201112068 四、指定代表圖： (一) 本案指定代表圖為：第（5 )圖。 (二) 本代表圖之元件符號簡單說明： 50 節點 142 電流源 100 觸控感測裝置 150 I/O介面 120 驅動電路 160 類比數位轉換器 122 PN碼產生器 170 感測電路 124 RDSW產生器 173 信號擷取器 140 電流源 五、 本案若有化學式時，請揭示最能顯示發明特徵的化學 無 予巧· 六、 發明說明： 【發明所屬之技術領域】 本發明係關於觸控感測，明確而言，是有關於可分散各種頻率之雜訊 干擾的觸控感測裝置。 【先前技術】 觸控面板係利用感測陣列來檢測由手指、觸筆之類物品所行之觸碰的 位置與強度。第1圖係顯示具有感測陣列U)之-般觸控感測裝置i (例如 觸控面板）_意圖。該__ H)包含-群縱向導電軌錢—群橫向導 電軌排列成X-Y座標的行與列，或者排列成極座標樣態，在各個交點有感 測元件（未圖示）。舉例而言，感測元件-般實施為電阻器或是電容器。控 制單元12經由多工器16傳送驅動信號以驅動感測陣列㈣列丨。被驅動 201112068 的列i之各行的感測信號係經過多工器14而由控制單元12依序或同時檢測 以判定觸碰位置與強度。藉由檢查感難號的值，可得知觸碰位置與強度。 舉例而言，假設一列有十六個節點（亦即，各列與十六行相交），對於一特 定列的十六個節點的感測信號的信號值為（0,00，1，2,3,43,2，10 0 0 0 〇’ 〇)，這表示第七個節點有較強的觸碰，而，感測元件對雜訊相當敏感。 因此’感測信號的值容易受到影響峨難以财地區分觸碰位置並判斷觸 碰強度。 如今觸控面板等觸控感測裝置已廣泛使用於各種應用並涉及許多複雜 的功能性#作，像是無線通鱗。目此，馳面板可能受到各種雜訊影響， 諸如ι/f雜訊、白色雜訊、電力雜訊、5_赫芝雜訊、微波（例如紅外線、 藍芽等）雜訊、背光雜訊等等。各種雜訊分散在不同的頻帶。第2圖顯示 各種雜訊以及信號如何與此等雜訊麵…上方_顯示諸如職訊23、60 赫芝雜訊25、區域雜訊27以及白色高斯雜訊29等各種雜訊的分佈。直流 信號係由-黑色箭頭21表示。中__示理想的感測信號^下方的圖顯 示輕合雜訊之感測信號一般而言，高頻雜訊可_低通紐器加以滤除。 然而’如果吾人嘗試減止頻率的傾觀賴除較低鮮的雜訊以 擷取直流項（亦即所要的信號），此紐器的響應時間會相當慢。舉例而言， 如果使用ίο赫芝之截止頻率來齡6G赫芝雜訊，則響應_將延遲〇 ι 秒。此種延遲會導致觸控面板操作不便。 在習知調變/_技術中，可顧鮮為fl的賊調變―電壓或電流驅 、號X驅動感測陣觸行與列。而後從感測陣列取得的感測信號係以頻 率為£2的解紙號加以解調。如此，可產生頻率為(行祀)以及(打_印的信 201112068 號。如果低通濾波器具有低於(fl+f2)/2之截止頻率，則可將高頻成份濾除， 而取得低頻成份。當fl=f2，則該低頻成份即為直流項，也就是所要的感測 信號。觸碰事件可從直流項得知《直流項的變化係對應筆因於觸碰的電容 值或電阻值變化。然而，用於調變驅動信號的載波必須選在低雜訊的頻帶。 如果載波是在高雜訊的頻帶，則感測信號的信號雜訊比（SNR)將會劣化。 因此’載波（亦即調變信號）必須選在低雜訊頻帶。為了知道何個頻帶具 有最低的雜訊，需要掃描並檢查所有的頻帶，如此會增加硬體與時間的成 本0 【發明内容】 本發明之一目的在於提供一種觸控感測技術以分散各種頻率的雜訊干 擾。 根據本發明之一特點，一種用於檢測感測陣列之觸碰事件的觸控感測裝 置，包含一驅動電路，提供一隨機脈波長度方波信號來調變一電氣信號以 產生用於驅動該感測陣列之一節點的調變驅動信號；以及一感測電路，從 該感測陣列之該節點測量一感測信號，並利用該隨機脈波長度方波信號來 擷取該節點之觸碰資訊。該隨機脈波長度方波信號具有複數個不同脈波長 度的脈波，以使該調變驅動信號亦具有同樣的不同脈波長度之脈波。 根據本發明之另一特點，一種用於檢測感測陣列之觸碰事件的觸控感測 方法，包含提供一隨機脈波長度方波信號；以該隨機脈波長度方波信號調 變一電氣信號以產生一調變驅動信號，用於驅動該感測陣列之一節點；從 該感測陣列之該節點測量一感測信號；以及利用該隨機脈波長度方波信號 201112068 擷取該㈣之_冑訊。概魏波長度枝域具有減織有不同脈 、長的脈波以使該調變驅動信號也具有同樣的不同脈波長度的複數個 脈波。 為讓本發明之上翻容缺賴練，下文特舉較佳實補，並配合 所附圖式’作詳細說明如下： 【實施方式】 本發明係利用正交向量的特性。假設一向量群的各向量為vi，其中i= 〇，1，…，η。如果一向量與不同向量的乘積為〇 (亦即Vixvj = (),其中 j)，且一向量與自身的乘積為丨（亦即νιχν」· =丨，其中丨=則此為正 交向量群。當 VI = (al，bl，cl，dl)且 V2 = (a2,叹 a，也），則 V1 χ % 的 雜等於al x a2 + bl x b2 + C1 x c2 + dl xd2。舉例而言，如果向量群包括 兩個向量：V1=(0, 0,0，！）以及 wmu)，則滿足 vlxV1 = 1，V1 x V2 = 0 ’以及V2xV2 =卜因此，VI與V2為正交。 任何信號皆可表示為正交向量群，如S = clVl+c2V2+c3V3+...+envn， 其中cl，c2,…，cn為係數。如果環境雜訊表不為n = lOOVi + 5〇V2 + 20V3 + 10V4 + 2V5 + 4V6 + 10V7...，其中各向量V1，V2,…代表一特定頻帶的分 董。對於一已知^號A來說，如果選定V5為調變向量，則經調變的作號 〈亦即輸入信號〉Si = AV5。可知該信號會耦合雜訊，因此輸出信號s〇 = AV5 + 100V1 + 50V2 + 20V3 + 10V4 + 2V5 + 4V6 + 10V7..· = 100V1 + 5〇V2 + 20V3 + 10V4 + (A+2)V5 + 4V6 + 10V7…。如果使用相同的向量V5作為解 調向量，則還原的信號 Sr = So χ V5 = 100 χ 0 + 50 χ 0 + 20 χ 〇 + 1() x 〇 + (Α+2) χ1+4χ〇+1〇χ 〇··. = A+2。 201112068 如果使用兩個不同的向量調變兩個信號，則可藉由利用這兩個不同的向 量作為解調向量榻取這兩個信號。舉例而古， 又5又選疋向量V5來調變信髀 A ’並選細_修來雙錢B，嶋人域為si·.暮 輸入信號會耦合雜訊，故輸出信號為s〇 = 20V3 + 1GV4 + 2V5 + 4V6 + l〇V7 =應 + l〇〇Vl + 50V2 + {A+2)Y5+^+4)V6+l0V7-^m^ + 10V4 + 置采解調輸出信號時，信號A 可還原為 SrA = So X V5 = 100 X 〇 + ν Λ 5〇X〇 + 2〇x〇+l〇x〇 + (A+2)xl + (Β+4) χ 0 + 1〇 χ 〇... = α+2。如果刹田人 θ ” D 1 V6來解調輸出信號時，信號Β 可還原為 SrB = So X V6 = 1〇〇 X 〇 + Λ 〇 + 5〇χ〇 + 2〇χ〇+1〇χ〇 + (Α+2)χ〇 + (Β+4) χ1 + 1〇χ〇.·.= β+4。籍由刹田爻 利用多個不同的向量’可同時處理感測陣 列之多點。稍後將詳細說明。 如所見者’還原的信號僅會留下少許雜訊。然而，如上所提及，為了降 低雜訊，應選定低雜賴分量〈例如本财的Μ〉作為調變與解調向量。 為了避免掃描全部頻帶以找出最低雜訊的頻帶，本發明係利用隨機展頻 (RSS)技術。母個選定用於調變與解調的向量均為複數個頻率的隨機組 〇因此還原的L號不會料特定頻帶之雜訊的嚴重衝擊。較佳而言，選 定的向量係時時改變。舉例而士 。’在時間t1’選定的向量為(1/4)V3+ (1/4)V5 (1/4)V7 (1/4)V8 ’而在時間t2，選定的向量為(I，聊+ (ι/3)ν5 + (寧8。實作上可利用偽隨機雜訊㈣碼技術。 第3圖為顯示三個不同_碼及其功率麟的示賴。各PN碼就像 是-把特定的鑰匙。如所見者，由黑色箭頭表示的三個抓碼之功率分量係 刀散在不同鮮SJ此，可_展躺目的。第4廳顯示根據本發明之 201112068 兩個信號的調變與解調的示意圖。信號八係以碼丄調變，而信號B 係以碼2調變。經調變的信號組合成一組合信號&。信號a可藉由利用碼 1去解調組合信號Sc而從組合信號Sc中還原。信號B可藉由利用碼2去 解調組合信號Sc而從組合信號Sc中還原。 第5圖係顯示根據本發明之觸控感測裝置1〇〇的示意圖。觸控感測裝 置100包含—驅動電路120 ’信號源，諸如用於將感測陣列（未圖示）之一 個電容性節點50充電放電的電流源140、142，I/O (輸入/輸出）介面15〇， 鲁類比數位轉換器（ADC) 160以及感測電路170。