TWI290221B - Two-dimensional motion sensor - Google Patents

Abstract

An optical sensor and method of using the same is provided for sensing relative movement between the sensor and a surface by detecting changes in optical features of light reflected from the surface. In one embodiment, the sensor includes a two dimensional array of photosensitive elements, the array including at least a first plurality of photosensitive elements arranged and coupled to sense a first combined movement along a first set of at least two non-parallel axes, and a second plurality of photosensitive elements arranged and coupled to sense a second combined movement along a second set of at least two non-parallel axes.

Description

1290221 IX. Description of the Invention: TECHNICAL FIELD OF THE INVENTION The present invention generally relates to optical navigation systems and methods of sensing motion using the same. [Prior Art] It is known that a negative input I such as a computer mouse, a touch display screen, a ruling ball, and the like is used to input data to a personal computer and a workstation, and a personal computer and a workstation establish an interface. These devices allow the cursor to be quickly monitored and can be used for many text, database read programs. The user controls the cursor for the cursor (for example) by moving the mouse on a plane to make the cursor It is realized in the direction and movement commensurate with the movement of the mouse. / α /, there are two forms of ancestor and mechanical. Mechanical mouse usually uses a rotating ball to detect motion, and 佶 code cries * " 接触 一 一 一 一 - 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对 对There is a problem with the mechanical mouse, that is, continuous capture, and after use, the mechanical slipperiness due to the accumulation of stains is prone to deviations and the original production of the temple is also early. In addition, the mechanical components ( The specific horse's core: the mechanical vehicle is coded by the encoder. The wear and tear caused by it undoubtedly limits the service life of the four garments. The optical guide wheel is used to solve the problem of the mechanical mouse discussed above. The program is for the development of the mouse using the Yuangan V navigation system. Pointing to the accuracy and less...: Finding first... Providing better, it has become very popular. "Early shirting, so it is used today for optics. The main technique of qizi...the gas depends on: the light source such as illuminating 106029.doc 1290221 (4)), whose illumination is at or near the surface of the grazing incidence; two = 2D) CM0S (complementary metal oxide semi-conducting (four) detector, (4) Synthesis and signal processing unit's and continuous shadow Thousands of features or points are associated to determine that the mouse has moved away from the "one (four) direction, distance, and speed. Although this technique is ambiguous", it is subject to more complex image processing requirements. ^为_ Improved, a coherent light source such as a laser can be used to illuminate the coarse to create a composite interference pattern (called speckle) that has a higher pair of effective light generation under incident illumination and even a standard Thunder $ Comparing the shirt image between the daytimes. The laser-based light produces a higher conversion efficiency and higher directionality, and the higher directionality allows the slave illumination to occupy an area specially designed to correspond to the area of the photodiode. In addition, the speckle pattern allows for tracking operations on virtually any surface: (wider surface coverage), such as a defocusing. (4) The maximum contrast is still maintained under the image. The connector or system has a photodiode (1D) array, such as a photodiode system, which is called a comb array. The photodiode in the 1D array can be straight: Wiring to the received signal Row analogy and parallel processing, which is processed by dimension (10) signal and easy to detect motion. Two methods of this method are used: multi-measurement and t, the world has proposed a multi-Korean linear array, in which two or more: more than four (four) column along Non-parallel wheels are arranged. The comb-like array device is relatively tender for the previous related types of optical mice, but there are still many unsatisfactory ones. Specific 106029.doc 1290221 These defects based on speckle-based devices It has limited accuracy along the direction of significant deviation from the (1) array orientation. This problem is particularly pronounced when the optical mouse moves in the off-axis direction, which (4) acts as a speckle pattern or enters the observation field too fast and leaves the 1D array. This makes the image less than a clear signal. This can be partially renewed by increasing the number of axes, which is the simplicity of the method of reducing the linear comb array. The optical pointing device and the method of using the same are combined with the accuracy of the buckle displacement measurement of the related type of device and the simplicity of the signal processing of the comb array type device. SUMMARY OF THE INVENTION The present invention provides a solution to these and other problems, and further provides advantages over conventional devices and methods of using the same. The present invention relates generally to optical navigation systems, and more particularly to optical sensors for sensing the relative lateral movement of the sensor with its surface above: the surface that moves above. Optical navigation systems may include (eg, optical computer sliders, trackballs, and the like, and are well known for inputting data to personal computers and satellite stations, and establishing interfaces with personal computers and workstations. In the following description, For the purpose of explanation, the invention may be described in detail in the details of the invention. The invention may be <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; It is not shown in detail or in the form of a block diagram to avoid unnecessary confusion of the description of the description. For the description of the "one" embodiment, reference is made to the description of the embodiment 106029.doc 1290221 A particular feature, structure or feature is included in at least one embodiment of the invention, and the phrase "an embodiment" that appears in various places in the specification is not necessarily all referring to the same embodiment. I conjugated to" may include both direct connection and interconnection via one or more intervening components. Preferably, 'the invention The sensor is a speckle-based sensor that senses the displacement based on the displacement of the composite intensity distribution pattern of light (generally known as speckle). The speckle is essentially a coherent ray that leaves the rough surface. a complex interference pattern generated by scattering and detected by the intensity photosensitive element, the intensity is sensitive

The component can be, for example, a photodiode having a field of view (or numerical aperture) of a limited angle. More preferably, the optical sensor comprises a two-dimensional (2D) array that combines the accuracy of the displacement measurement of the 2D correlator with the simplicity and efficiency of the signal processing of a linear or one-dimensional (1D) comb array. The 2D array can be: a periodic 2D comb array comprising a plurality of regularly spaced photosensitive elements having a period of 1 or 2 cycles, a quasi-periodic 21) array (such as Penr〇se paving); or a non- A circumferential shoulder 2D array that has a regular pattern but does not contain periodicity. A comb array means a planar array of a plurality of regularly spaced and electrically connected photosensitive elements that extend substantially in at least two non-parallel directions and have periodicity in two dimensions. [Embodiment] The following description will gradually explain the theory of 2D comb array signal processing, and describe various exemplary embodiments of a 2D comb array architecture. Image-related comb-like array processing The comb-like array technique in the dimension (1D) is useful for comparing image-related signal processing with a single technique. 106029.doc 1290221 ID correlation between f and g signals can be expressed as follows: Bribe (/, Zha = £仏«-0 ( i ) Assuming that the average of both f and g is zero. Otherwise, such The signal can always be redefined by offsetting with an individual mean. If g and f have some similarities, then at a particular value of the displacement m, the correlation peaks, and (iv) the two signals are common. The characteristics are generally optimally aligned or, related. For a "1D" mouse, g is considered to be (to a greater extent) a displacement of f, meaning gn = fn + x. Therefore, the correlation is y ( (, / + 丄 = | / ; / 2 (2) where X is the displacement. When m = x, the correlation function (Eq2) reaches the peak. Therefore, the knowledge of the peak position is known. The displacement can be determined. In the conventional optical mouse, the captured signal f is used as a short-term template to correlate with the capture after Ik. Once the displacement is determined, the new capture will replace the old template, etc. This dynamic template It is advisable for random signals. If the type of signal (such as periodic signal) is already scheduled, a fixed template can be used. This removes the need to continuously update the signal template. This greatly simplifies the related operations and the construction of the device. In fact, the comb array is a device as will be described in more detail below. To achieve this, the signals can be Expressed as follows: Discrete Fourier transform 106029.doc -10- (3) 1290221 Change (DFT) expansion:

