201210189 六、發明說明： 【發明所屬之技術領域】 β本發明係有關於-種用於觸控感測器的滤波器，特別 是有關於—制於職❹指之移除纽㈣與高頻雜訊 的濾波器及其方法。 【先前技術】 目前觸控感測控制裝置已普遍使用於許多電子裝置， 例如行動電話、ΜΡ3以及個人數㈣理料。觸控 感測裝置可設置於顯示螢幕，例如液晶顯示螢幕的前面， =以使用者可藉由觸碰特定顯示的選單的一個位置選擇特 疋功能。 、‘ 、有許多種型態的觸控感測控制裝置是使用電容感測器 去感應身體部位’例如手指的存在。雖賴控感測電容 :器已經报成功地應用於許多裝置，然而在某些場合二 些裝置可能报難偵測電荷變化去感應身體部位。舉例來 說、’存在於裝置的雜訊可能混亂電容電荷鍵盤的電荷數量 的測量精確度。 =傳統上液晶顯示器切換以及更新像素時會有切換雜 =。其他形式的顯示器可能在掃描或更新像素時有其他型 態的雜訊。 丄，則技術的電容感測器採用各種信號處理技術從想要 / ^被中濾'除雜訊。原則上，可能使用標準it波器，例如 3^車寺宇^廣、、古-JL- ^、向通濾波器、低通濾波器以及帶通濾波器。 基於雜訊強度有需要量身訂做濾、波技術。舉例來說， 201210189 依據雜訊的數量及特性可能需要切換某些濾波器開與關， 或基於雜訊狀況設定濾波器參數。 因此，有必要提供一種用於移除不想要的直流成份以 及雜訊，例如特定高頻突波的電容式觸控感測器的濾波器。 【發明内容】 本揭露提供一種濾波器系統，適用於移除一觸控感測 信號的直流成份以及高頻雜訊成份，包括：一類比數位轉換 器，用於轉換該觸控感測信號為一離散時間輸入信號；一 • 直流移除單元，用於移除該離散時間輸入信號的直流成份 以得到一微分信號；一解雜訊單元，用於移除該微分信號 的雜訊成份以得到一解雜訊信號；以及一重建單元，用於 處理該解雜訊信號以得到一輸出信號。 本揭露另提供一種濾波方法，適用於移除觸控感測信 號的直流成份以及高頻雜訊成份，包括:藉由一類比數位轉 換器轉換該觸控感測信號成為一離散時間輸入信號；藉由 一直流移除單元移除該離散時間輸入信號的直流成份以得 • 到一微分信號；藉由一解雜訊單元移除該微分信號的雜訊 成份以得到一解雜訊成分；藉由一重建單元處理該解雜訊 信號以得到一輸出信號。 【實施方式】 為使本發明之上述目的、特徵和優點能更明顯易懂， 下文特舉較佳實施例，並配合所附圖式，作詳細說明如下: 第1圖顯示發明的濾波器系統的示意圖。濾波系統100 包括類比數位轉換器110以及數位信號處理單元120。數 201210189 位信號處理單元120包括直流移除單元122、解雜訊單元 124以及重建單元126。數位信號處理單元120可能是數位 信號處理器，但不限於此。 第2圖係說明發明的濾波器系統的數位輸入波形與數 位輸出波形的離散時間波形。參考第1圖，來自觸控面板 的信號是觸控感測器所感應的類比信號，且稱為觸控感應 信號。類比數位轉換器110轉換觸控感應信號成為數位信 號。數位信號是第2圖所示的ADC輸出信號，也是離散時 間信號。ADC輸出信號可能包括影響信號品質的的雜訊成 份，或可能是ADC輸出信號中的不重要部份但佔據大量濾 波系統的儲存容量的直流成份。數位信號處理單元120必 須濾除ADC輸出信號的這些成份。第2圖所顯示的濾波器 輸出信號是沒有DC成份與雜訊成份的乾淨信號。這個乾 淨信號清楚顯示觸控狀態，因此有助於MCU(微控制單 元）130執行後續處理。 第3圖係根據直流移除方法說明發明的濾波器系統的 直流移除單位的數位輸入波形與數位輸出波形的離散時間 波形。直流移除單元122用於移除離散時間ADC輸出信號 或離散時間輸入信號IN[N]的直流成份。在一實施例中， 微分方法是用於移除直流成份。微分方法將在目前取樣時 間的取樣值減去在前一取樣時間的取樣值。此方法可表示 成方程式： X[N]=IN[N]-IN[N-1]， 其中X[N]是在取樣時間N的微分訊號，且IN[N]是在 取樣時間N的離散時間輸入信號。如第3圖所示，微分信 201210189 號X[N]沒有DC成份的離散時間信號。 第4圖係根據解雜訊方法說明發明的滤波器系統的解 雜訊單it的數位輸人波形與數位輪出波形的離散時間波 形。解雜訊單元m狀雜微分信號·的高頻率雜 訊。於-實施例中，解雜訊方法科移除高頻率雜訊。高 頻率雜訊是定義成依附於穩定信號上的不想要的突波。解 雜訊方㈣概念在於當在目餘賴分信號的第一 =樣值相反於在後-個取樣時間的微分信號的第二取樣值 時，互相關或縮減第—取樣值及$二取樣值。 Γ之反的成雙成對的取樣值會互相抵銷 的解雜訊方法描述如p 更詳細 先第ΙΓΓΓ雜訊單元的解雜訊方法的實施例。首 先，二個方程式定義如下： 自 Y[N]=X[N] - Y[N-1], Z[N]=Y[N] — X[N+1]， 其中耶财取樣的微分跡z同是在取樣 時間N的解雜訊信號，且γ「Νι是镇 凡明疋“信號。運算子符號”―，， 疋義為：當一第一運算元A與一第_ '蓄狀一 ^ 弟〜運鼻元B大於或等於201210189 VI. Description of the invention: [Technical field to which the invention pertains] β The present invention relates to a filter for a touch sensor, and more particularly to the removal of a button (four) and a high frequency Noise filter and its method. [Prior Art] At present, touch sensing control devices have been commonly used in many electronic devices, such as mobile phones, ΜΡ3, and personal number (4) materials. The touch sensing device can be placed on the display screen, such as the front of the liquid crystal display screen, = the user can select a feature by touching a location of the particular displayed menu. ‘ 、 There are many types of touch sensing control devices that use capacitive sensors to sense the presence of body parts such as fingers. Although the sensing capacitors have been successfully applied to many devices, in some cases, some devices may report difficulty in detecting charge changes to sense body parts. For example, the noise present in the device may confuse the measurement accuracy of the amount of charge on the capacitive charge keyboard. = Traditionally, when the LCD is switched and the pixels are updated, there will be a switch. Other forms of display may have other types of noise when scanning or updating pixels. Alas, the technology's capacitive sensors use a variety of signal processing techniques to remove noise from the desired / ^ being filtered. In principle, it is possible to use standard iterators, such