(/7):加法信號 C(/7) : 1位元的進位信號 5(/7-/):延遲加法信號 32(77):類比輸出 d(n):三元信號 yz(/7) : N位元調變輸出信號之(N-1)較低有效位元 M;?) : N位元調變輸出信號的最高有效位元 310、320 :多工器 400 :全差動運算放大器And the description of the specific embodiments of the present invention is only a detailed description of the specific embodiments of the present invention and the drawings. All the scope of the present invention shall be based on the following, and the following is applicable to those skilled in the art. In the field, the changes or modifications of Fan Fanyi can be covered in the following patents defined in this case [Simple Description]: ^ is a functional block diagram of the self-knowledge - 卩 兀 position 兀 triangle integral modulator; An embodiment of the functional block diagram of the multi-bit 7° delta-sigma modulator of the present invention is further implemented in the functional block diagram of the multi-bit three-distribution variable of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 3 is a schematic diagram of a functional block diagram of a three-dimensional digital converter of the present invention. FIG. Figure. Fifth _4 Shuming's silk-type capacitive surface - an embodiment of the schematic [main component symbol description] Conventional: First-order triangular integral modulator 100 First totaling circuit U〇22 1355808 Correction replacement on May 4, 100 Page first integrator 120 second summing circuit 130 second integrator 140 quantizer 150 first digital analog converter 170 second digital analog converter 160 modulated input signal X / 7) first feedback signal /; ( / ?) Second feedback signal / 2 (/?) N-bit modulation output signal X /?) The present invention: 200A, 200B, 200C: multi-bit triangular integral modulator 21.0A, 210B: the first total circuit 220: first integrating circuit 230A, 230B, 230C: second summing circuit 240: filter circuit 245: third summing circuit 250: quantizer 260: second feedback circuit 265: delay circuit 270: (N-1) bit Meta-digital analog converter 275A: differentiator 275B: scaling parameterizer 280: (N-1) bit adder 290, 300: ternary digit analog converter 1355808 May 4, 2014 Correction replacement page 293: 1-bit Adder 295A, 295B: first feedback circuit / (/?): modulation input signal 7 * 1 (77), corpse 1 (77): first intermediate signal 7 * 2 (77): Two intermediate signals / ι (/7), corpse 1 (77): first integral signal / 2 ( η): filtered signal y (77): Ν bit modulated output signal aO?): first feedback signal a, (/7) '0.53' 2(77-1): second feedback signal a" 2(77-1): delay analog output <33(7?): third feedback signal S(/7): addition signal C(/7) : 1-bit carry signal 5 (/7-/): delayed addition signal 32 (77): analog output d(n): ternary signal yz (/7): N-bit modulation output (N-1) lower effective bit M of the signal; ?) : most significant bit of the N-bit modulated output signal 310, 320: multiplexer 400: fully differential operational amplifier

Cfbl、Ofb2 :反饋電容Cfbl, Ofb2: feedback capacitor

Cinl、Cin2第一對輸入電容Cinl, Cin2 first pair of input capacitors

Cin3、Cin4第二對輸入電容 24Cin3, Cin4 second pair of input capacitors 24

Claims (1)