2man! N α=0 Therefore, the correlation (1) becomes: Α^-1 ί Λ/_ι «=〇 V〇=0 yv-i n-\ a — Q b=0

I2m(a-b)n/N \

a~0 b=0 AM AM

ΣΣά, 2 ¢7=0 AM

Mbm / N (4) and the correlation (2) becomes: N-\ (5) 〇〇n(fj+x)m = J^FqF; e2ma(mx)/ N &lt;2=0 ID comb array line 1 The ± or 1-array array is an array having a plurality of photosensitive elements, and the photosensitive 7L pieces are connected in a periodic manner, so that the array acts as a fixed template for the interrogation signal spatial frequency: component. One such U-comb-like embodiment is illustrated in Figure 1, and will be described in more detail below. The plurality of photosensitive elements connected in a circumferential manner enable the comb array to function as a correlator at the drag-and-drop position (the sy, the ratio & the ratio of the A &lt; 7L pieces in the array and the optical collector). The signal can now be regarded as displacement X) with seven K = FAF; where c is the amplitude of the slowly varying, and (6) κ - 2πΑ / Ν is the selected space 1 〇 6 〇 29. doc continuation - II - 1290221 rate . The factor elKm can be considered to be the stage in which the selected spatial frequency component of the code is aligned with the initial alignment of the template. Thus, it can be inferred that the 1D comb array is essentially m-correlated within a spatial frequency. Two-Dimensional Comb Array The above observations lead to the conclusion that 21) comb arrays can be constructed and grouped to provide a 2d correlation at a spatial frequency 犮 = (Κχ, Ky). The image f and its own displacement form [(x, illusion displacement) 2〇 are related to: ν-\ν~λ

=ΣΣ仏f,F*b e2Kiaim-x)/Ne27tib{ny)IN a=0 6=0 V / / Similar to Equation 6 above, the 2D comb array signal is: (8) elKxKm-x) As described above, (Kx ' Ky) Ξ (2πΑ/Ν, 2πΒ/Ν) is the selected 21) null $ frequency. The comb signal is simply the product of the harmonics of the displacement 乂 and 丫. It should be noted that the comb signal is periodic and peaks as long as the template and the spatial frequency of the image are spatially in phase. For the sake of simplicity, set m, (4), the product of the angle in Equation 8. The product is expanded to CC = cos(Kxx)cos(Kyy) CS = cos(Kxx)sin(Kyy) (9) SC = sin(Kxx)cos(Kyy) SS = sin(Kxx)sin(Kyy) The next step is to determine the 106029.doc -12-1290221 2D array that produces the four signals as in (9) φ above. configuration. I first recall that it would be beneficial to generate in-phase and positive-female nicknames in a 1D comb array configuration with 4 components per cycle. Figure j shows such as photodiodes! The general configuration of the 1D comb array 1 〇2 of the photosensitive element (along an axis), wherein the combination of the photosensitive elements of the 廛 group will be brighter by the speckle (or non-speckle) image. The spatial frequency of the dark signal acts as a periodic filter. In the non-embodied embodiment, the 1D comb array 102 is comprised of a plurality of photodiode sets or cycles, each having four photodiodes 104, labeled A, B and D herein. The current or signal from the corresponding or similarly labeled photodiode 104 in each cycle is electrically connected (total wiring) to form four line signals 1 〇 6 emanating from array 1. Background suppression and signal enhancement are accomplished by using differential analog circuits 1〇8 to generate in-phase differential current signals, denoted herein as C^t, and by using differential analog circuit 11() to generate positive-parent differential current signals. , marked here as SQut. Comparison of the phase of the in-phase and quadrature signals may allow for determining the magnitude and direction of the motion of the 1D comb array 102 relative to the scattering surface. Referring to Figure 1, the in-phase c〇ut and quadrature s〇ut signals are obtained by taking the underlying speckle pattern and processing them according to the cosine and sine templates 112 and 114, respectively. Preferably, the system is designed such that the optical "bright-dark" signal pattern (^P political spot) is substantially the same size as the comb array (four (4) photodiodes in the embodiment of Figure 1). The period of 4 or pixel is equal. As shown in Figure 1, the in-phase current is obtained from c〇ut = A-C, and the current of the orthogonal signal is self-obtainable. The cosine and sine assignments above are now applicable to the 2D case. The result is the four matrices shown in Figures 106029.doc 1290221 2A through 2D, which are the four harmonic product products shown in Equation 9 above. In particular, Figure 2A shows a matrix of CC or cos(Kxx)cos(Kyy) signals having a 2D comb array of photosensitive elements grouped in groups of 4x4 elements per unit. To simplify the symbol, the subscript "out" will be omitted from now on. Similarly, Figure 2B shows a matrix of CS signals, Figure 2C shows a matrix of SC signals, and Figure 2D shows a matrix of S S signals.

As shown in Figures 3A and 3B, a 2D comb array can now be constructed from the above matrix. Here, the 2D comb array 302 has a plurality of photosensitive elements 304 configured or grouped into units 306, each unit having a photosensitive element grouped in a configuration of 4x4 elements per unit (or 4x4 elements/cycle). As shown in detail in FIG. 3B, the photosensitive elements 3〇4 having the same letter and the same number in the unit 306, and the corresponding elements of all the units having the same number in the 2D comb array 302 are electrically connected or collectively wired to obtain A1 to D2. The eight signals. The eight total wiring signals are further combined with the difference amplifiers 3〇8 to give the following four signals. · COA1-A2 CS=B1-B2 (1〇) SOC1-C2 SS = D1-D2 These four signals contain In-phase and quadrature information in the X and y directions. Using the dihedral identity, the product of the 谙 function can be converted into a simple harmonic function (the sum and the spectral function of the difference):

Cos(Kxx+Kyy) = CC-SS sin(Kxx+Kyy) = SC + CS 106029.doc (11) -14· 1290221

C〇s(Kxx-Kyy) = CC + SS sin(Kxx-Kyy)=SC-CS The coordinate system or array can be rotated 45 as needed. To get an expression expressed only by X and y. In either orientation, the 2D displacement can be determined subsequently. In practice, and Ky can take the same value. The 2D comb array presents a simple design and a number of better practices --- , one ---, yv- \ 厶 邛曰