as 3^车寺宇广, 古-JL-^, pass-through filters, low-pass filters, and band-pass filters. Based on the noise intensity, it is necessary to tailor the filter and wave technology. For example, 201210189 may need to switch some filters on and off depending on the number and characteristics of noise, or set filter parameters based on noise conditions. Therefore, it is desirable to provide a filter for a capacitive touch sensor that removes unwanted DC components and noise, such as specific high frequency surges. SUMMARY OF THE INVENTION The present disclosure provides a filter system for removing a DC component and a high frequency noise component of a touch sensing signal, including: an analog-to-digital converter for converting the touch sensing signal to a discrete time input signal; a DC removal unit for removing a DC component of the discrete time input signal to obtain a differential signal; and a noise cancellation unit for removing the noise component of the differential signal to obtain a noise cancellation signal; and a reconstruction unit for processing the noise cancellation signal to obtain an output signal. The disclosure further provides a filtering method, which is suitable for removing a DC component and a high frequency noise component of the touch sensing signal, comprising: converting the touch sensing signal into a discrete time input signal by using an analog converter; Removing a DC component of the discrete-time input signal by a DC removal unit to obtain a differential signal; removing a noise component of the differential signal by a noise cancellation unit to obtain a noise cancellation component; The de-noise signal is processed by a reconstruction unit to obtain an output signal. The above described objects, features and advantages of the present invention will become more apparent from the following description of the preferred embodiments of the invention. Schematic diagram. The filtering system 100 includes an analog digital converter 110 and a digital signal processing unit 120. The number 201210189 bit signal processing unit 120 includes a DC removal unit 122, a de-noise unit 124, and a reconstruction unit 126. The digital signal processing unit 120 may be a digital signal processor, but is not limited thereto. Figure 2 is a diagram showing the discrete time waveforms of the digital input waveform and the digital output waveform of the inventive filter system. Referring to Figure 1, the signal from the touch panel is an analog signal sensed by the touch sensor and is called a touch sensing signal. The analog digital converter 110 converts the touch sensing signal into a digital signal. The digital signal is the ADC output signal shown in Figure 2 and is also a discrete time signal. The ADC output signal may include a noise component that affects signal quality, or a DC component that may be an unimportant part of the ADC output signal but that occupies a large amount of filtering system storage capacity. Digital signal processing unit 120 must filter out these components of the ADC output signal. The filter output signal shown in Figure 2 is a clean signal with no DC components and noise components. This clean signal clearly shows the touch state, thus facilitating the MCU (Micro Control Unit) 130 to perform subsequent processing. Figure 3 is a diagram showing the discrete time waveforms of the digital input waveform and the digital output waveform of the DC removal unit of the inventive filter system according to the DC removal method. The DC removal unit 122 is configured to remove the DC component of