1355808 _ 100年5月4曰修正替換頁 十、申請專利範圍: 1、 一種三角積分調變器,包含有: -第-加總電路’係依據由一調變輸入信號及一第一反饋信 號所加=而*之—第—組錢來產n中介信號; • 一第積刀電路’係用以對該第-中介信號進行積分以產生 : 一第一積分信號; -第路’係依據該第一積分信號及一第二反饋信號 的權重加總所得之一第二組信號來產生一第二中介信號; ϋ電路”以處理該第二中介信號以產生—遽、波信 號’。其中該據波器電路包含有_第二積分電路; 里化裔係用以對該據波信號進行量化,以產生一 Ν位元 調變輸出信號;以及 -反饋電珞Hx接收該Ν位元調變輸出信號,並使用一 第了數位類比轉換器來產生該第-反饋信號及-(Ν-1) 位元數位類比轉換器來產生該第二反饋信號。 2、 如申請專利範圍第4所述之調變器,其中該反饋電路更包含 有: - 一(Ν_1)位元累加器,用以接收該Ν位元調變輸出信號中之 ' (Ν—丨)較低有效位元,以產生一(Ν-1)位元累加信號及 一 1位元的進位信號,其中該(.丨)位元累加信^係耦 接至該(Ν-1)位元數位類比轉換器之一輸入端;以及、 一加法器,係用以接收該Ν位元調變輸出信號之最高有次位 元(most significant bits)及該進位信號,以產生一力绅 號’其中該加總信號係搞接至該第一數位類比轉換哭、^ 25 13558081355808 _ May 2014 4 曰Revision and replacement page X. Patent application scope: 1. A triangular integral modulator, comprising: - a first-plus total circuit is based on a modulated input signal and a first feedback signal Adding = and * - the first group of money to produce n intermediate signals; • a first knife circuit 'is used to integrate the first intermediate signal to produce: a first integrated signal; - the first road based on The weights of the first integrated signal and the second feedback signal are summed to obtain a second intermediate signal to generate a second intermediate signal; the circuit is configured to process the second intermediate signal to generate a chirp, a wave signal. The data circuit includes a second integration circuit; the chemistry system is used to quantize the data signal to generate a 调 bit modulation output signal; and the feedback power 珞 Hx receives the Ν bit 调Varying the output signal and using a digital analog converter to generate the first feedback signal and a -(Ν-1) bit digital analog converter to generate the second feedback signal. 2. For example, claim 4 Demodulator, wherein the feedback The circuit further includes: - a (Ν_1) bit accumulator for receiving the '(Ν-丨) lower significant bit in the 调 bit modulated output signal to generate a (Ν-1) bit An accumulation signal and a 1-bit carry signal, wherein the (.丨) bit accumulation signal is coupled to one of the input terminals of the (Ν-1)-bit digital analog converter; and, an adder, The highest significant bits of the 调 bit modulated output signal and the carry signal are used to generate a force ' ', wherein the sum signal is connected to the first digital analog to convert the cry , ^ 25 1355808 100年5月4日修正替換頁 輸入端。 其中該第二組信號更包 如申請專利範圍第1項所述之調變器 含一第三反饋信號。 一第二數位類比 並產生該第三反 如申請專利範圍第3項所述之調變器,更包含 轉換器,係用以接收該N位元調變輪出信號’ 饋信號。 °〜’ 如申請專利範圍第1項所述之調變器,苴 h ^^咕 ,、甲έ亥濾波電路更包含 第二加總電路,係依據一第三組信號之栺 · Qnm\ A-b ^ t 羅重和(weighted sum)來產生一第三中介信號,其中該 -弁八$ , 乐—級信號包括有該第 一矛只刀电路之一輸入及一輸出。 j請專利範圍第丨項所述之調麵,其中轉波電路更包含 一弟二加總電路,係依據一第三組信號之權重和來產生一第二 中介信號,其中該第三組減包括有該第, ^ 及該第二加總電路之—輸出。 ^之粉出 7、如申請專利範圍第!項所述之調變器,其中該第一積分電路與 該第二積分電路皆為-離散時間積分器,而該離散時間積分器 (differential switch-capacitor^ 積分器二且該差動交換電容積分器包含有一對輸入電容器,一 差動運算放大益,-對反饋電容器及複數個由複數個時派信號 所控制之開關。 26 1355808Corrected the replacement page input on May 4, 100. The second group of signals further includes a third feedback signal as described in claim 1 of the scope of the patent application. A second digital analogy produces the third inverter as described in claim 3, and further includes a converter for receiving the N-bit modulated round-out signal. °~' If the modulator described in the first application of the patent scope, 苴h ^^咕, the Jiahaohai filter circuit further includes a second summing circuit, based on a third set of signals 栺· Qnm\ Ab ^ t weighted sum to generate a third intermediate signal, wherein the - 弁 eight $, the music-level signal includes one of the input and an output of the first lance circuit. j. Please refer to the adjustment surface described in the third paragraph of the patent scope, wherein the hopping circuit further comprises a second two-plus circuit, which generates a second intermediate signal according to the weight of the third group of signals, wherein the third group is subtracted Including the first, ^ and the second summing circuit - output. ^The powder out 7. If you apply for the patent scope! The modulator according to the item, wherein the first integration circuit and the second integration circuit are both - discrete time integrators, and the discrete time integrator (differential switch-capacitor^ integrator 2 and the differential exchange capacitance integral The device includes a pair of input capacitors, a differential operation to amplify, - a feedback capacitor and a plurality of switches controlled by a plurality of time signals. 