More advantages of off-and/or multi-axis 1D comb arrays include: (1) faster signal processing; (Π) reduced power consumption; (iii) higher angular accuracy; and (iv) non-crossing relative to The effectiveness of the limit of the direction of the array orientation. "Compared, because the 2d comb array produces much less data to be processed and thus produces a much simpler algorithm to be executed, it has significantly faster signal processing. For example, the more zero debt can be used The algorithm is used to determine the displacement. To specify the displacement in a plane, two real numbers, X and 平移, are required. In the conventional optical mouse based on correlation, these two real numbers can be correlated from successive images. Decision. Because each image in the correlation-based method usually includes about 1 〇 3 pixels, only the translation values of the two χ and y- are required to process a large amount of data. In contrast, the 2 〇 comb array is only Produces four (4) positive real numbers 'which are equivalent to only two (2) real numbers with positive and negative points. In a sense, 'parallel processing is constructed in the interconnect architecture of the 2D comb array. By treating this"布 ” ” 兮 槐 槐 槐 槐 槐 槐 槐 槐 , , , , , , , , 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余 剩余And increased capital The source implements a complex digital signal processing (Dsp) algorithm to further enhance the use of the optical salt of the present invention. This is especially true, and the optical navigation system of the benefit is traced to 106029.doc -15-1290221. Compared to correlation-based devices, because 2D comb arrays have much less data to process, and therefore have much simpler algorithms to implement, they are expected to consume less power. This is a highly desirable feature for power sensitive applications such as wireless optical mice. Electrical energy consumption can be further reduced by combining with effective laser illumination, such as in a laser based laser speckle. Compared to the angular accuracy of the conventional 2D correlator mouse, the angular accuracy of the 2] comb array can be more easily scaled. The minimum angle detectable by the 2D sensor is inversely proportional to the number of photosensitive elements in a column or row. The improvement in angular accuracy generally depends on the increase in the number of photosensitive elements within the array. This puts a heavy burden on the 2D correlator mouse because the amount of data to be processed rises quadratically with the number of components in a column or row. In contrast, the amount of data or signals to be processed in a 2D comb array is independent of the number of components. That is, in a 2D comb array having a configuration similar to that shown in Figs. 3 and 3], the number of differential signals output from the 2E&gt; comb array is always equal to four, and Angle accuracy is therefore limited only by the size of the array that can be constructed. Finally, the performance of a 2D comb array is independent of the direction of movement relative to the array compared to a 1D comb array. Referring to Figures 4A and 4B, since on average, each point in the image passes through the path 408 in the active region of the 2-comb array 4〇2 in all directions than in the 1D comb array 406. The 410 is longer' and therefore more conducive to the estimation of the displacement, so the performance of the 2d comb array 402 is superior to the optical sensor 404 having a linear or 1D comb array 406. This 106029.doc • 16 - 1290221 is 'because the embodiment of the 2D comb array described previously works with the pixel geometry of symmetry (10) such as a square), it is easier to achieve a &quot;bright&quot; signal pattern (ie The matching of the speckles with the period of the 20 comb arrays results in improved signal contrast and higher front-end SNR compared to the conventional (1) comb-like arrays that typically use a highly "asymmetric" pixel shape. Finally, this Compared to a fish multi-axis (1) comb array, the 2D array can be more effectively illuminated, and thus the power consumption is also lower. Illustrative Embodiments and Experimental Verification Figure 5 is shown - having an embodiment in accordance with the present invention An exemplary embodiment of an optical navigation system based on a speckle-based 2D comb array. Referring to Figure 5, the optical navigation system 502 generally includes an optical head 504 having a light source 5〇6, such as a VCSEL. (Vertical cavity surface-emitting lasers broadly include illumination optics for a first or collimating lens 508 for collimating divergent beams, imaging optics comprising a second or imaging lens 51 for roughing The illuminated portion of the scattering surface 512 is mapped or imaged to a 2D comb array 514 at the image plane of the second lens. Preferably, the illumination optics are configured to be illuminated at a predetermined angle of incidence selected to allow for ascending prediction The surface 512' is such that if the optical head 504 or the data input device is spaced from the surface 512 by more than a predetermined interval, the device will suspend the tracking of the action, and the imaging optical device may include a back surface of the second lens 51. Focusing the aperture 516 of the plane to provide an ankle imaging system that maintains the integrity of the excellent speckle pattern during motion, and matching the average size of the speckle to the period of the 2] comb-like array. Advantages of the optical navigation system of the 2D comb array 514 of the invention 5〇2 106029.doc -17-1290221, we have made 32 χ 32 photodiodes (PD) or components and are similar to those shown in Figs. 3A and 3B Square, symmetrical 2) comb-like array. And shown in Figures 6A and 6B at various speeds and on two different surfaces

The results of the circular trajectory verify the disclosed method. The experiments from Figure 6 and Dip's diagrams were carried out on a test platform where the extremely high-precision (four) optical navigation system &lt; the relative movement between the optical head and the surface was possible. Figure 6A is a diagram showing the circular trajectory produced when the optical head is moved four times on a white surface at a speed of 1 cm (10 cm/s, 10 cm/s, 25 cm/a 4 〇 cm/s). . Figure 6B illustrates a circular trajectory produced when the optical head is moved at the same speed over a wood texture. The reference circle of the dotted line in Fig. 6A and "indicated by reference numeral 6〇2, and the obstruction or circular trajectory generated by the optical navigation system is indicated by a solid black line. The number along these axes can be any unit. From this trajectory, it is known that an optical navigation system having a sensor using the 2D comb array of the present invention can sense movement at a speed of up to 4 G em/s on a patterned and unpatterned surface and has typically less than 5 % path error. Subsequent reliance on accurate tracking of a wide variety of surfaces and wide range of motions. A general rule of the array for a linear or m comb array is also applicable to the 2D comb array of the present invention, the 2D comb array comprising: (1) having a different from 4 χ 4 elements per unit _ comb array; (8) a 2D comb array having a plurality of sub-arrays of a given spatial frequency; (iii) a 2D comb array having a plurality of sub-arrays of different spatial frequencies; and (iv) a comb-like connection having a dynamic re-resonance between the photosensitive elements The array of roles, which can cause the spatial frequency (four) state to be changed 106029.doc -18-1290221 'for example, the intensity of the signal from the array can be optimized. It is to be further understood that the 2D comb array in accordance with the present invention may also comprise the above-described general or combination of embodiments. Some alternative embodiments of a 2D comb array comprising one or more of the above general principles will now be described in more detail with reference to Figures 7 and 8. An alternative embodiment of the 2D comb array has a different unit than 4 &lt; 4 elements per unit. For example, as shown in FIG. 7, the 2D comb array 7〇2 includes a plurality of photosensitive elements such as a photo-electric body 704, which has 6χ6 elements per unit (or 6×6 elements/cycle) in the unit 7〇6. Configuration grouping or configuration. As described above with reference to the examples of Figures 3 and 3, the corresponding elements of all of the elements in each of the cells 7〇6 and 2〇 comb array 702 are coupled to thirty-six (36) One of the output lines. The 36 total wiring signals are further combined with a weighting factor in accordance with matrix 708 to produce four output signals _cc, cs, and SS. The details of the matrix used to generate each of these four signals are detailed in the table below.