the discrete time ADC output signal or the discrete time input signal IN[N]. In an embodiment, the differentiation method is for removing DC components. The differential method subtracts the sampled value at the current sampling time from the sampled value at the previous sampling time. This method can be expressed as an equation: X[N]=IN[N]-IN[N-1], where X[N] is the differential signal at sampling time N, and IN[N] is the dispersion at sampling time N Time input signal. As shown in Figure 3, the differential signal 201210189 X[N] has no discrete time signal of DC components. Fig. 4 is a diagram showing the discrete time waveform of the digital input waveform and the digital rounded waveform of the noise signal of the inventive filter system according to the method of denoising. High frequency noise of the m-shaped heterodifferential signal of the noise unit. In an embodiment, the Noise Removal Methodology section removes high frequency noise. High frequency noise is an unwanted glitch that is defined as being attached to a stable signal. The concept of the noise canceling party (4) is to cross-correlate or reduce the first-sampling value and the second-sampling when the first=sample of the target lag signal is opposite to the second sample value of the differential signal at the next-sampling time. value. The anti-noise method of the double-paired sample values offset each other is described in more detail. First, the first embodiment of the noise-cancellation method of the noise unit. First, the two equations are defined as follows: From Y[N]=X[N] - Y[N-1], Z[N]=Y[N] — X[N+1], where the derivative of the sample is taken z is the same as the noise signal at the sampling time N, and γ "Νι is the town of the city" signal. The operator symbol "-," is defined as: when a first operand A and a _ 'storage one ^ brother ~ Yun nose B is greater than or equal to
零或該弟-運异疋A與一第二運算元B小於零，則A --B-A，當該第一運算元a大於零 ^ ,,, 令且該第二運算元B小於 零’二且71'…’則ΠΑ+Β，否則A„B = 〇; 以及虽該第一運算元A小於雯且誃坌 0 , A , . , v 碎第二運算元B大於零， I>|B|，則 A — B=A+B，否則 Α — Β = 0。第 5 201210189 圖表格中的數字是解雜訊單元的解雜訊方法的實施例之範 例。 第6圖係根據重建單元的重建方法說明重建單元的數 位輸入波形與數位輸出波形的離散時間波形。方法的概念 是當輸出信號的取樣值大於第一臨界值且小於第二臨界值 時’使用IIR方程式產生輸出信號的取樣值；否則使用FIR 方程式產生輸出信號的取樣值。當輸出信號的取樣值大於 第一臨界值時意指觸碰行為被感應到’所以使用IIR方法 找出輸出信號的取樣值。相對地，當輸出信號的取樣值小 於第一臨界值時意指觸碰行為沒有被感應到，所以使用fir 方程式偵測輸出信號的取樣值。IIR方程式定義如下： OUT[N]=〇UT[N-l]+Z[N]，其中Z[N]是在取樣時間n的解 雜訊彳§號’且OUT[N]是在取樣時間]Si的輸出信號。fir方 程式定義如下:OUT[N]=Z[N]+Z[N-l]+Z[N-2] + ….Z[N-n]， 其中η是FIR方程式的階數，Z[N]是在取樣時間N的解雜 訊信號，且OUT[N]是在取樣時間N的輸出信號。在發明 的實施例中，FIR方程式的階數n是2，但不限於此。在實 施例中’第一臨界值與第二臨界值是可變的或可調整的。 以此方式，重建單元的輸出取樣值可限制在適當範圍而不 會導致觸控感應信號失真。濾波系統根據第一臨界值可能 偵測觸碰行為是碌觸發。m统可能根據第二臨界值 偵測觸碰行為是否將要完成。 舉例來說，如第6圖所示，假設第一臨界值與第二臨 界值都等於數字B。在取樣時間n·5之前，解雜訊信號z[N] 是零且輸出信號〇UT[N]也是零。在取樣時間η·5，輸出信 201210189 號OUT[N]是40，且低於第一臨界值B，所以使用fir方 程式。在取樣時間n-4 ’輸出信號〇UT[N]是60，且高於第 一臨界值B。因此，在下一取樣時間n_3使用nR方程式。 在取樣時間n+1，輸出信號0UT[N]是4〇且低於第二=界 值B所以在下一取樣時間使用nR方程式。上述僅僅是範 例，但不限於此。 第7圖係說明重建單元的另一重建方法。此方法的概 念是當輸出信號的取樣值小於第三臨界值且符合無接觸函 • 數，使用FIR方程式產生輸出信號的取樣值；否則使用IIR 方程式產生輸出信號的取樣值。IIR方程式定義如下： OUT[N]=OUT[N-l]+Z[N] ’其中Z[N]是在取樣時間n的解 雜訊#说，且OUT[N]是在取樣時間]si的輸出信號。fir方 程式定義如下：〇UT[N]=Z[N]+Z[N-l]+Z[N-2]+....ZfN_nJ， 其中η是FIR方程式的階數，z[N]是在取樣時間N的解雜 訊信號，且OUT[N]是在取樣時間N的輸出信號。無接觸 函數定義如下：OUT[N]+k<OUT[N-l]<〇UT[N]-k&& • OUT[N]+k<OUT[N-2]<OUT[N]-k&&…"&&〇UT[N]+k<OU T[N-M]<OUT[N]-k，其中OUT[N]是在取樣時間N的輸出 信號，k是常數且Μ是無接觸函數的階數。當沒有觸碰行 為觸發或影響觸控感測信號時，觸控感測信號應該是穩定 於一個電壓範圍内。所以根據無接觸函數，濾波系統可確 定沒有更進一步的觸碰行為被觸發。 第8圖係根據發明的實施例說明濾波方法。在步驟 8 〇 2，藉由類比數位轉換器將觸控感應信號轉換成離散時間 輸入信號。在步驟804，藉由直流移除單元將直流成份自 201210189 離散時間輸入信號中移除以得到微分信號。在步驟806， 藉由解雜訊單元將微分信號的雜訊成份移除以得到解雜訊 信號。在步驟808，藉由重建單元處理解雜訊信號以得到 輸出信號。 