26 1355808 日修正替換頁 ,8、如申請專利範圍第1項所述之調變器,其中該第一積分電路與 §亥第一積分電路皆為一連續時間積分器。 .9、如帽專概®第1賴述之調賴,其巾域波電路更包含 . 有至少—額外加總電路及至少一額外積分電路,以增加該三角 積分調變器之階數(order)。 10、 如中請專賴圍第1項所述之調魏,其中該第—數位類比 轉換器係被實現在一差動電路拓樸(T〇p〇1〇gy)上,該' 路拓樸包含有-第一 3對i (3_ΐ〇_υ多工器,一第二 (3-to-l)多工器,一 2位元的輪入信號及三個參考電壓,其 :’該第-3對1 (3-t〇-l)多工器係用以產生一差動^號^ 上正向端’該第二3對1 (3-to-l).多工器係用以產生—差動 信號之一負向端,該2位元的輸入信號係用以作為該第—數位 類比轉換器中之該第一 3對1與該第二3對丨多工器之選擇 線,該三個參考電壓係被用以作為該第一 3對1與該第二3對 1夕工态之輸入端的輸入信號。 11、 如申請專利範圍第1項所述之調變器,其中該第一加總電路 於—單路差動交換電容電路中與該第—積分電路相結^,其 中,該單路差動(single differential)交換電容電路包含 有—耦接於該調變輸入信號之第一對電容,一耦接於該第一數 位類比轉換器之一輸出端的第二對電容,一對反饋電容,一差 動運昇放大器,及複數個由複數個時脈信號所控制之開關。 27 12The modulating device of claim 1, wherein the first integrating circuit and the first integrating circuit are both a continuous time integrator. .9, as in the Cap's 1st review, the towel domain circuit is further included. There are at least - an additional summing circuit and at least one additional integrating circuit to increase the order of the delta-sigma modulator ( Order). 10. For example, please refer to the adjustment mentioned in item 1, where the first-to-digital analog converter is implemented on a differential circuit topology (T〇p〇1〇gy). Park contains - the first 3 pairs of i (3_ΐ〇_υ multiplexer, a second (3-to-l) multiplexer, a 2-bit round-in signal and three reference voltages, which: 'The The -3 to 1 (3-t〇-l) multiplexer is used to generate a differential ^ ^ upper forward end 'the second 3 to 1 (3-to-l). multiplexer system Generating a negative signal of one of the differential signals, the input signal of the two bits being used as a selection of the first three-to-one and the second three-pair multiplexer in the first-digital analog converter a line, the three reference voltages are used as input signals of the input ends of the first three-to-one and the second three-to-one-single state. 11. The modulator according to claim 1, The first summing circuit is coupled to the first integrator circuit in a single differential switching capacitor circuit, wherein the single differential switched capacitor circuit includes - coupling to the modulation The first pair of capacitors of the input signal, one coupled One output of the first digital to analog converter of the second capacitor, a feedback capacitor, a differential operational amplifier liters, and a plurality controlled by a clock signal of a plurality of switches. 2712 =請專利範圍第i項所述之調變器,I中 ;早路差動交換電容電路加〜、電路 Ϊ二組錢之權重和係、藉由設Γ-位目結合,且該 电路中之電容比值來實現。 、^早路差動交換電容 13 在多位元三角積分 5玄方法包含有: 於一第 之方法,其中 調變器中抑制電路誤差 組信號上施行一第一權 ,,號ί括有—調變輸人信號及-第-反饋信^⑼一組仏 對5玄第一中介信號予以進行積分以產生一第: 於一第二組信號上施行-第二權重加法以=積ft 反饋信號; ㈣包括有該第―積分信號及—第二 對該第二中介賴扣進行驗來產生 對該遽波信號予以量細產生 =㈣, 對該Ν位摘變輸·號 ^調讀出信號; 運算以產生一_ 又氐有欢位元施行一累加 將該Ν位元調變輪出^、f SUm)信號及-進位信號; 相加,以產生—加總錢取=效位元與該進位信號予以 透=數位類比轉換器將該加總信號轉換成該第一反饋信 更包含有: 14、如申請專利範圍第13項所述之方法, 28 1355808 ,_ • . 100年5月4日修正替換頁 . 透過一反饋路徑來將該(N-1)位元的和信號轉換成該第二反 饋信號,其中該反饋路徑包含有一(N-1)位元的數位類 比轉換器。 ' 15、如申請專利範圍第13項所述之方法,其中對該N位元調變輸 - 出信號之(N-1)較低有效位元施行一累加運算之步驟更包含有: 透過一(N-1)位元的加法器來將該N位元調變輸出信號之 (N-1)較低有效位元與該(N-1)位元的和信號予以相加; 以及 延遲該(N-1)位元的加法器之一輸出以產生該(N-1)位元 的和信號。 16、 如申請專利範圍第13項所述之方法,其中該第二組信號更包 含有一第三反饋信號,且該第三反饋信號係為透過該N位元調 變輸出信號之一數位類比轉換而產生。 17、 如申請專利範圍第13項所述之方法,其中該對該第二中介信 號予以進行濾波之步驟更包含有: . 對該第二中介信號予以進行積分。 29 1355808 100年5月4日修正替換頁 .七、指定代表圖: (一) 本案指定代表圖為:第(二A)圖。 (二) 本代表圖之元件符號簡單說明: (本案代表圖為步驟流程圖,故無元件代表符號) 200A :多位元三角積分調變器 210A :第一加總電路 220 :第一積分電路 230A:第二加總電路 240 :濾波電路 250 :量化器 260 :第二反饋電路 265 :延遲電路 270 ·· (N-1)位元數位類比轉換器 275A :微分器 280 : (N-1)位元加法器 290 :三元數位類比轉換器 293 : 1位元加法器 295A :第一反饋電路 X/7):調變輸入信號 r2(/?):第二中介信號 /2(7?):濾·波信號 y(/7) : Ν位元調變輸出信號 ^(/7):第一反饋信號 i 2(77):第二反饋信號 <93(77):第三反饋信號 5 1355808 J * ., ,_ 100年5月4日修正替換頁 .S〔/7):加法信號 C(/7) : 1位元的進位信號 .5(77-/):延遲加法信號 32(7?):類比輸出 ' d(n):三元信號 八、本案若有化學式時,請揭示最能顯示發明特徵的化學式:= Please refer to the modulator described in item i of the patent range, I; the early differential exchange capacitor circuit plus ~, the circuit Ϊ two groups of weights and the system, by setting the Γ-bit combination, and in the circuit The capacitance ratio is achieved. , ^ Early differential exchange capacitor 13 in the multi-element triangular integral 5 Xuan method includes: In the first method, in the modulator to suppress the circuit error group signal to perform a first right, the number includes - The modulated input signal and the -first-feedback signal (9) are grouped to integrate the 5th first intermediate signal to generate a first: a second set of signals is applied - the second weighting is added to the = ft feedback signal (4) including the first-integrated signal and the second detecting the second intervening buckle to generate a fine-grained output of the chopping signal = (4), and extracting the signal from the de-emphasis The operation is to generate a _ and the 欢 氐 施 施 施 施 施 累 累 累 累 累 累 累 累 累 累 累 累 累 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 The carry signal is converted into the first feedback signal by the trans-digital analog converter. The method further includes: 14. The method according to claim 13 of the patent application, 28 1355808, _ • . Correction of the replacement page on the 4th. The sum of (N-1) bits is passed through a feedback path. It converts the number into a second feedback signal, wherein the feedback path comprises an (N-1) bit digital converter than class. 15. The method of claim 13, wherein the step of performing an accumulation operation on the (N-1) lower significant bit of the N-bit modulated output signal further comprises: (N-1) a bit adder to add the (N-1) lower significant bit of the N-bit modulated output signal to the sum signal of the (N-1) bit; and delaying the One of the (N-1) bit adders is output to generate the sum signal of the (N-1) bit. The method of claim 13, wherein the second group of signals further comprises a third feedback signal, and the third feedback signal is a digital analog conversion through the N-bit modulated output signal. And produced. 17. The method of claim 13, wherein the step of filtering the second intermediate signal further comprises: integrating the second intermediate signal. 29 1355808 May 4, 100 revised replacement page. 7. Designated representative map: (1) The representative representative of the case is: (2A). (2) A brief description of the symbol of the representative figure: (The representative figure in the present case is a step flow chart, so there is no component representative symbol) 200A: Multi-bit triangular integral modulator 210A: First summing circuit 220: First integrating circuit 230A: second summing circuit 240: filter circuit 250: quantizer 260: second feedback circuit 265: delay circuit 270 · (N-1) bit digital analog converter 275A: differentiator 280: (N-1) Bit adder 290: ternary analog converter 293: 1-bit adder 295A: first feedback circuit X/7): modulation input signal r2 (/?): second intermediate signal /2 (7?) : Filter wave signal y (/7): Ν bit modulation output signal ^ (/7): first feedback signal i 2 (77): second feedback signal < 93 (77): third feedback signal 5 1355808 J * ., , _ May 4, 100 revised replacement page. S [/7): addition signal C (/7): 1-bit carry signal. 5 (77- /): delayed addition signal 32 ( 7?): Analog output 'd(n): ternary signal 8. If there is a chemical formula in this case, please reveal the chemical formula that best shows the characteristics of the invention:
TW96140230A 2006-10-27 2007-10-26 Method and apparatus to reduce internal circuit errors in a multi-bit delta-sigma modualtor TW200828825A (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US86313406P 2006-10-27 2006-10-27