在本發明中，驅動電路12〇 具有一偽隨機雜訊（PN)碼產生器122，其產生並提供抓碼。該PN碼係 送置衣隨機脈波長度方波（簡稱為RDSW)產生器124〇RDSW產生器124 根據產生PN碼產生器122所提供的PN碼而產生具有多個脈波的信號，此 等脈波具有不同的脈波長度。 第6圖顯示第5圖之觸控感測裝置所產生的隨機脈波長度方波(rdsw) 信號。如所示，該RDSW信號至少包含八個具有不同脈波長度τ卜丁2、...、 ® T8的多個脈波。該RDSW信號的各個脈波長度係由亂數決定。舉例而言， RDSW 信號 Ti (i = 1 到 8)係依據一串亂數 101、235、76、104、223、 94、160、112產生。該RDSW信號的第一脈波長度T1對應1〇1，第二脈 波長度T2對應235，以此類推。該RDSW信號係用於調變電流源140、142 所提供的電流信號以形成一 RDSW調變驅動信號。於本實施例中，該RDSW 調變驅動信號係經由I/O介面送去驅動該電容性節點50。亦即，該電容性 節點係依據該RDSW調變驅動信號充電/放電。如此項領域所周知，因為觸 碰而造成的電容性節點50的電容改變會反應在感測信號Vin的電壓變化。 201112068 感測信號Vm係經由I/O介面ι5〇從電容性節點5〇測得。該感測信號 Vin為電壓信號。為了以數位形式處理該感測信號，該感測信號vin係 藉由ADC 160轉換成數位。然而，可省略adc 16〇而直接處理類比的感測 信號Vin。感測電路170中的信號擷取器173係利用該RDSW產生器124 所產生的相同RDSW信號擷取前述的電壓變化，此電壓變化表示電容性節 點50的電容改變。信號擷取器173根據該RDSW信號產生一解調信號以解 調該感測信號Vin。因此，該感測電路170可輸出對應該電容改變的電壓變 化資訊’此代表該電容性節點50的觸碰資訊。 第7圖係顯示根據本發明之產生隨機脈波長度方波（rjqsw)信號之方 法的流程圖，該RDSW信號具有各種脈波長度Ti(i = 1到η)的多個脈波。 程序於步驟S10開始，計數i = 1。於步驟S20，產生亂數。於步驟S30，判 定該亂數是否在適當範圍内以使脈波長度落在Tmax與Tmin之間的範圍 内°上限Tmax與下限Tmin係用於限制RDSW信號各脈波長度的範圍。該 ^SW信號最大的脈波長度不能超過Tmax以避免電容節點50被過充電。 而RDSW信號最小的脈波長度應大於Tmin以便能確實檢測到感測信號的 脈波。如果所產生的亂數不在適當範圍，則程序返回步驟S2〇以重新產生 新的亂數。如果所產生的亂數是在適當範圍内，則程序進行至步驟s3〇以 檢查該計數i是否超過一預定數η。若是，則程序結束。若否，則脈波長度 Τι則在步驟S40中根據該亂數決定。此外，該計數i係加1，且該程序返回 步驟S20再次循環。 第8圖顯示根據本發明利用隨機脈波長度方波（RDSW)信號之調變與 解調波形。如所述，RDSW調變驅動信號係藉由第5圖之觸控感測裝置1〇〇 201112068 之驅動電請繼物sw信號調變,，該卿讎用作為 調紅調變電流源所提供的電流信號。當調變驅動信號為高位準時， 該電容性節點5〇被充電。當機驅動信號為低位準時，該電容性節點50 被放電。該感測電請依據細sw信號產生一解調信號，但與該·w 信號有90度的相位移。如所示，解調信號具有與調變信號相同的波形但有 Ti/4之相位延遲。亦即，解調信號的第—方波脈波相對於調變信號延遲 T1/4 ’解調信號㈣二方波脈波相對於觀信號延遲谓1此類推。 由信號齡们73所實行的解調可藉由乘加法實施。舉例而言，可利 用數位信號處理微處理器控制單元（Dsp MCU)應（乘與加）累積器（未 圖示；由乘加指令碼實施)。如果感測信號％為(2, 2 3, % 2 & 3丄％…） 且RDSW域為（m ls !，！，…），則累積的結果 +2.6X (-1)+2. 3x (1)+3.1χ(ΐ)+3.4χ(ΐ)201112068 IV. Designated representative map: (1) The representative representative of the case is: (5). (2) A brief description of the component symbols of the representative figure: 50 node 142 current source 100 touch sensing device 150 I/O interface 120 driving circuit 160 analog digital converter 122 PN code generator 170 sensing circuit 124 RDSW generator 173 Signal Extractor 140 Current Source 5. If there is a chemical formula in this case, please disclose the chemical that can best display the characteristics of the invention. 6. Description of the Invention: [Technical Field of the Invention] The present invention relates to touch sensing, In other words, there is a touch sensing device that can disperse noise interference of various frequencies. [Prior Art] The touch panel uses a sensing array to detect the position and intensity of a touch by an object such as a finger or a stylus. Figure 1 shows the general touch sensing device i (e.g., touch panel) with the sensing array U). The __H) includes a group of longitudinal conductive rails - the group of lateral conductors are arranged in rows and columns of X-Y coordinates, or are arranged in a polar coordinate state, with sensing elements (not shown) at each intersection. For example, the sensing element is typically implemented as a resistor or a capacitor. Control unit 12 transmits drive signals via multiplexer 16 to drive the sense array (4) train. The sensed signals of the rows of the column i that are driven 201112068 are passed through the multiplexer 14 and sequentially or simultaneously detected by the control unit 12 to determine the touch position and intensity. By checking the value of the difficulty number, the touch position and intensity can be known. For example, suppose a column has sixteen nodes (ie, each column intersects sixteen rows), and the signal value of the sensing signal for sixteen nodes of a particular column is (0, 00, 1, 2, 3,43,2,10 0 0 0 〇' 〇), which means that the