CC 1 0.5 -0.5 -1 «Π ς Λ C 0.5 0.25 -0.25 -0.5 -___ VJ · 3 〇oc -0.5 -0.25 0.25 0.5 -------- --- 0.5 -0.25 -1 -0.5 0.5 ^__0.5 -0.5 -0.5 -0.25 0.25 0 〇oc 0.5 0.25 -0.25 -0.5 -U.ZD Π 0&lt; _ cs 0 r\ 0.866 0.866 0 Π on 0 r\ 0.433 0.433 0 -/ -Uo / A ΛΊ 0 r\ -0.43 -0.43 0 —— 0 0 -0.87 -0.87 0 —〕 〇〇Qf:fi 0 r\ -0.43 -0.43 0 —_______y · o uu 0 U.OOD Π 0 0.433 0.433 0 __ '0.43 - 0.43 106029.doc -19- 1290221 sc 0 0 0 0 0 0 0.866 0.433 -0.43 -0.87 -0.43 0.433 0.866 0.433 -0.43 -0.87 -0.43 0.433 0 0 0 0 0 0 -0.87 -0.43 0.433 0.866 0.433 -0.43 -0.87 -0.43 0.433 0.866 0.433 -0.43 ss 0 0 0 0 0 0 0 0.75 0.75 0 -0.75 -0.75 0 0.75 0.75 0 -0.75 -0.75 0 0 0 0 0 0 0 -0.75 -0.75 0 0.75 0.75 0 -0.75 -0.75 0 0.75 0.75 In other alternative embodiments, the optical sensor can comprise a plurality of 2D comb arrays or sub-arrays of a given spatial frequency or different spatial frequencies. For example, FIG. 8 shows an optical sensor 8〇2 having two 2D comb array pairs disposed in quadrants 8〇4, 8〇6, 808, and 810 in accordance with an embodiment of the present invention. Unintentional block diagram. The diagonally opposite quadrants 8〇4 and 8〇6 are connected and form a first array pair or a first 2D comb array. The phase object is limited to 8 〇 8 and 8 j 〇 , and the two are connected and form a second single array pair or a second 2D comb array. In the example described in the first paragraph, the quadrants 8〇4, 8(10) and each of the individual members, and the corresponding components of all the units in the array pair are coupled into sixteen (16) total wiring signals 814. . The 16 total wiring signals 106029.doc -20· 1290221 814 are further combined with the difference amplifier gig to generate eight (8) signals:

CC1, CS1, SCI, SS1 of the first 2D comb array, and CC2, CS2, SC2, SS2 from the second 2D comb array. In the operation of the towel, since the selected spatial frequency component is weak at a certain position on the surface, or because the effects from different parts of the array are uniformly added to zero, whether from the 2D comb array or from the array The intensity of the signal can be reduced. However, it should be understood that the weakening of any pair of arrays is unlikely to result in other internal weakening, so the multiple array or sub-array configurations are generally suitable for mitigating the weakening of the signal. Furthermore, the square symmetrical configuration of the optical sensor 802 enables all of the photosensitive elements 8丨8 within the optical sensor to be illuminated simply and efficiently. The general rule of law will now describe a general theory that leads to unexpected results, that is, by changing the weighting coefficients of various photosensitive elements in the optical sensor, any array can be used to extract speed from any finite set of arbitrary axes. vector. To date, only comb arrays having 1D or 2D periodicity have been considered, but other patterns having non-periodic or undivided units may be used. For example, the buckle array can have a pattern of sunflower blossoms, or even the number of photosensitive elements can be pseudo-randomly located, and it can still extract 1D or 2D motion information. All of the following examples are special cases of the general concept, meaning that a 1 〇 or 2D array is used to capture 1D or 2D motion. A few new light wind sensor configurations are also described. There are two basic concepts in optical navigation systems based on spotting. First, the pattern is a superposition of a plurality of 2D spatial frequencies of a plurality of frequencies determined by the optical characteristics of the system. Second, the m action (ie, along the selected axis or 106029.doc 21 1290221 ^ upward action) can be selected by selecting the specific distribution of the light of B as the speckle pattern and observing how it follows the Selecting, 2, or changing in the direction of the selected direction, the distribution system, - in addition to the multiplicative odd amount (the value can get the action value = :), 'does not change shape with the action Distribution. This means that if the distribution; ten privately, keep a constant y. The heart&apos; requires that the components within the 1D sensor respond to shifts in either direction&apos; but only extract components along the selected axis or in the selected action. For example, the optical head of an optical navigation system can be moved in the direction of the data 7 on the surface, but only the component of the motion along the x-axis is extracted. 4::: The distance that Γ crossed in the χ direction is defined as (10), and the function that the distance of the thousand-way opener is crossed in the 7 direction (4) is the function of the kidney. The following equation is required: ^(^χ) ^γ(^γ)Φ(χ,γ) = Λφ(χ^γ) , (12) For a certain constant Λ, meaning that, in addition to doubling the constant, the translation still leaves a flaw that has not changed. Good, may extract from the constant and search for the action.,,, 2, 乂匕ί can be recognized as the equation, that is, the searched light distribution ^ is known as the flat and private talent 7; (meaning) and its eigenfunction. The eigenfunctions of the translation operators are complex exponents in the direction of translation, and have any functional form in the vertical direction. That is, for translation in the 乂 direction, the number is: 〇xp(2mx /x)Qxp(2my/y) and the eigenvalues are: 106029.doc -22- (13) 1290221 A = ^p(2mdxfx)Qxp(2mdyfy) (14) where fx is the spatial frequency in the x direction And fy is the spatial frequency in the y direction. Therefore, by measuring the light distribution before and after the action, extracting the coefficients of the searched eigenfunction and calculating the eigenvalue, The scaled eigenvalues are derived from λ to extract the distance traveled in the χ (or y) direction. This extraction is done by using the eigenfunction in the inner product of the orthogonality. Therefore, given a special concern The distribution of the eigenfunction can be found by finding the inner product together with the eigenfunctions sought. The translation operator is Hermitian under 5 nautical miles: (u,v)= \\ u{x,y)v {x,y) dxdy ()5) means that if a speckle pattern 6Xx, y) contains a portion of the searched eigenfunction having a coefficient of the ratio c, Obtain the eigenfunction of interest and the inner product to extract c:

c = ]]%,3;)Which (-2;τ〇ρ(-ώ(16) At this point, a smaller complexity occurs. These eigenfunctions and eigenvalues are plural However, the speckle pattern grandchild, 4 is a real number, and it is only possible to perform real number algorithm using the photosensitive element and weighting coefficient in the hard body. Therefore, the real part and the imaginary part of the complex coefficient will be calculated separately. In general: c Ξ Cr + i〇i - l\S(x,y)y{[exp (^2πι (xfx + yfy + φ〇))] dxdy + i J{^(x^)3[exp (xfx + yfy + φ 〇dxdy ( 1 7 ) 106029.doc -23- 1290221 should be added to the general rule of interest. Equation 1 7 is integrated to obtain the value of complex c. After moving the sensor or array, repeat The above integral is obtained to obtain a complex number. The new value, c, 纟 should be equal to the value obtained by obtaining the ratio of these two values to obtain the eigenvalue 乂, thereby finding that the sensor or array has moved Distance. This can be made by calling the following formula: (18) Also = exP(2;ri("/x+a))

Therefore, both the X-action and the y_action are mixed in ^. If you only need to extract the X action, you can remove the base value by selecting, =〇, meaning that there is no change in the y direction. Then A or Equation 18 can be simplified as: A = exp(2^/(^)). (19) and &lt; can be extracted directly from the calculated A value. Therefore, in order to detect the action of the x_component of any action, the selected eigenfunction 姒, less) is equal to eXp(2; dXfx). Means: do, 7) = exp(2;r^). The main signal and the orthogonal signal can be calculated from the following equation: C = cr + icf = jj5(x5&lt;y)9i [exp (-2^:/(χ/Λ + ^〇))] dxdy + 1 JJ5(x ^P[exp + ^0))] dxdy ( 2 1 ) As described above in McCaw's equations 17, 18 and 19, the eigenvalue A is subsequently obtained from the two measurements, and in the X direction. The distance of the upper movement is also extracted from the following equation: (22) It should be understood that the above method can be generalized to detect the ID movement in any desired direction. 106029.doc -24-1290221

In the knife, not just along the χ (or y) direction. This is most easily accomplished by the rotation of the trigger so that the X-axis or direction can be extended in the desired direction, and all formulas described above can be applied to the rotated coordinate system. U. A description of the method of performing the above integration in a real hard body will now take advantage of the available visible light of the trigger sensor or array, requiring an array with a fill factor close to the uniform-illumination. That is, the array will contain a plurality of photosensitive elements, each of which is at a position within the array. ^: With a fill factor that attempts to cover an array of regions as a whole of the photosensitive element region. (d) When the action of the array is determined, it is finally possible to sum the outputs of the photosensitive elements with certain weighting coefficients. In order to accurately calculate these coefficients, use for the purpose of illustration... To satisfy:

Cr Σ J] takes less) 9i[exp(-2; π·(χ/χ ”.))) (23) However, it is unlikely that the above equation is accurately constructed in the hard body. Instead, each-photosensitive The components are given a weighting coefficient of ';, / this is the weighting of the following weight-plus-sum. (24) ~=ΣΊ(χ,γ)^· Its closest For the formula (23): ~=Σ+ Π取洲[exP(-2;Π·(^Λ + furnace.))] Office: Ά (25) Given the value of the weighting factor: (26) ==令沢[exp(一两吨乂).))]=令 cos(—2;r(x乂坤〇)) 106029.doc -25- (27) 1290221 Similarly, for the imaginary part (positive Pay) signal: c, = Σ 'Ji take less) by: (28) -4exp(-2,-(,X + ,〇))] = Jsin(-2,(^ + ,〇)), It should be noted that "丨" is a two-fourth (4): the first means that it represents the imaginary part of C; and the second is the index of the photosensitive element.

The above-described framework can now be used for (4) in the arbitrary-space frequency-(four) column to detect the 1D component of the action in the any-direction direction. It should be noted that the use of a particular 2-column or 2D array having a particular shape is not mentioned above. However, several array strategies with certain desired characteristics will be described in detail below. Several different strategies or methods can be employed in the placement of the array. First, it should be reiterated that this is beneficial in at least three 1D directions. It is necessary to have at least two directions to obtain the two components of the action, but because the speckle is compounded, any given eigenfunction will fade out, causing signal loss. It should be noted that the value of the ^Kalman filter does not provide correct action if the user changes direction when the signal has faded. Therefore, it is necessary to provide at least one signal to some degree of resistance to fading. Moreover, while the signal processing required to process additional information is more complex, it would be more beneficial to offset some of the advantages achieved by using speckle-based optical sensors with 2D arrays. There is some scalability in how to provide this redundancy, including: (1) several spatially separated arrays (spatial redundancy); (u) given parties 106029.doc -26-1290221 2 directions: no n? Spatial frequency (space frequency redundancy: multiple axes in two directions (direction redundancy), of course 'in this case, the action value via &lt;# is no longer necessarily orthogonal. For simplicity, in the following In the description, the single-action split of any K test will be called &quot;signal/orthogonal pair&quot; (SQ-pair). If two or more SQ-pairs are detected, it must be The initial decision is made to determine how to use the sensing elements in the array. In detail, the components in the array can be connected such that: (1) any given photosensitive element is only fed into -SQ·-; (8) signals from each photosensitive element Splitting and giving different weighting wires, so that 'a given photosensitive π piece can be mSQ_pair; or (iii) is combined with (1) and (u)n. Since the signal splitter and the buffer consume energy budget and lc (integral) Circuit) or wafer size without using signals from each photosensitive element in each SQ_pair Ratio (SNR), so which method is used will be partly related to the trade-off between energy budget and 1C or wafer size. The configuration of the ideal sensor element for detecting only 10 actions will be considered first. It is configured to have photosensitive elements with the following weighting system and number in each place. ·~=fC〇s(—2 small 乂”❶)) (29) 'Ί3ΐη(-2; Γ(Ό)) (30) Since the weighting coefficients are not dependent at all, all the photosensitive elements in the vertical line have exactly the same weighting factor. Alternatively, there are many equal weighting coefficients by wiring only all of the photosensitive elements in a vertical line 106029.doc -27- 1290221 Avoiding the use or use of thinner, thinner photosensitive elements. This implies that the use of 2D arrays of equal length and width will generally produce the desired m action away from the axis. Also note the cosine and sine Has been staggered to zero. If the weighting coefficient is zero, then there is no need to waste a photosensitive element here, so if the photosensitive elements are separated by 1/4 cycle, the total signal and the orthogonal signal photosensitive element can and only Help each sensitization of a single signal The components are staggered. Therefore, a preferred 2D array for m motion detection is disposed in a plurality of vertical strips of photosensitive elements having alternating signals that contribute to the main signal and the quadrature signal. Furthermore, since the eigenfunction of the search is The y direction is continuous, so there is no need to make the stripe of the photosensitive element continuous in the vertical direction. Therefore, it is possible to sample the eigenfunction on y by interrupting the photosensitive element in the y direction. Only every other photosensitive element can be omitted and only The alternate blank columns, rather than having each photosensitive element in a continuous row, are now refilled with two sets of photosensitive elements in this special blank, which are designed to detect vertical motion. This is primarily a 21) comb-like array of symmetrical and y-direction axes, which have been described above with respect to Figures 3A and 3B. However, it should be noted that it is possible to construct a similarly staggered 2D array with axes of more than two directions. For example, Figure 9 shows a hexagonal array 902 that is staggered and routed to create three (3) spaces 120. The ID SQ-pair along the axes 904, 906 and 908. Referring to Figure 9, the grid of hexagonal photosensitive elements 9〇4 is routed to produce a 1D action along three different axes 904, 906 and 908. Photosensitive elements 904 for detecting in-phase signals and intersecting each axis are indicated by the same number j, 2 or 3. The alternate columns of photosensitive elements 9〇4 for detecting orthogonal signals are indicated by similar numbers 1, 2, and 3. From the edge of the photosensitive element, every axis, 9〇6, 106029.doc -28-1290221. The in-phase and quadrature signals of the 8 configuration are generally in phase (1) and quadrature (.) signals. And the I# payment number is not the photosensitive element _ ^ 丄 Therefore, one of the disadvantages of the county q interrupt array is the fact that the signal is sampled and the == number is affected. In particular, the interrupted array will have any spatial frequency of multiple potential periods of σ'. The influence of the fake signal can be difficult, &gt;, small, and the vertical 3 is reduced by the signal from the mother-photosensitive element.

