finger print sensor(.pdf)

IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 50, NO. 4, APRIL 2003 1109

MEMS Fingerprint Sensor Immune to Various FingerSurface Conditions

Norio Sato, Member, IEEE, Katsuyuki Machida, Member, IEEE, Hiroki Morimura, Member, IEEE,Satoshi Shigematsu, Member IEEE, Kazuhisa Kudou, Masaki Yano, and Hakaru Kyuragi

Abstract—This paper describes a new fingerprint sensor thatdetects the topography of finger ridges and valleys. We propose amicroelectromechanical systems (MEMS) cavity structure for thepixels arrayed on the sensor surface and a fabrication process thatstacks the cavity structures on a CMOS LSI. A thin film on top ofthe cavity structures is bent by finger ridges mechanically, whichis detected by the sensing circuits below them electronically. Basedon an analytical model, we designed a cavity structure suitable forfingerprint detection. We fabricated the cavity structures on thesensing circuits by a fabrication process that entails gold electro-plating, sacrificial layer etching for cavities, and a sealing tech-nique. The fabricated fingerprint sensor consists of about 57 334pixels in the area of 11.2 mm 12.8 mm, which yields a high spa-tial resolution of 508 dots-per-inch (dpi). With these pixels workingtogether, fingerprint images were obtained. The sensor producesclear fingerprint images regardless whether the finger was dry orwet, which confirms its potential for various practical applications.

Index Terms—Fingerprint image, fingerprint sensor, LSI, mi-croelectromechanical systems, process.

I. INTRODUCTION

A S THE network society becomes more pervasive, highersecurity is required. Personal identification guarantees

that the person is authenticated as a proper user of networkterminals, such as mobile units. Personal identification byfingerprint has become attractive because fingerprint identi-fication is more secure than conventional methods based onpasswords and personal identification numbers (PINs). As a keydevice for fingerprint identification, semiconductor capacitivesensors have been developed using LSI interconnect technology[1]–[4]. The capacitive fingerprint sensor has an array of smallsensor plates and detects the capacitance between a fingersurface and the sensor plates. Capacitive fingerprint sensorshave some advantages. First, they are small and thin comparedto other optical fingerprint sensors, since they are fabricatedon semiconductor substrates. Second, the sensing circuits canmeasure small capacitances with little noise because they aredirectly below the sensor plates. The sensing circuits can alsodo various kinds of functional operations like signal processingto obtain the best fingerprint image corresponding to eachindividual [5]. Third, sensor reliability that is high enough

Manuscript received September 9, 2002; revised December 11, 2002. Thereview of this paper was arranged by Editor K. Najafi.

N. Sato, K. Machida, H. Morimura, S. Shigematsu, and H. Kyuragi are withthe NTT Microsystem Integration Laboratories, NTT Corporation, Atsugi,Kanagawa, 243-0198 Japan (e-mail: [email protected].)

K. Kudou and M. Yano are with the NTT Advanced Technology Corporation,Atsugi, Kanagawa, 243-0198 Japan.

Digital Object Identifier 10.1109/TED.2003.812490

for practical use has been established [6]. These advantagesoriginate from the fact that the sensing array is stacked on thesensing circuits on the semiconductor substrate. However, sincetheses sensors work on the principle of capacitance detection,they are too sensitive to finger surface conditions and humidityin the atmosphere. For example, it is difficult to obtain clearimages from fingers that are too dry or too moist without animage processing step. It is almost impossible to obtain animage of a finger wetted with water. The sensors can neitherbe used in the rain nor in extremely dry weather. This meansthat the applications of fingerprint sensors are restricted tomoderate finger surface conditions and moderate atmospherelike in the room.

Another proposed sensing technique detects the topographyof the finger surface to overcome the above problems [7], [8].These fingerprint sensors have arrays of small pixels. Each pixelitself is a capacitive pressure sensor. When a finger touches thesensor, each pixel detects the magnitude of the pressure fromthe finger ridges. The values output from every pixel compose afingerprint image. This sensing technique will enable the sensorto detect directly the topography of the finger surface regardlessof the finger surface conditions. However, theses sensors did notintegrate the sufficient large number of small pixels required fordetecting the topography of a finger. The sensor array consistedof 16 16 pixels at most. In addition, the moving parts of one ofthe sensors were not sealed, which made it vulnerable to waterthat invaded the cavity structures of the pixels.