seventh node has a strong touch, and the sensing component is quite sensitive to noise. Therefore, the value of the sense signal is easily affected, and it is difficult to distinguish the touch position and determine the touch strength. Touch sensing devices such as touch panels have been widely used in various applications and involve many complicated functions, such as wireless scales. For this reason, the Chi panel may be affected by various noises, such as ι/f noise, white noise, power noise, 5_Hezhi noise, microwave (such as infrared, Bluetooth, etc.) noise, backlight noise, etc. Wait. Various noises are scattered in different frequency bands. Figure 2 shows how the various noises and signals are related to these noise levels... Above _ shows the distribution of various noises such as 23, 60 Hertzian 25, Area Noise 27, and White Gaussian Noise 29. The DC signal is indicated by the -black arrow 21. The __ shows the ideal sensing signal ^ The picture below shows the sensing signal of the light and noise. Generally speaking, the high frequency noise can be filtered out by the low pass. However, if we try to reduce the frequency, we will rely on the lower noise to extract the DC term (that is, the desired signal). The response time of this device will be quite slow. For example, if you use the cutoff frequency of ίο赫芝 to age 6G Hertzian noise, the response_ will be delayed by ι ι sec. This delay can cause the touch panel to be inconvenient to operate. In the conventional modulation/_ technology, the thief can be changed to the voltage of the fl. - voltage or current drive, number X drive sense array touch and column. The sensed signal taken from the sense array is then demodulated with a paper release number of £2. In this way, the frequency can be generated (line 祀) and (printed letter 201112068. If the low-pass filter has a cutoff frequency lower than (fl+f2)/2, the high-frequency component can be filtered out and obtained Low-frequency component. When fl=f2, the low-frequency component is the DC term, which is the desired sensing signal. The touch event can be learned from the DC term that the change of the DC term corresponds to the capacitance value of the touch or The resistance value changes. However, the carrier used to modulate the drive signal must be selected in the low noise band. If the carrier is in the high noise band, the signal to noise ratio (SNR) of the sensed signal will be degraded. 'Carrier (that is, modulation signal) must be selected in the low noise band. In order to know which frequency band has the lowest noise, it is necessary to scan and check all frequency bands, which will increase the cost of hardware and time. An object of the present invention is to provide a touch sensing technology for dispersing noise interference of various frequencies. According to one feature of the present invention, a touch sensing device for detecting a touch event of a sensing array includes a driving Circuit, providing one a random pulse length square wave signal to modulate an electrical signal to generate a modulated drive signal for driving a node of the sensing array; and a sensing circuit to measure a sensing signal from the node of the sensing array And using the random pulse length square wave signal to extract the touch information of the node. The random pulse length square wave signal has a plurality of pulse waves of different pulse lengths, so that the modulated drive signal has the same According to another feature of the present invention, a touch sensing method for detecting a touch event of a sensing array includes providing a random pulse length square wave signal; The wave length square wave