最後，熟此技藝者可體認到他們可以輕易地使用揭露 的觀念以及特定實施例為基礎而變更及設計可以實施同樣 目的之其他結構且不脫離本發明以及申請專利範圍。 201210189 【圖式簡單說明】 第1圖顯示發明的濾波器系統的示意圖； 第2圖係說明發明的濾波器系統的數位輸入波形與數 位輸出波形的離散時間波形； 第3圖係根據直流移除方法說明發明的濾波器系統的Zero or the younger brother - A and a second operand B are less than zero, then A - BA, when the first operand a is greater than zero ^,,, and the second operand B is less than zero 'two And 71'...' then ΠΑ+Β, otherwise A„B = 〇; and although the first operand A is smaller than Wen and 誃坌0, A, . , v, the second operand B is greater than zero, I>|B |, then A - B = A + B, otherwise Α - Β = 0. The number in the table of 201210189 is an example of an embodiment of the method of denoising the noise cancellation unit. Figure 6 is based on the reconstruction unit The reconstruction method describes the discrete input time waveform of the digital input waveform and the digital output waveform of the reconstruction unit. The concept of the method is to generate the sampling value of the output signal using the IIR equation when the sampling value of the output signal is greater than the first critical value and less than the second critical value. Otherwise, the FIR equation is used to generate the sampled value of the output signal. When the sampled value of the output signal is greater than the first threshold, it means that the touch behavior is sensed 'so the IIR method is used to find the sampled value of the output signal. Relatively, when the output When the sampled value of the signal is less than the first critical value, it means that the touch behavior is not It is sensed, so the fir equation is used to detect the sampled value of the output signal. The IIR equation is defined as follows: OUT[N]=〇UT[Nl]+Z[N], where Z[N] is the rectification at sampling time n彳 § ' and OUT [N] is the output signal at the sampling time] Si. The equation of fir is defined as follows: OUT[N]=Z[N]+Z[Nl]+Z[N-2] + ....Z [Nn], where η is the order of the FIR equation, Z[N] is the de-noise signal at sampling time N, and OUT[N] is the output signal at sampling time N. In an embodiment of the invention, FIR The order n of the equation is 2, but is not limited thereto. In the embodiment, the 'first threshold and the second threshold are variable or adjustable. In this way, the output sample value of the reconstruction unit can be limited to appropriate The range does not cause distortion of the touch sensing signal. The filtering system may detect that the touch behavior is a trigger according to the first threshold. The system may detect whether the touch behavior is to be completed according to the second threshold. For example, As shown in Fig. 6, it is assumed that both the first critical value and the second critical value are equal to the number B. Before the sampling time n·5, the de-noising signal z[N] is zero and the output signal UT[N] is also zero. At the sampling time η·5, the output signal 201210189 OUT[N] is 40 and lower than the first critical value B, so the fir equation is used. At the sampling time n-4 'output signal 〇UT [N] is 60 and is higher than the first critical value B. Therefore, the nR equation is used at the next sampling time n_3. At the sampling time n+1, the output signal OUT[N] is 4 〇 and lower than the second = boundary value B Therefore, the nR equation is used at the next sampling time. The above is merely an example, but is not limited thereto. Figure 7 illustrates another reconstruction method of the reconstruction unit. The concept of this method is to generate a sampled value of the output signal using the FIR equation when the sampled value of the output signal is less than the