We have developed a novel fingerprint sensor to solve theabove problems. The sensor uses a technique that detects thetopography of a finger directly. We propose a new MEMS struc-ture as a pixel. The MEMS structure has a mechanically mov-able part and sensing circuits directly below it detect the me-chanical movements electronically. In order to achieve a densearray of small pixels, the array of MEMS structures is stackedon the sensing circuits. This makes it possible to detect small ca-pacitances of a few femtofarads. We design the MEMS structureand construct a sensor fabrication process to realize the finger-print sensor.

This paper first describes the concept and sensing principle ofour fingerprint sensor, which we named as a MEMS fingerprintsensor. Next, the analytical model of the MEMS structure is de-scribed. Then, a fabrication process to stack the MEMS struc-tures directly on the sensing circuits is presented. In the process,a novel sealing technique, STP (Spin coating film Transfer andhot-Pressing) [9]–[11], is applied. Finally, we discuss the fabri-cation results and evaluate the sensor’s ability to obtain finger-print images regardless of whether the finger is dry or wet.

0018-9383/03$17.00 © 2003 IEEE

1110 IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 50, NO. 4, APRIL 2003

Fig. 1. (a) MEMS structure with cavity in the MEMS fingerprint sensor. (b) Itssensing mechanism.

II. MEMS FINGERPRINTSENSOR

We propose the MEMS fingerprint sensor as shown in Fig. 1.A finger touches the sensor surface where a large number ofsmall pixels are arrayed. The detailed structure of the sensor isshown in Fig. 1(a). Each pixel is separated by a grounded wallin an area of 50-m square. The pixels have MEMS structuresstacked on the sensing circuits. Each MEMS structure com-prises a protrusion, a cavity, a pair of electrodes, and a groundedwall. The upper electrode and sealing layer are made of metaland dielectrics, respectively, and are deformable thin films. Theprotrusions are also made of dielectrics. The upper electrode isgrounded through the grounded wall.

The sensor works as follows. The ridge of a finger surfacepushes the protrusion down, and the protrusion deflects theupper electrode as shown in Fig. 1(a). The protrusion transfersthe pressure from a finger to the center of the upper electrode.(Since the pixels are small compared to the width of fingerridges, it is difficult for the ridges to directly deflect the upperelectrodes.) The deflection of the upper electrode increasesthe capacitance between it and the lower electrode. Then, thecapacitance is converted into the output voltage of the sensingcircuit just under the lower electrode [5]. The value of theoutput voltage is translated into digitized signal levels. On theother hand, the valley of a finger surface does not push theprotrusion, and the capacitance is kept small. Therefore, thecapacitance under a ridge is larger than that under a valley.These values of the capacitance are translated into the digitizedsignal levels. With this sensing, the detected signals from allthe pixels generate one fingerprint image.

We had to design the MEMS structure and determine thestructural parameters so that the sensor would produce the max-imum capacitance change for a certain pressure from a fingersurface. We also had to develop a new sensor fabrication processbecause the MEMS structure is stacked on the sensing circuitsand has a cavity. It is necessary to fabricate the MEMS struc-

Fig. 2. Analytical model for numerical calculations of the MEMS structurefrom bird’s-eye in the Cartesian coordinates. Thez-axis is directed to the centerof the lower electrode.

ture without damaging the underlying sensing circuits and sealthe cavities to keep the water and contaminants out of the cavi-ties. Next, we describe the structural design and the fabricationprocess of the MEMS fingerprint sensor.

III. STRUCTURAL DESIGN

In order to check analytically whether the upper electrodebends enough, we modeled the mechanical and electrical dy-namics of the MEMS structure, and calculated two relation-ships: that between pressure from a finger and bending displace-ment, and that between the bending displacement and capac-itance change. Based on these relationships, the structure pa-rameters are determined that will produce sufficient capacitancechange to be detected by the sensing circuits.

A. Structure Modeling

The MEMS fingerprint sensor model is shown in Fig. 2. Theorigin is set at the center of the upper electrode. The upper elec-trode is assumed to be a square plane. The side length of theupper electrode is, and its thickness is. The lower electrodeis also a square plane with a side length of. The distance be-tween the upper and lower electrodes is. We assume that thepressure from a finger is loaded equally on the whole area ofthe top surface of the upper electrode at in the direc-tion of -axis.