signal modulates an electrical signal to generate a modulated drive signal for driving a node of the sensing array; measuring a sensing signal from the node of the sensing array; and utilizing the random pulse length The square wave signal 201112068 captures the (4) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ In order to make the present invention overturn, it is better to supplement the following, and the detailed description of the present invention is as follows: [Embodiment] The present invention utilizes the characteristics of orthogonal vectors. Each vector of the vector group is vi, where i = 〇, 1, ..., η. If the product of a vector and a different vector is 〇 (that is, Vixvj = (), where j), and the product of a vector and itself is 丨(ie νιχν”· =丨, where 丨= then this is the orthogonal vector group. When VI = (al, bl, cl, dl) and V2 = (a2, sigh a, also), then V1 χ % Equivalent to al x a2 + bl x b2 + C1 x c2 + dl xd2. For example, if the vector group includes two vectors: V1 = (0, 0, 0, !) and wmu), then vlxV1 = 1, V1 is satisfied. x V2 = 0 ' and V2xV2 = Bu Therefore, VI and V2 are orthogonal. Any signal can be represented as an orthogonal vector group, such as S = clVl + c2V2 + c3V3 + ... + envn, where cl, c2, ..., cn are coefficients. If the environmental noise table is not n = lOOVi + 5〇V2 + 20V3 + 10V4 + 2V5 + 4V6 + 10V7..., where each vector V1, V2, ... represents a specific frequency band. For a known number A, if V5 is selected as the modulation vector, the modulated value <ie, the input signal>Si = AV5. It can be seen that this signal will couple noise, so the output signal s〇= AV5 + 100V1 + 50V2 + 20V3 + 10V4 + 2V5 + 4V6 + 10V7..· = 100V1 + 5〇V2 + 20V3 + 10V4 + (A+2)V5 + 4V6 + 10V7.... If the same vector V5 is used as the demodulation vector, the restored signal Sr = So χ V5 = 100 χ 0 + 50 χ 0 + 20 χ 〇 + 1() x 〇 + (Α+2) χ1+4χ〇+1 〇χ 〇··. = A+2. 201112068 If two signals are modulated using two different vectors, the two signals can be taken by using these two different vectors as demodulation vectors. For example, in ancient times, another 5 is selected from the vector V5 to modulate the letter A' and select the fine _ repaired double money B, and the human domain is si. The input signal will couple the noise, so the output signal is s〇= 20V3 + 1GV4 + 2V5 + 4V6 + l〇V7 = should be + l〇〇Vl + 50V2 + {A+2)Y5+^+4)V6+l0V7-^m^ + 10V4 + When demodulating the output signal, the signal A can be restored to SrA = So X V5 = 100 X 〇+ ν Λ 5〇X〇+ 2〇x〇+l〇x〇+ (A+2)xl + (Β+4) χ 0 + 1〇χ 〇 ... = α+2. If the motorized person θ ” D 1 V6 demodulates the output signal, the signal Β can be restored to SrB = So X V6 = 1〇〇X 〇+ Λ 〇+ 5〇χ〇+ 2〇χ〇+1〇χ〇 + (Α+2)χ〇+ (Β+4) χ1 + 1〇χ〇.·.= β+4. The multiple points of the sensing array can be processed simultaneously by the use of multiple different vectors by Sha Tin. As will be explained later. As seen, the 'reduced signal will only leave a little noise. However, as mentioned above, in order to reduce the noise, a low-pitched component (such as the Μ of the financial account) should be selected as the modulation and Demodulation vector. In order to avoid scanning all frequency bands to find the frequency band of the lowest noise, the present invention utilizes random spread spectrum (RSS) technology. The vectors selected for modulation and demodulation are random groups of complex frequencies. Therefore, the restored L number does not affect the severe impact of noise in a specific frequency band. Preferably, the selected vector changes from time to time. For example, the vector selected at time t1 is (1/4) V3+. (1/4) V5 (1/4) V7 (1/4) V8 ' and at time t2, the selected vector is (I, chat + (ι/3) ν5 + (Ning 8. Practically available pseudo Random noise (four) code technology. Figure 3 is a diagram showing three different _ codes and their power linings. Each PN code is like - a specific key. As you can see, the power components of the three grab codes represented by black arrows are scattered. The fresh SJ can be used for display purposes. The fourth hall shows a schematic diagram of modulation and demodulation of two signals according to the present invention 201112068. The signal eight is modulated by code, and the signal B is modulated by code 2. The modulated signals are combined into a combined signal & The signal a can be restored from the combined signal Sc by demodulating the combined signal Sc using the code 1. The signal B can be demodulated by using the code 2 to demodulate the combined signal Sc. Reduction in combination signal Sc. Figure 5 is a schematic diagram showing a touch sensing device 1 according to the present invention. Touch sensing device 100 includes a signal source for driving circuit 120', such as for sensing arrays (not A capacitive node 50 is shown in FIG. 1 as a current source 140, 142 for charging and discharging, an I/O (input/output) interface 15A, a Lu analog digital converter (ADC) 160, and a sensing circuit 170. In the present invention, The driving circuit 12A has a pseudo random noise (PN) code generator 122, which A PN code is generated and provided with a random pulse length square wave (abbreviated as RDSW) generator 124 〇 RDSW generator 124 generates a plurality of pulses according to the PN code provided by the PN code generator 122. Wave signals, which have different pulse lengths. Figure 6 shows a random pulse length square wave (rdsw) signal generated by the touch sensing device of Figure 5. As shown, the RDSW signal is at least Contains eight pulses with different pulse lengths τ, 2, ..., ® T8. The respective pulse lengths of the RDSW signal are determined by the random number. For example, the RDSW signal Ti (i = 1 to 8) is generated from a string of random numbers 101, 235, 76, 104, 223, 94, 160, 112. The first pulse length T1 of the RDSW signal corresponds to 1〇1, the second pulse length T2 corresponds to 235, and so on. The RDSW signal is used to modulate the current signal provided by current sources 140, 142 to form an RDSW modulated drive signal. In this embodiment, the RDSW modulation drive signal is sent to drive the capacitive node 50 via the I/O interface. That is, the capacitive node is charged/discharged according to the RDSW modulation drive signal. As is well known in the art, the change in capacitance of the capacitive node 50 due to a touch will reflect the voltage change in the sense signal Vin. 201112068 The sensing signal Vm is measured from the capacitive node 5〇 via the I/O interface ι5〇. The sensing signal Vin is a voltage signal. In order to process the sensed signal in digital form, the sensed signal vin is converted to a digital bit by the ADC 160. However, the analog signal Vin can be processed directly by omitting adc 16〇. The signal extractor 173 in the sensing circuit 170 draws the aforementioned voltage change using the same RDSW signal generated by the RDSW generator 124, which represents a change in capacitance of the capacitive node 50. The signal extractor 173 generates a demodulated signal based on the RDSW signal to demodulate the sensing signal Vin. Therefore, the sensing circuit 170 can output voltage change information corresponding to the change in capacitance 'this represents the touch information of the capacitive node 50. Fig. 7 is a flow chart showing a method of generating a random pulse wave length square wave (rjqsw) signal