third threshold and conforms to the contactless function; otherwise the sampled value of the output signal is generated using the IIR equation. The IIR equation is defined as follows: OUT[N]=OUT[Nl]+Z[N] 'where Z[N] is the de-noise # at the sampling time n, and OUT[N] is the output at the sampling time]si signal. The fir equation is defined as follows: 〇UT[N]=Z[N]+Z[Nl]+Z[N-2]+....ZfN_nJ, where η is the order of the FIR equation and z[N] is the sampling The noise signal of time N, and OUT[N] is the output signal at sampling time N. The contactless function is defined as follows: OUT[N]+k<OUT[Nl]<〇UT[N]-k&& • OUT[N]+k<OUT[N-2]<OUT[N]- k&&..."&&〇UT[N]+k<OU T[NM]<OUT[N]-k, where OUT[N] is the output signal at sampling time N, k is a constant And Μ is the order of the contactless function. When no touch action triggers or affects the touch sensing signal, the touch sensing signal should be stable within a voltage range. So according to the no-contact function, the filtering system can determine that no further touch behavior is triggered. Figure 8 illustrates a filtering method in accordance with an embodiment of the invention. In step 8 〇 2, the touch sensing signal is converted into a discrete time input signal by an analog digital converter. At step 804, the DC component is removed from the 201210189 discrete time input signal by the DC removal unit to obtain a differential signal. In step 806, the noise component of the differential signal is removed by the de-noising unit to obtain a noise-cancellation signal. At step 808, the noise cancellation signal is processed by the reconstruction unit to obtain an output signal. In the end, it is obvious to those skilled in the art that they can easily use the concept of the disclosure and the specific embodiments to change and design other structures that can perform the same purpose without departing from the scope of the invention and the scope of the claims. 201210189 [Simple diagram of the diagram] Figure 1 shows a schematic diagram of the inventive filter system; Figure 2 shows the discrete-time waveform of the digital input waveform and digital output waveform of the inventive filter system; Figure 3 is based on DC removal Method illustrating the inventive filter system
直流移除單位的數位輸入波形與數位輸出波形的離散 波形； B ^ 4圖係根據解雜訊方法說明發明的濾波器系統的解 Φ雜訊單元的數位輸入波形與數位輸出波形的離散時間波 形； 第5圖係說明解雜訊單元的解雜訊方法的實施例； 第6圖係根據重建單元的重建方法說明重建單元的數 位輸入波形與數位輸出波形的離散時間波形； 弟7圖係說明重建單元的另一重建方法；以及 第8圖係根據發明的實施例說明濾波方法。 春 【主要it件符號說明】 1 〇〇〜濾、波系統 110〜類比數位轉換器 120〜數位信號處理單元 122〜直流移除單元 124〜解雜訊單元 126〜重建單元 130〜微控制單元 s 11 201210189 802、804、806、808〜方法步驟 12DC removes the discrete waveform of the digital input waveform and the digital output waveform; B ^ 4 shows the discrete-time waveform of the digital input waveform and digital output waveform of the Φ noise unit of the inventive filter system according to the method of de-noising FIG. 5 is a diagram illustrating an embodiment of a method for denoising a noise of a noise canceling unit; FIG. 6 is a diagram illustrating a discrete time waveform of a digital input waveform and a digital output waveform of a reconstruction unit according to a reconstruction method of the reconstruction unit; Another reconstruction method of the reconstruction unit; and FIG. 8 illustrates a filtering method according to an embodiment of the invention. Spring [Main id symbol description] 1 〇〇 ~ filter, wave system 110 ~ analog digital converter 120 ~ digital signal processing unit 122 ~ DC removal unit 124 ~ de-noise unit 126 ~ reconstruction unit 130 ~ micro control unit s 11 201210189 802, 804, 806, 808~ method step 12