Classical mechanics describes the bending displacementwhen a pressure is loaded. (See the Appendix for

details.) The bending displacement is proportionalto . The relationship between and capacitancechange is given as

(1)

where is permittivity in a vacuum. Next, we determine thestructural parameters of the MEMS structure so that the capac-itance change would be large enough to be detected by thesensing circuit.

Though the real structure has a protrusion, a sealing layer,and etch holes that are explained later, we considered the simplemodel described above as the first approximation. The modelincluded those factors are almost impossible to solve analyti-cally, and needs numerical calculations based on a finite-ele-

SATO et al.: MEMS FINGERPRINT SENSOR 1111

Fig. 3. (a) Calculated results of the bending displacement of the upper electrode. (b) Relationship between the pressure and bending displacement. (c) Relationshipbetween the pressure and capacitance change.

ment method. In order to discuss the relationships between, ,, and , we selected the simple model that can be expressed by

explicit analytical solutions.

B. MEMS Structure Analysis for Structure Determination

The bending displacement is a function of ,and the capacitance change is a function of and . Wefocus on four parameters,, , and . According to our pre-vious investigation [6], the measured maximum pressure of afinger among 100 fingers is 0.6 MPa. We consider a range offinger pressure from 0 to 2 MPa to be wide enough. A ridgeon a finger surface is about 100 to 400-m wide, so the size ofeach pixel is set to 50-m square to achieve sufficient resolu-tion. Considering that the width of the grounded wall is 8-m,the side length of the upper electrodeis 42 m. Thus, thesevalues are applied to the range ofand value of . The thicknessof the upper electrodeand the distance should be determinedbased on the analytical model described earlier.

The capacitance change calculation is explained here. Thebending displacement is numerically calculated asshown in Fig. 3(a) from the analytical solution of . InFig. 3(a), a pressure of 1 MPa and a thickness of 1 mare assumed. The upper electrode is gold, whose modulusof elasticity and Poisson’s ratio are Pa and 0.44,respectively. The bending displacement has a maximum at thecenter of the upper electrode . The relationshipbetween the pressureand bending displacement isshown in Fig. 3(b) when thicknessis 1 m. The bendingdisplacement is proportional to the pressure. Sincethe curved surface of the upper electrode is obtained,the capacitance between the upper and lower electrodes isdetermined. The relationship between capacitance changeand pressure is calculated as shown in Fig. 3(c) for thickness

of 1 m. Thus, we can predict the capacitance change for agiven pressure.

Using the relationship between the capacitance change andpressure, we determined the value ofand so as to producea sufficient capacitance change. We considered three require-ments in determiningand : In normal use, when a maximumpressure of 1 MPa is applied by a finger, the capacitance changeshould be as large as possible. The upper electrode should notcontact the lower electrode even if a pressure of 2 MPa is appliedas a worst case. The structure given by the parameters ofand

should be achievable in our fabrication process. For the firstrequirement, the capacitance change when the pressureis 1 MPa was calculated as a function ofand , and is shownwith contours in Fig. 4. The capacitance change becomeslarger as and becomes smaller. Whenand are too small,however, the upper electrode comes into contact with the lowerelectrode. In this case when (at MPa) is sat-isfied, the capacitance change can not be calculated. (Thisregion is shown with bold lines in the graph.) Within the regionwhere (at MPa) is satisfied, we consider theother two requirements so as to make the capacitance change aslarge as possible. In Fig. 4, the second requirement is also ex-pressed with the dotted line of (at MPa).We have to choose the most appropriateand from the regionwhere region (at MPa) is satisfied. Due tothe variations of the film thickness in each fabrication step,orthat is too small is not preferable. Taking this third requirementinto consideration, we set to 1 m and to 1 m as shownin Fig. 4. For these values ofand , the bending displacementof the upper electrode at with the pressure of 2 MPa is0.97 m as shown in Fig. 3(b). The upper electrode hardly con-tacts the lower electrode, even when strong pressure is appliedby a finger. The capacitance change is several femtofarads as


Fig. 4. Distance and thickness dependence of capacitance change. Thecontours show the capacitance change when the pressure is 1 MPa. In theregion with bold lines, the upper electrode contacts the lower electrode whenthe pressure is 1 MPa. The dotted line shows the limit where the upper electrodecontacts the lower electrode at the pressure of 2 MPa. The parameters wereselected from the region over the dotted line in this paper.

shown in Fig. 3(c), when a pressureof 1 MPa is applied. Inthis way, the structural parameters of the MEMS structure areproperly determined by the analytical model and the numericalcalculations.