having a plurality of pulse waves of various pulse lengths Ti (i = 1 to η) according to the present invention. The program starts at step S10 and counts i = 1. In step S20, a random number is generated. In step S30, it is determined whether or not the random number is within an appropriate range such that the pulse wave length falls within a range between Tmax and Tmin. The upper limit Tmax and the lower limit Tmin are used to limit the range of the pulse length of the RDSW signal. The maximum pulse length of the ^SW signal cannot exceed Tmax to avoid the capacitor node 50 being overcharged. The minimum pulse length of the RDSW signal should be greater than Tmin so that the pulse of the sensed signal can be detected. If the generated random number is not in the proper range, the program returns to step S2 to regenerate a new random number. If the generated random number is within the appropriate range, the program proceeds to step s3 to check whether the count i exceeds a predetermined number η. If yes, the program ends. If not, the pulse length Τι is determined based on the random number in step S40. Further, the count i is incremented by 1, and the program returns to step S20 to cycle again. Figure 8 shows the modulation and demodulation waveforms using a random pulse length square wave (RDSW) signal in accordance with the present invention. As described, the RDSW modulation driving signal is modulated by the driving signal of the touch sensing device 1〇〇201112068 of FIG. 5, which is provided as a red modulation current source. Current signal. When the modulation drive signal is at a high level, the capacitive node 5 is charged. When the motor drive signal is at a low level, the capacitive node 50 is discharged. The sensing power generates a demodulated signal according to the fine sw signal, but has a phase shift of 90 degrees with the ·w signal. As shown, the demodulated signal has the same waveform as the modulated signal but has a phase delay of Ti/4. That is, the first-square pulse wave of the demodulated signal is delayed with respect to the modulated signal by a T1/4' demodulated signal. The four-square wave is delayed relative to the observed signal. The demodulation performed by the signal ages 73 can be implemented by multiplication and addition. For example, a digital signal processing microprocessor control unit (Dsp MCU) should be used (multiply and add) accumulator (not shown; implemented by multiply-add instruction code). If the sense signal % is (2, 2 3, % 2 & 3丄%...) and the RDSW field is (m ls !, !,...), the cumulative result is +2.6X (-1) + 2. 3x (1) +3.1χ(ΐ)+3.4χ(ΐ)
為了增加RDSW信號的亂度，可在兩個脈波長度之間加入虛擬間隔 TD °第9圖顯不利用本發明之改良隨機脈波長度方波⑽·)信號的調 變與解調波形。在虛擬聽+，無域傳輸。亦即，驅動健的信號值在 虛擬間隔為零《如圖所示，在第一脈波長度Ή與第二脈波長度T2之間插 入第一虛擬間隔TD1。在第二脈波長度丁2之後插入第二虛擬間隔TD2。以 此類推。較佳而言，各個虛擬間隔的長度亦為隨機決定。 為使本發明更臻顯明，將參照第1〇圖說明調變與解調波形，第10圖 顯示根據本發明之觸控感測裝置1〇〇在觸碰與無觸碰情況下的調變與解調 波形以及感測信號。最上方顯示調變電流信號（亦即驅動信號）的 波形。感測信號Vin顯示在中間部份。實線表示無觸碰情況下的波形’而 201112068 虛線表示有觸碰情況下的波形。如所示’感測信號Vin可表示為交流（ac) 項與直流（DC)項的總和。以RDSW信號解調之後，DC項的正負分量相 銷。AC項部份，只有與調變信號之頻率（亦即信號的頻率）相同 的分量留下。由於RDSW信號的頻率係根據亂數（例如PN碼）決定，雜 訊干擾係隨機分散在整個頻譜。 第5圖之信號擷取器173可擷取與RDSW產生器124所產生之 號具有相同的多種隨機脈波長度的感測信號部份。第11圖顯示從 根據本發明之觸控感測裝置100的信號擷取器173輸出的被擷取信號。如 所不’ DC項的正負分量相銷。在解調之後，具有與奶撕信號相同的多 種隨機脈波長度的信號部份其信號值為所有劃線區塊的總和。該信號值表 示所測得的電容值。其他信號部份之正負分量相銷。 該觸控感測裝置100可應用來測量一群圖樣化的導體中至少一個節點 的自電容或互電容。第12圖係顯示根據本發明之觸控感測裝置1〇〇應用例 的示意圖。該觸控感測裝置100係連接至感測陣列200。該感測陣列2⑻具 有一群排列成N行x N列的圖樣化導體。該觸控感測裝置1〇〇提供 調變驅動k號予列i ’並從行j測量感測信號。因此，可取得列丨與行】的互 電容變化。列i與行j的互電容變化表示節點25〇的觸碰資訊，該節點25〇 為列i與行j的父點。第13圖係顯示根據本發明之觸控感測裝置i〇Q之另 —應用例的示意圖。該觸控感測裝置1〇〇係連接至感測陣列3〇〇。該感測陣 列300具有一群排列成>^行x N列的圖樣化導體。該觸控感測裝置1〇〇提 供RDSW調變驅動信號予列i ’並從該列i _感測信號。因此，可檢測該 列上之所有節關自電容變化。