Publications (2)

Publication Number Publication Date
TW200828825A TW200828825A (en) 2008-07-01
TWI355808B true TWI355808B (en) 2012-01-01

Family

ID=44817780

Family Applications (1)

Application Number Title Priority Date Filing Date
TW96140230A TW200828825A (en) 2006-10-27 2007-10-26 Method and apparatus to reduce internal circuit errors in a multi-bit delta-sigma modualtor

Country Status (1)

Country Link
TW (1) TW200828825A (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9225349B2 (en) * 2013-08-20 2015-12-29 Skyworks Solutions, Inc. Dither-less multi-stage noise shaping fractional-N frequency synthesizer systems and methods

Also Published As

Publication number Publication date
TW200828825A (en) 2008-07-01

Similar Documents

Publication Publication Date Title
TWI360955B (en) Sigma-delta modulator with dac resolution less tha
JP4890503B2 (en) Delta-sigma modulator
JP5696508B2 (en) ΔΣ modulator and signal processing system
JP2994497B2 (en) D / A converter DC offset calibration method and D / A converter DC offset calibration system
TWI255097B (en) Variable order delta- sigma modulator and DA converter
US5061925A (en) Phase equalization system for a digital-to-analog converter utilizing separate digital and analog sections
CN101882931B (en) Deltasigma modulator
US20180063644A1 (en) Digital Silicon Microphone with Configurable Sensitivity, Frequency Response and Noise Transfer Function
US8264390B2 (en) Conversion of a discrete time quantized signal into a continuous time, continuously variable signal
TWI232639B (en) Sigma-delta data conversion employing quantization error feedback
EP2041875A1 (en) Sigma-delta modulation with offset
JP2006514479A (en) Nonlinear filter
JP2005507599A (en) Incremental delta analog / digital converter
JP2012160816A (en) Δς modulator and signal processing system
US20140159929A1 (en) Delta-modulation signal processors: linear, nonlinear and mixed
TW200929966A (en) Circuit and method for calibrating direct current offset
US9204215B2 (en) One-bit digital-to-analog converter offset cancellation
CN110249533A (en) Signal processing apparatus and signal processing method for Hall sensor
TWI355808B (en)
CN109889203B (en) Semiconductor device and method of operating the same
TWI523413B (en) System and method for amplifying a digital input signal to generate an analog output signal
TWI362188B (en) Adaptive dynamic range control
US20180167081A1 (en) Digital to analogue conversion
WO2009034494A1 (en) Adjustable-resistor array type circuit of a semi-digital ratiometric finite impulse response digital-to-analog converter (firdac)
TW513861B (en) Improved third order sigma-delta modulator