IV. FABRICATION PROCESS

We have developed a CMOS-compatible MEMS fabricationprocess that includes the STP technique for sealing. In stackingthe MEMS structures, it is important that the MEMS fabrica-tion process does not deteriorate the underlying CMOS sensingcircuits. The MEMS structures must be sealed to prevent waterand contaminations from entering the cavities. To meet theserequirements, we select low temperature processes in order notto damage the underlying CMOS LSI and adopt a new sealingtechnique. We use gold electroplating to form thick films. STPseals the cavity structures to protect them from water and con-taminations. The fabrication process is illustrated in Fig. 5.

First, the sensing circuits are fabricated in the 0.5-m CMOSLSI and three-metal interconnection process [Fig. 5(a)]. Next,seed layers for electroplating are deposited by evaporatingAu–Ti on the dielectric fillm of SiN. The seed layers are each0.1- m thick. After resist patterning, the lower electrodesof 1- m thick are electroplated. Following that, the resistpattern is removed. The 2-m grounded wall is electroplatedin the same way. The lower electrodes and grounded wall arepatterned by wet-etching of the seed layers [Fig. 5(b)]. Then,the sacrificial layer of photosensitive polyimide is spin-coated.The top of the grounded wall is exposed by photolithography sothat it makes contact with the upper electrode that is fabricatednext [Fig. 5(c)]. And, the upper electrode is electroplated on itin the same way as the lower electrodes [Fig. 5(d)]. We madeetch-holes at the four corners of the upper electrode by coveringthe corners with resist material patterned like islands. There areetch-holes 5-m square at the four corners in each pixel. Then,the sacrificial layer is etched away through the etch-holes inorder to make cavities [Fig. 5(e)]. The etch-stoppers are goldor titanium. The oxygen radicals etch away the polyimide.

Fig. 5. Fabrication process of the MEMS fingerprint sensor.

Fig. 6. STP process. First, polyimide is spin-coated on a base film of EthyleneTetraFluoroEthylene copolymer (ETFE) as shown in (a). Next, the polyimide ishot-pressed against the sensor surface in a vacuum as shown in (b). Then, thebase film is peeled off, leaving the polyimide film on the sensor surface. Finally,the polyimide film is annealed at 310C, and the cavities are sealed as shownin Fig. 5(f).

(The gold or titanium of the upper and lower electrodes andgrounded wall do not react with the oxygen radicals.) Thecavities are sealed with a 1-m-thick sealing layer by usingthe STP technique [Fig. 5(f)]. The technique is developed forplanarization in the multilevel interconnects process for LSIs[9]–[11]. The process of STP for cavity sealing is shown inFig. 6. STP can seal vertical etch-holes while preventing thesealing material from flowing into the cavities. Finally, largeprotrusions about 10-m square and 10-m high are patternedusing photosensitive polyimide, and the polyimide film isannealed at 310 C. In this process flow, the MEMS structureis fabricated [Fig. 5(g)].


TABLE ICHIP SPECIFICATIONS

Fig. 7. Chip photograph.

Fig. 8. SEM photograph of the MEMS fingerprint sensor.

V. RESULTS

A. Fabricated MEMS Fingerprint Sensor

We fabricated the MEMS fingerprint sensor chip as shownin Fig. 7. The chip specifications are summarized in Table I.The chip has pixels in the sensing areaof 11.2 mm 12.8 mm. Each pixel is 50-m square and has asensing circuit with 102 transistors [5]. The scanning electronmicroscope (SEM) photograph in Fig. 8 shows that the MEMSfingerprint sensor has a lot of protrusion 50-m apart on thesurface. A magnified image of a pixel cross section obtained by afocused ion beam (FIB) is shown in Fig. 9. The MEMS structurewas fabricated properly as designed. The upper electrode does

not contact the lower electrode; it is positioned over the lowerelectrode with the proper spacing. The sealing layer seals thecavity; no sealing material flowed into it. The MEMS structureswere stacked on the sensing circuits. The achievement of thisMEMS structure with a cavity is a significant step forward inenhancing the ability of fingerprint sensing.