S〇 w The output of each component and the copy to -1D:::Γ. This will increase the ratio of the samples (due to the use of each sensor, light: every second or third component in !), and also means using the -: first...like function (due to the sampling function) Not a series of her '2疋舁 step function winding' which inhibits a higher harmonic. ::Second: The pattern has a strong periodicity. For example, it is an optical pattern from the braided, my cloth or patterned surface, which is expected to have a false effect. A method of reducing the susceptibility to false signals (4) - a completely aperiodic array of life, and in particular, an array with no strong peaks at any spatial frequency. Such arrays can be obtained by using a pseudo-random distribution of photosensitive elements to confuse the periodicity, or by using a regular pattern that does not include periodicity. In this aperiodic array, the output from each of the photosensitive elements is typically required because there is no chance of zeroing the interlaced set of components with little or no cost. i A particularly interesting non-periodic pattern is the so-called leaf sequence array, or, to the hollyhock &quot; array. It has two interesting characteristics: it is based on the Golden Ratio, and the golden ratio is the strongest irrational number of all numbers, 106029.doc -29- 1290221 means the height of its higher harmonics in its spectrum. minimize. This is also quite easy to produce. In the polar coordinates, the first point is located at:

(P/, more 7) = [ C

(31) where 'Φ is the golden ratio 1618 ···. An embodiment of a dot pattern having a leaf sequence array 1002 of 2" elements 1004 is shown in FIG.

For this array, the optimum photosensitive element size is the Voronoi diagram (Wigner-Seitz unit) at the center of the photosensitive element. Thus, the pattern of photosensitive elements 1102 of array 11〇4 will thus appear similar to that shown in the figure. This leaf sequence array 1002 has no strong peaks in its Fourier spectrum, but the size of the photosensitive element 1004 has a rough average. When used to pattern a surface, it will therefore be resistant to false signals. By using the weighting coefficients calculated from the coordinates given above, it should be possible to extract any number of SQ-pairs from this array, such as the one shown in Figure 9, which has a possible limitation, ie for any: axis In contrast, the distribution of photosensitive elements that contribute to the motion of each axis is relatively sparse. The desired characteristics of the previously described 2-comb array are such that each element contributes to the action of the two axes by appropriately dividing the total number of inductive optical elements, without: assigning an independent weight to each element The coefficient, the Jinge wafer surface area and:: consumption, each element is given an independent weighting factor equivalent: expensive. Therefore, there is a need for a multi-axle array in which the mother of the photosensitive elements contributes to the motion signals from each axis. In detail, the concept can be applied to a 3-axis 歹J similar to the array described in Figure 9. In this 3-axis array, each photosensitive element contributes to the action of all three vehicles by 106029.doc -30-1290221 However, there are still only a small number of weighting coefficients that are applied only after the output of the component group or column is summed. Therefore, this embodiment is a 3-axis analog array of the above 2D comb array. Referring to Figure 12, there is shown a schematic diagram of a hexagonal array 1202 that is routed for 3-axis motion detection. In Fig. 12, each hexagon represents a single photosensitive member 12A, such as a photodiode. Each of the photosensitive elements 12A4 in the array 1202 is coupled to at least one of the three sets of signal lines 1206, 1208, and 1210 to detect motion in a direction perpendicular to the signal line. Therefore, the vertically oriented signal line 1206 is used to detect horizontal motion. It should be noted that each of the lines within the group of signal lines 12〇6, 1208, and 1210 exhibit a solid line pattern or a dashed line pattern. The solid lines are the group of the main signal line or the in-phase signal line, and the dotted lines are orthogonal signal lines. The symbol + and the weighting coefficients indicated as +1 and one are respectively indicated. If the photosensitive element 1204 is traversed by a line, it means that the element will contribute to the signal encoded by the line. In the illustrated embodiment, each photosensitive element 1204 is traversed by three lines associated with the three different sets of signal lines 丨2〇6, 1208, and 1210, which means that each element contributes The signal of each of the three axes. For example, the photosensitive element 1204 at the top of the hexagonal array 12〇2 contributes to a main signal of a group of signal lines 1206 having a weighting coefficient of one, which contributes to a signal with a weighting coefficient of +1. The main signal of the group of lines 12〇8, and contributes to the main signal of a group of signal lines 12 1〇 with a weighting coefficient of +1. The photosensitive element just below its lower right will contribute to the orthogonal signal of a group of signal lines 12 0 6 with a weighting coefficient of -1, a weighting coefficient of +1. 106029.doc -31· 1290221 line 12〇 The orthogonal signals of the group of 8 and the orthogonal signals of the group of signal lines 12 10 whose weighting coefficients are +1. So on and so forth. In theory, with three axes and two types of signals, and each of the two signals has two possible weighting factors, there will be 64 different types of I&quot; possibility. But in fact, because of some combinations, it will not come out; see, so there are only 16 possibilities. It can be seen that since the overall pattern is periodic, the thick black solid line (4) is one of the periodic patterns of the unit cell 1212, and there are only "components in the unit cell. Therefore, there are mainly 16 different flavors". , the photosensitive element axis, each element is characterized by applying a weighting coefficient of each of the two axes, and whether it is toward the main part or the orthogonal part of the axis. (It should be noted that the thick black dots indicate photosensitive elements having the same flavor or the same help for each axis.) Therefore, in the wiring mechanism, all the photosensitive elements 12〇4 of a given flavor are wired together to give one output « . The signal from each flavor can be split into three cases, and the weighting system is adapted to each of the three signals applied, and then the output signals are combined to the main signals of each of the three axes. And orthogonal signals. The above embodiments enable the collection of 3-axis information that can be combined to resist any-single-axis attenuation, and that each of the three axes can be used to give better results. And compared to the arrays in the optical sensors previously used in speckles, the soil resistance/, and the false twists have a more awkward/possible application of the above method to the square array of photosensitive (four), and: The four-axis education is obtained from the viewpoint of using different spatial frequencies in the array. In particular, the four axes can be derived from a square 2D array similar to the array of the image from (4) 106029.doc -32-1290221, which is generally obtained by adding more connections to the photosensitive elements. An embodiment of this will be described with reference to FIG.