賴上述實施例與細例皆咖電容性感 201112068 測陣列加以說明，但應可輕易理解本發明之技術亦可使用於電阻性感測陣 列。 雖然本發明已就較佳實施例揭露如上，然其並非用以限定本發明，本 發明所屬技術領域中具有通常知識者，在不脫離本發明之精神和範圍内， 當可作各種之更動與潤飾，因此本發明之保護範圍當視後附之申請專利範 圍所界定者為準。 【圖式簡單說明】 第1圖顯示具有感測陣列之一般觸控感測裝置（例如觸控面板）的示 意圖； 第2圖顯示各種雜訊以及信號如何與此等雜訊耦合； 第3圖為顯示三個不同的PN碼及其功率頻譜的示意圖； 第4圖係顯不根據本發明之兩個信號八與B的調變與解調的示意圖； 第5圖係顯示根據本發明之觸控感測裝置的示意圖； 第6圖顯示第5圖之觸控感測裝置所產生的隨機脈波長度方波(RDSW) 信號； $ 7圖係、顯碰據本發明之產蛾機脈波長度方波⑽，）信號之方 法的流程圖； 第8圖顯不根據本發明利用隨機脈波長度方波信號之調變與 解調波形； 第圖‘’肩示利用本發明之改良隨機脈波長度方波（RDSW)信號的調變 與解調波形； 11 201112068 第ίο圖顯示根據本發明之觸控感測裝置在有觸碰與無觸碰情況下的調 變與解調波形以及感測信號； 第11圖顯示從根據本發明之觸控感測裝置的信號擷取器擷取並輸出的 信號； 第12圖係顯示根據本發明之觸控感測裝置應用例的示意圖；以及 第13圖係顯示根據本發明之觸控感測裝置之另一應用例的示意圖。 【主要元件符號說明】 1 觸控感測裝置 120 驅動電路 10 感測陣列 122 PN碼產生器 12 控制單元 124 RDSW產生器 14 多工器 140 電流源 16 多工器 142 電流源 18 節點 150 I/O介面 21 直流信號 160 類比數位轉換器 23 Ι/f雜訊 170 感測電路 25 60赫芝雜訊 173 信號擷取器 27 區域雜訊 200 感測陣列 29 白色高斯雜訊 250 節點 50 節點 300 感測陣列 100 觸控感測裝置In order to increase the turbulence of the RDSW signal, a virtual interval TD ° can be added between the two pulse lengths. Fig. 9 shows the modulation and demodulation waveforms of the modified random pulse length square wave (10)·) signal of the present invention. In virtual listening +, no domain transmission. That is, the value of the drive signal is zero at the virtual interval. As shown, the first virtual interval TD1 is inserted between the first pulse length Ή and the second pulse length T2. The second dummy interval TD2 is inserted after the second pulse length D2. And so on. Preferably, the length of each virtual interval is also randomly determined. In order to make the present invention more apparent, the modulation and demodulation waveforms will be described with reference to FIG. 1 , and FIG. 10 shows the modulation of the touch sensing device 1 according to the present invention in the case of touch and no touch. And demodulate the waveform as well as the sensed signal. The waveform of the modulated current signal (ie, the drive signal) is displayed at the top. The sensing signal Vin is displayed in the middle portion. The solid line indicates the waveform in the case of no touch, and the dotted line in 201112068 indicates the waveform in the case of touch. As shown, the sense signal Vin can be expressed as the sum of the alternating current (ac) term and the direct current (DC) term. After demodulation with the RDSW signal, the positive and negative components of the DC term are matched. The AC term is only left in the same component as the frequency of the modulated signal (ie, the frequency of the signal). Since the frequency of the RDSW signal is determined by random numbers (e.g., PN codes), the noise interference is randomly dispersed throughout the spectrum. The signal extractor 173 of Fig. 5 can capture the portion of the sensed signal having the same plurality of random pulse lengths as the number generated by the RDSW generator 124. Fig. 11 shows the captured signal output from the signal extractor 173 of the touch sensing device 100 according to the present invention. If not, the positive and negative components of the DC term are sold. After demodulation, the portion of the signal having the same plurality of random pulse lengths as the milk tear signal has a signal value that is the sum of all the scribe blocks. This signal value represents the measured capacitance value. The positive and negative components of other signal parts are sold. The touch sensing device 100 can be applied to measure the self capacitance or mutual capacitance of at least one of a group of patterned conductors. Fig. 12 is a view showing an application example of the touch sensing device 1 according to the present invention. The touch sensing device 100 is connected to the sensing array 200. The sensing array 