B. Fingerprint Images

We obtained a fingerprint image with the MEMS fingerprintsensor as shown in Fig. 10(a). The image demonstrated that thestructural design was appropriate and all the pixels and tran-sistors described in Table I worked together. This means thatthe MEMS structure was fabricated without destroying the tran-sistors in the sensing circuits; the MEMS structure fabricationprocess is compatible with the CMOS LSI and metal intercon-nection processes. After the finger in Fig. 10(a) was taken offthe sensor surface, the image in Fig. 10(b) was captured. Theimage is white, which means that the upper electrode returnedto its initial state after deflection.

We compared our MEMS fingerprint sensor with a conven-tional capacitive fingerprint sensor. Fingerprint images of a dryfinger are shown in Fig. 11. The fingerprint image from theMEMS fingerprint sensor in Fig. 11(a) is very clear. On theother hand, the raw image of the same finger from the capacitivefingerprint sensor in Fig. 11(b) is unclear. The MEMS finger-print sensor can clearly capture the image of a dry finger easilywithout any operation to enhance the quality of the image. Wealso captured a fingerprint image of a finger wetted with water.The MEMS fingerprint sensor captured the fingerprint imageas clearly as it did the normal finger, as shown in Fig. 12(a),while the capacitive fingerprint sensor could not as shown inFig. 12(b). The water filled the valleys of the finger surface,which made the capacitance at the valleys almost equal to thatat the ridges. Therefore, the image is quite dark in Fig. 12(b). Incontrast, the MEMS fingerprint sensor could detect the fingertopography directly regardless of the water in the valleys. Theseresults confirm that the MEMS fingerprint sensor is immune tofinger surface conditions and environmental conditions, whichmakes it suitable for wide practical use.

VI. SUMMARY

We proposed a novel MEMS fingerprint sensor with arrayedcavity structures stacked on CMOS sensing circuits. A MEMSstructure was devised to implement a principle for detectingthe topography of a finger. An analytical model of a MEMSstructure was described, and the optimized structural parame-ters were determined by numerical calculations. We developeda MEMS sensor fabrication process that is compatible with thestandard CMOS LSI process. The process includes the STPtechnique to seal the cavities of the MEMS structure. By usingthe process, we were able to stack the MEMS fingerprint sensoron a CMOS LSI. It was confirmed that the fabricated sensor de-tects the topography of a finger directly, regardless of the fingersurface conditions. This MEMS fingerprint sensor has the po-tential to widen the application of fingerprint identification forvarious people in various environments.


Fig. 9. FIB cross section of the pixel.

Fig. 10. Captured fingerprint images with finger (a) on the MEMS fingerprintsensor and (b) after taken off.

Fig. 11. Fingerprint images of a dry finger captured by (a) the MEMS and(b) a conventional capacitive fingerprint sensor.

Fig. 12. Fingerprint images of a finger wetted with water captured by (a) theMEMS and (b) a conventional capacitive fingerprint sensor.

APPENDIX

The formulation expressing the mechanical and electric dy-namics of the MEMS structure is described here. First, elasticdeflection of the upper electrode of the MEMS structure is ob-tained analytically from classical mechanics in a simple case

[12]. To express the bending displacement at the pres-sure , the flexural rigidity of the upper electrode is intro-duced

(A1)

where is a modulus of elasticity andis Poisson’s ratio of thematerial. The relationship between the deflection of the upperelectrode and pressure is given by

(A2)

where it is assumed that the bending displacement does not de-pend on . Equation (A2) is solved under the boundary condi-tions that all the four edges of the upper electrode are clampedby the grounded wall. The boundary conditions for the edge at

are

(A3)

(A4)

Equation (A4) means that the upper electrode does not rotatealong the edge at . In the same manner, the boundaryconditions for the other three edges are given. The analyticalsolution of (A2) under boundary conditions (A3) and (A4) isgiven by

(A5)

and

(A6)

(A7)


(A8)

where

(A9)

The constants can be determined from an infinite numberof the linear equations

(A10)

Since the absolute value becomes small as increases, itis sufficient to consider several terms at the beginning

(A11)

Thus, the bending displacement of the upper electrode atat the pressure is given. It is noted that the bending displace-ment is proportional to . We can determine thestructural parameters of the MEMS structure from the relation-ship.