Figure 13 shows a square 2D array 丨3〇2 and a wiring diagram for four-axis motion detection. The connection of the photosensitive element 1304 is similar to that described with reference to Figure 12, but there are four directions at this time, and each photosensitive element contributes to each of the main signals (in-phase) in each of the four directions or Orthogonal signal. The first set of signals 1306 is coupled to all of the photosensitive elements to detect motion in the horizontal direction. The second set of signal lines 1308 are connected to detect vertical movement, and the third set of signal lines 13 10 are connected to detect _45 in the vertical direction. The movement in the direction, and the fourth set of signal lines 13 12 are connected to detect +45 in the vertical direction. Private movement in the direction. It should be noted that the signal lines within the group of signal lines 13 1 0 and 13 12 are spaced closer than the signal lines within the group of signal lines 1306 and 1308. This indicates that the spatial frequency detected by it is different from and higher than the group of signal lines 1306 and 13〇8. Each of the lines within the group of signal lines 1306, 1308, 13 10, and 13 12 is again in a solid or dashed pattern. The solid lines are the main signal line or the in-phase signal line group, and the dotted lines are orthogonal signal lines. The symbols + and - are indicated as weighting coefficients of +1 and -1, respectively. The outline of the early unit cell 1314, and the thick solid line draws a periodic pattern. There are 16 photosensitive elements in the unit cell. Therefore, there are mainly 16 different &quot;flavored&quot; photosensitive S pieces 1300, each of which is applied to the main axis or the orthogonal part of the axis by the weighting coefficient applied to each of the axes And the thick black dots indicate the photosensitive elements having the same flavor or having four (4) help for each axis. After all the photosensitive elements i of the single-flavor combination are taken, the line-flavor signal is split into four cases, and Appropriate weighting factor—from 106029.doc -33 - 1290221 Leading to the main signal and the orthogonal signal in each of the four axes. It should be noted that this concept can be generalized to any one by superimposing multiple Ls on the array. Periodic array. For example, the grid owner extracts multiple price direction 4s, and two = one calculation without adding individual weighted h numbers. We can also based on other periodicity within the array. The array adds more directions; of course, the number of flavors of the unit cell will increase significantly.

In summary, a method of measuring the displacement within 2D using a bright_dark_case on a 2D array and various embodiments of such an array have been described. In general, the method uses a two-dimensional array of pixels or photosensitive elements. The array is connected in such a way that signal processing for a given spatial frequency can be simplified, which is extremely important for positioning shift measurement applications. . Various embodiments of the pixel connection mechanism can be constructed to allow for different (or more) spatial frequency geometries. This method allows speckle-based 2D displacement measurements to have lower energy requirements from signal processing electronics than 2D related types of devices, and does not compromise measurement as previously used with linear (1) comb array devices. </ RTI> </ RTI> </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; It is to be understood that many modifications, improvements and variations are possible within the scope of the invention as described in the above teachings. The scope of the invention is intended to be embraced by the scope of the invention and the scope of the invention. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 (previous technique) is one of four (4) photosensitive elements per cycle configuration 106029.doc -34-1290221 linear, one-dimensional (1D) comb array and associated cosine And a schematic block diagram of a sinusoidal template; FIGS. 2A through 2D are diagrams showing a matrix of cosine and sinusoidal distribution of a two-dimensional (2d) comb array according to an embodiment of the present invention; FIGS. 3A and 3B are diagrams according to the present invention. A schematic block diagram of a 2D comb array of the photosensitive elements of the embodiment constructed from Figures 2-8 to 2D and having photosensitive elements grouped by 4 X 4 elements per unit configuration;

4A and 4B are diagrams for comparing two vertical (or 1D χ 1D) linear comb arrays with a 2D comb array according to an embodiment of the present invention; FIG. 5 is a dispersion based on an embodiment of the present invention. Schematic block diagram of an optical navigation system for a 2d comb array; Figures 6A and 6B are optical navigation systems having a 2D comb array in accordance with an embodiment of the present invention at various speeds and on a variety of surfaces versus the system FIG. 7 is a schematic block diagram of a 2D comb array having photosensitive elements of 6 χ 6 elements per unit according to an embodiment of the present invention; Figure 8 is a schematic block diagram of an optical sensor having two 2D comb arrays arranged in quadrants in accordance with an embodiment of the present invention; Figure 9 is a wiring diagram for detecting according to another embodiment of the present invention. A schematic block diagram of an optical sensor having an array of hexagonal photosensitive elements along a 1D motion of three different axes; FIG. 10 is a non-periodic leaf for a photosensitive element in accordance with yet another embodiment of the present invention. Point pattern of the array; Figure U is Voronoi 106029.doc -35-1290221 FIG. 12 is an optical sensing device having a hexagonal 2D array of hexagons according to another embodiment of the present invention. FIG. 13 is a wiring and [main component symbol description] 102 1D comb array 104 photodiode 106 four-line signal 108 differential analog circuit 110 according to an embodiment of the present invention. Difference analog circuit 112 cosine template 114 sinusoidal template 302 2D comb array 304 photosensitive element 306 unit 308 difference amplifier 402 2D comb array 404 optical sensor 406 1D comb array 408 path 41 〇 path 502 optical navigation system 504 optical head diagram I06029.doc -36 - 1290221

506 light source 508 first/collimating lens 510 second/imaging lens 512 surface 514 2D comb array 516 aperture 602 dashed reference drawing reference numeral 702 2D comb array 704 photodiode 706 tj mouth early morning 708 matrix 802 optics 804 quadrant 806 quadrant 808 quadrant 810 quadrant 812 early 814 total wiring signal 816 differential amplifier 818 photosensitive element 902 hexagonal array 904 axis 906 axis 908 axis 106029.doc -37- 1002 1290221