2 (8) has a group of patterned conductors arranged in N rows x N columns. The touch sensing device 1 〇〇 provides a modulation drive k number to the column i ′ and measures the sensing signal from the row j. Therefore, the mutual capacitance change of the column and row can be obtained. The mutual capacitance change of column i and row j represents the touch information of node 25, which is the parent point of column i and row j. Figure 13 is a schematic view showing another application example of the touch sensing device i〇Q according to the present invention. The touch sensing device 1 is connected to the sensing array 3A. The sense array 300 has a group of patterned conductors arranged in >^ rows x N columns. The touch sensing device 1 provides an RDSW modulation drive signal to the column i' and senses the signal from the column i_. Therefore, all of the self-capacitance changes on the column can be detected. The above embodiments and the detailed examples are described in the 201112068 array, but it should be readily understood that the technique of the present invention can also be applied to the resistance sensing array. While the invention has been described above in terms of the preferred embodiments, the invention is not intended to limit the invention, and the invention may be practiced otherwise without departing from the spirit and scope of the invention. The scope of protection of the present invention is therefore defined by the scope of the appended claims. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic diagram showing a general touch sensing device (for example, a touch panel) having a sensing array; FIG. 2 is a view showing how various noises and signals are coupled with such noises; A schematic diagram showing three different PN codes and their power spectrum; FIG. 4 is a schematic diagram showing modulation and demodulation of two signals eight and B according to the present invention; FIG. 5 shows a touch according to the present invention. Schematic diagram of the control sensing device; Figure 6 shows the random pulse length square wave (RDSW) signal generated by the touch sensing device of FIG. 5; $7 image system, the wavelength of the moth-making pulse according to the present invention A flowchart of a method for a square wave (10), a signal; Figure 8 shows a modulation and demodulation waveform using a random pulse length square wave signal according to the present invention; the figure ''s shoulder shows the improved random pulse using the present invention Modulation and demodulation waveform of wave length square wave (RDSW) signal; 11 201112068 The figure shows the modulation and demodulation waveform and sense of the touch sensing device according to the present invention in the case of touch and no touch. Measuring signal; Figure 11 shows the touch from the present invention a signal captured and output by the signal extractor of the sensing device; FIG. 12 is a schematic diagram showing an application example of the touch sensing device according to the present invention; and FIG. 13 is a view showing the touch sensing device according to the present invention. A schematic diagram of another application example. [Main component symbol description] 1 touch sensing device 120 driving circuit 10 sensing array 122 PN code generator 12 control unit 124 RDSW generator 14 multiplexer 140 current source 16 multiplexer 142 current source 18 node 150 I / O interface 21 DC signal 160 analog digital converter 23 Ι / f noise 170 sensing circuit 25 60 Hertz noise 173 signal picker 27 area noise 200 sensing array 29 white Gaussian noise 250 node 50 node 300 sense Array 100 touch sensing device
1212