ACKNOWLEDGMENT

The authors would like to thank T. Ogura for helpful supportand encouragement. They also would like to thank Y. Okazaki,H. Ishii, Y. Tanabe, S. Yagi, Y. Komine, and T. Kumazaki fortheir helpful discussion and fabrication.

REFERENCES

[1] N. D. Young, G. Harkin, R. M. Bunn, D. J. McCulloch, R. W. Wilks,and A. G. Knapp, “Novel fingerprint scanning arrays using polysiliconTFT’s on glass and polymer substrates,”IEEE Electron Device Lett.,vol. 18, pp. 19–20, Jan. 1997.

[2] M. Tartagni and R. Guerrieri, “A fingerprint sensor based on the feed-back capacitive sensing scheme,”IEEE J. Solid-State Circuits, vol. 33,pp. 133–142, Jan. 1998.

[3] D. Inglis, L. Manchanda, R. Comizzoli, A. Dickinson, E. Martin, S.Mendis, P. Silverman, G. Weber, B. Ackland, and L. O’Gorman, “A ro-bust, 1.8 V 250�W direct-contact 500 dpi fingerprint sensor,” inISSCCDig. Tech. Papers, Feb. 1998, pp. 284–285.

[4] S. Jung, R. Thewes, T. Scheiter, K. Goser, and W. Weber, “A low-powerand high-performance CMOS fingerprint sensing and encoding archi-tecture,”IEEE J. Solid-State Circuits, vol. 34, pp. 978–984, July 1999.

[5] H. Morimura, S. Shigematsu, T. Shimamura, K. Machida, and H.Kyuragi, “A pixel-level automatic calibration circuit scheme for sensinginitialization of a capacitive fingerprint sensor LSI,” inSymp. VLSICircuits Dig. Tech. Papers, June 2001, pp. 171–174.

[6] K. Machida, S. Shigematsu, H. Morimura, Y. Tanabe, N. Sato, N.Shimoyama, T. Kumazaki, K. Kudou, M. Yano, and H. Kyuragi, “Anovel semiconductor capacitive sensor for a single-chip fingerprintsensor/identifier LSI,”IEEE Trans. Electron Devices, vol. 48, pp.2273–2278, Oct. 2001.

[7] R. J. D. Souza and K. D. Wise, “A very high density bulk microma-chined capacitive tactile imager,” inProc. Transducers ’97, 1997, pp.1473–1776.

[8] P. Rey, P. Charvet, M. T. Delaye, and S. Abou Hassan, “A high densitycapacitive pressure sensor array for fingerprint sensor application,” inProc. Transducers ’97, 1997, pp. 1453–1456.

[9] K. Machida, H. Kyuragi, H. Akiya, K. Imai, A. Tounai, and A.Nakashima, “Novel global planarization technology for interlayerdielectrics using spin on glass film transfer and hot pressing,”J. Vac.Sci. Technol., vol. B 16, pp. 1093–1097, 1998.

[10] N. Sato, K. Machida, K. Kudou, M. Yano, and H. Kyuragi, “Advancedtransfer system for spin coating film transfer and hot-pressing in pla-narization technology,”J. Vac. Sci. Technol, vol. B 20, pp. 797–801,2002.

[11] N. Sato, K. Machida, M. Yano, K. Kudou, and H. Kyuragi, “Advancedspin coating film transfer and hot-pressing process for global planariza-tion with dielectric-material-viscosity control,”Jpn. J. Appl. Phys., vol.41, pp. 2367–2373, 2002.

[12] S. P. Timoshenko and S. Woinowsky-Krieger,Theory of Plates andShells, 2nd ed. New York: McGraw-Hill, 1959.

Norio Sato (M’02) was born in Tokyo, Japan, in1974. He received the B.S. and M.S. degrees inphysics from Tokyo University, in 1997 and 1999,respectively.

In 1999, he joined Nippon Telegraph and Tele-phone Corporation (NTT), Tokyo. He is now withNTT Microsystem Integration Laboratories, Kana-gawa, Japan. Since 1999, he has been engaged inthe research and development of the semiconductorfabrication process and MEMS.

Mr. Sato is a member of the Japan Society of Ap-plied Physics and the Physical Society of Japan.

Katsuyuki Machida (M’99) was born in Nagasaki,Japan, on April 16, 1954. He received the B.E., M.E.,and Dr. Eng. degrees in electronics engineering fromKyushu Institute of Technology, Kitakyusyu, Japan,in 1979, 1981, and 1995, respectively.