1004 1102 1104 1202 1204 1206 1208 1210 1212 1302 1304 1306 1308 1310 1312 1314 Leaf sequence array photosensitive element photosensitive element array hexagonal array photosensitive element signal line signal line signal line early unit cell square 2D array photosensitive element signal line signal line signal Line signal line early unit cell 106029.doc -38-

Claims (1)

%年广月9日修(More 撝1 129(^®|l38_ Patent Application^ Chinese Application Patent Scope Replacement (May 96) Ten, Patent Scope·· 1 · An Optical Sensing Detecting relative movement between the sensor and the surface by detecting a change in optical characteristics of light reflected from a surface, the sensor comprising a two-dimensional (2D) comb of the photosensitive element An array comprising at least a first plurality of photosensitive elements configured to be adapted to sense a first combination movement along one of at least two non-parallel axes of a first set; and a second plurality a photosensitive element configured to be coupled to sense a second combination movement along one of at least two non-parallel axes of a second set; wherein the array is a 2-inch comb array, and wherein the first and the Waiting for a second plurality of photosensitive elements to be regularly spaced, and having a periodicity in at least a dimension of the 2D comb array; and wherein the sensor is a speckle-based sensor configured Reflecting on the surface by a continuous light source based on 2. The optical sensor of claim 1, wherein the first plurality of photosensitive elements are non-parallel along at least two of the first plurality of photosensitive elements. The axes are periodically arranged; and the first plurality of photosensitive elements are periodically disposed along at least two non-parallel axes of the second set. / 3. As requested by the optical sensors, wherein the first and the first The second plurality of photosensitive elements are regularly spaced and have a periodicity within two dimensions of the array. 4 · An optical sensor for detecting optical properties of light reflected from a surface by a 矣The change to sense the relative movement between the aligning surfaces of the sensor'. The sensor comprises one of the photosensitive elements, a difficult (2D) array, the array package 106029-960517.doc 1290221 Having at least a plurality of photosensitive elements configured to combine to sense a first combined movement along at least two non-parallel axes of a first set: and: a second plurality of photosensitive elements 'configured and _ The second combination movement of at least two non-parallel axes of the second group; Wherein the array is a non-periodic array, and wherein the first and second plurality of photosensitive elements are regularly spaced but not periodic. / 5. An optical sensor 'for sensing relative movement between the sensor and the surface by detecting a change in optical characteristics of light reflected from a surface'. The sensor comprises a photosensitive element a two-dimensional (2D) array comprising at least a plurality of first-plurality photosensitive elements configured and coupled to sense a movement of at least two non-parallel axes along a --set And a first plurality of photosensitive elements configured to be coupled to sense a second combination movement along one of at least two non-parallel axes of a second set; wherein the array is a quasi-periodic array, and wherein The first and the first plurality of photosensitive elements are regularly spaced and have a quasi-periodicity. 6. An optical sensor for sensing relative movement between the sensor and the surface by detecting a change in optical characteristics of light reflected from a surface, the sensor comprising a photosensitive element a two-dimensional (2D) array comprising at least a first plurality of photosensitive elements configured to be coupled to sense a first combination movement of one of at least two non-parallel axes along a group; and a second plurality of photosensitive elements configured to be coupled to sense a second combination movement along one of at least two non-parallel axes of a second set; wherein the first and second plurality of photosensitive elements comprise At least one shared photosensitive element configured to sense a corpse along the first and 106029-960517.doc -2- 1290221? Japanese repair (more) is replacing the purchase - -.......... The movement of at least two non-parallel axes of the second group. 7. An optical navigation system for use in a data input device for sensing displacement of the data input device relative to the surface by detecting changes in optical properties of light reflected from a surface, the system comprising An illuminator that illuminates a portion of the surface; An imaging optics that maps the illuminated portion of the surface to an optical sensor, wherein the sensor is a speckle-based sensor configured to be based on one a coherent light source that reflects to the surface and senses movement due to a change in a complex interference pattern created by the light reflected from the surface, and wherein the sensor comprises a two-dimensional (2D) comb array of photosensitive elements, The array includes at least a first plurality of photosensitive elements configured to be coupled to sense at least two non-parallel axes along the -th set - a first combined movement 'and a plurality of photosensitive elements' Configuring and coupled to sense a second combination movement along at least two non-parallel axes of the second group; and wherein the second and second plurality of photosensitive elements are regularly spaced and two-dimensionally in the array It has periodicity. First, wherein the first plurality of photoreceptor strips are non-parallel axes are periodically arranged, and at least two non-parallel axis circumferences of the second group of the fifth group are arranged as in the optical navigation system of claim 7 At least two second plurality of photosensitive elements of the first group are arranged in a phased manner. 106029-960517.doc 1290221, repair (U is a mixed page signal and a first orthogonal signal, and is used to combine the outputs of the second plurality of photosensitive elements into a second in-phase signal and a second 10. The optical navigation system of claim 9, wherein the front end circuit further comprises means for applying a weight to the output before the weighted outputs are totaled to the equivalent phase and the orthogonal signal. The optical navigation system of claim 7, wherein the imaging optics comprises an aperture configured to provide eschar imaging during the action to maintain the integrity of the speckle pattern and to average the size of the speckle with the comb 12. The optical navigation system of claim 7, wherein the illuminator is configured to illuminate the surface at a predetermined angle of incidence of the data input device that allows the debt to rise from the surface. This part. 13. A method of sensing the displacement of a speckle-based sensor relative to a surface (2D), the optical sensor comprising a two-dimensional (2D) comb of photosensitive elements shape a column comprising at least a first plurality of photosensitive elements and a first plurality of photosensitive elements, wherein the first and second plurality of photosensitive elements are regularly spaced and have a periodicity within two dimensions of the array The method comprises the steps of: illuminating a portion of the surface, mapping the illuminated portion of the surface to the array; sensing the -to-combination movement along a non-parallel axis of the strip with the first plurality of photosensitive elements; ^ Use "the first plurality of photosensitive elements to sense at least two of the second group 106089-960517.doc 1290221 *乂..........a "p ten" (3⁄4 positive non- The second combination of one of the parallel axes moves ~; and wherein the steps of sensing the movement along at least two non-parallel axes of the first and second sets comprise - based on - reflecting from the surface of the source to the surface And the step of sensing the movement from the change of the complex interference pattern established by the light reflected by the surface. The method of claim 13, wherein sensing at least two non-zero along the first and second groups The steps of the parallel axis movement Combining the outputs of the first plurality of photosensitive elements into a first in-phase signal and a first _ ^ and combining the outputs of the second plurality of photosensitive elements of the e-hai brothers into a first in-phase signal and a second orthogonal The method of claim 14, wherein the step of sensing movement along at least two non-parallel axes of the first and second groups further comprises summing the additional outputs to the The first and second in-phase and quadrature signals are applied to the steps of the output before the ~4 weight. 106029-960517.doc