In 1981, he joined the Musashino ElectricalCommunication Laboratory, Nippon Telegraph andTelephone Public Corporation (NTT), Musashino,Tokyo, Japan. Since then, he has been engaged inthe research on ECR plasma CVD, the developmentof LSI process and manufacturing technologies. He

is now a Senior Research Engineer Supervisor with the NTT MicrosystemIntegration Laboratories, Atsugi, Kanagawa. He is currently engaged inresearch and development on the material and manufacturing technologies forMEMS.

Dr. Machida is a member of the Japan Society of Applied Physics.

Hiroki Morimura (M’96) was born in Saitama,Japan, on January 9, 1968. He received his B.E.degree in physical electronics and an M.E. degreein applied electronics from the Tokyo Institute ofTechnology, Japan, in 1991 and 1993, respectively.

In 1993, he joined Nippon Telegraph and Tele-phone Corporation (NTT), Tokyo. He is now at NTTMicrosystem Integration Laboratories, Kanagawa,Japan. He has been engaged in the research anddevelopment of low-voltage, low-power SRAMcircuits. He is currently doing research on sensing

circuits for CMOS fingerprint sensors and developing single-chip fingerprintsensor/identifier LSIs for portable equipment.

Mr. Morimura is a member of the Institute of Electronics, Information, andCommunication Engineers of Japan.


Satoshi Shigematsu(M’93) was born in Tokyo,Japan, on August 2, 1967. He received the B.S. andM.E. degrees in system engineering from TokyoDenki University, in 1990 and 1992, respectively.

In 1992 he joined Nippon Telegraph and Tele-phone Corporation (NTT), Tokyo. Since 1992 hehas been engaged in the research and developmentof low-voltage low-power CMOS circuit. He is nowwith NTT Microsystem Integration Laboratories,Kanagawa, Japan. His research interests includebiometrics sensor technology and low-power and

high-speed circuit design technique. He is currently doing research on parallelprocessing circuits for CMOS fingerprint identifier and developing single-chipfingerprint sensor/identifier LSIs and user authentication system.

Mr. Shigematsu is a member of the IEICEJ and Information Processing So-ciety of Japan.

Kazuhisa Kudou was born in Miyazaki, Japan, onMay 28, 1968. He received the B.E. degree in elec-tronics engineering from the North Shore College,Kanagawa, Japan, in 1989.

In 1989, he joined Nippon Telegraph and Tele-phone Technology Transfer Corporation (NTEC),Atsugi, Kanagawa. Since then, he has been engagedin the development of plasma etching and LSI fabri-cation process. His current work is the developmentthe thick film patterning technology for MEMS.He is now an Engineer with the NTT Advanced

Technology Corporation, Atsugi, Kanagawa.Mr. Kudou is a member of the Japan Society of Applied Physics.

Masaki Yano was born in Kanagawa, Japan, onMarch 11, 1967. He received the B.E degree inelectronics engineering from the North ShoreCollege, Kanagawa, Japan, in 1987.

In 1987, he joined Nippon Telegraph and Tele-phone Technology Transfer Corporation (NTEC),Atsugi, Kanagawa. Since then, he has been engagedin the development of the CVD and LSI fabricationprocess. His current work is the electroplating andthick film technologies. He is now a Staff Engineerwith the NTT Advanced Technology Corporation,

Atsugi, Kanagawa.

Hakaru Kyuragi was born in Fukuoka, Japan, onOctober 4, 1954. He received the B.E., M.E., and Dr.Eng. degrees in electronic engineering from KyotoUniversity, Kyoto, Japan, in 1978, 1980, and 2000,respectively.

He joined the Nippon Telegraph and TelephonePublic Corporation in 1980, where he has beeninvolved in the research of SR-excited photopro-cesses such as SiN film deposition and etching,an BiCMOS device process technology using SRlithography, and application of Si-based technology

to fingerprint sensor, MEMS, and millimeter-wave component module. He isnow Executive Manger in the Smart Devices Laboratory, NTT MicrosystemIntegration Laboratories. His current research interest is in the development ofhardware for ubiquitous service.

Dr. Kyuragi is a member of the Japan Society of Applied Physics, and theInstitute of Electronics, Information and Communication Engineers.

Documents

finger print sensor(.pdf)