An integrated heterojunction bipolar transistor cascode opto-electronic ...

1358

IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 47, NO. 7, JULY 1999

An Integrated Heterojunction Bipolar Transistor Cascode Opto-Electronic Mixer Yoram Betser, Jacob Lasri, Victor Sidorov, Shimon Cohen, Dan Ritter, Meir Orenstein, Associate Member, IEEE, Gadi Eisenstein, Fellow, IEEE, Alwyn J. Seeds, Fellow, IEEE, and Asher Madjar, Fellow, IEEE

Abstract— An integrated electrically pumped opto-electronic mixer consisting of two InP/GaInAs heterojunction bipolar transistors in a cascode configuration is demonstrated. Intrinsic downconversion gains of 18.2 and 8.9 dB at RF optical modulation frequencies of 3 and 9.5 GHz were obtained. The performance of the cascode mixer and a single heterojunction bipolar transistor (HBT) opto-electronic mixer are compared. The performance of the cascode mixer was superior to the single HBT mixer, mainly at high frequencies. Up and down mixing conversion gains were measured and found comparable. A simulation was carried out by solving the nonlinear differential equations that correspond to the large-signal equivalent circuit. The results of the simulation enabled us to identify the principal nonlinear components in the equivalent circuit. Index Terms—Cascode, conversion gain, mixer, opto-electronic, photo-HBT.

I. INTRODUCTION

O

PTO-ELECTRONIC mixers (OEM’s), based on InP/GaInAs heterojunction bipolar transistors (HBT’s), are attractive front-end components for optical subcarrier multiplexed systems [1], [2]. They exhibit a large inherent nonlinearity, excellent high-frequency performance, and high optical responsivity [3]. Due to their high optical responsivity, HBT OEM’s are much more efficient than OEM’s based on unipolar devices [4], [5]. The two major reasons for that are the thickness of the optical sensitive layer, which is much thicker in an HBT compared with an FET, and the intrinsic gain of the photo-induced current in an HBT. Recently, we have reported a single-stage three-terminal HBT opto-electronic mixer with an intrinsic down-conversion gain of over 10 dB [6]. In this paper, we report on the performance of an integrated opto-electronic mixer consisting of a cascode pair of InP/GaInAs HBT’s. The cascode stage provides wider bandwidth than a single HBT in a 50- system. We compare the cascode-mixing performance, as a function of frequency Manuscript received September 29, 1998; revised March 18, 1999. This work was supported by the U.K. Israel S&T Research Fund and by the Israel Ministry of Science Project 5858. Y. Betser was with the Department of Electrical Engineering, TechnionIsrael Institute of Technology, Haifa 32000, Israel. He is now with the Department of Electrical and Computer Engineering, University of California at Santa Barbara, Santa Barbara, CA 93106 USA. J. Lasri, V. Sidorov, S. Cohen, D. Ritter, M. Orenstein, and G. Eisenstein are with the Department of Electrical Engineering, Technion-Israel Institute of Technology, Haifa 32000, Israel. A. J. Seeds is with the Department of Electronic and Electrical Engineering, University College London, Torrington Place, London WC1E 7JE, U.K. A. Madjar is with RAFAEL, Haifa, Israel. Publisher Item Identifier S 0018-9480(99)05208-4.

Fig. 1. Schematic diagram of the epitaxial layer structure and mesa structure. The optical window is located on the base mesa.

and bias, with the performance of a single HBT OEM, fabricated on the same wafer. Since both up-conversion and down-conversion mixers are of interest for various applications, we have measured both conversion gains as a function of the base–emitter voltage and local oscillator (LO) power. The up-conversion and downconversion efficiencies were comparable. The experimental results are compared with the numerical solution of the nonlinear differential equations that model the opto-electronic mixer. The model enables us to identify the principal nonlinear elements in the equivalent circuit. II. DEVICE FABRICATION AND HIGH-FREQUENCY CHARACTERIZATION A schematic diagram of the epitaxial layer structure and mesa structure is shown in Fig. 1. The epitaxial layers were grown on a semi-insulating InP substrate by a compact metalorganic molecular beam epitaxy system [7]. The emitter Ti/Pt/Au metal served as a mask for wet etching of the emitter mesa. Self-aligned nonalloyed Pt/Ti/Pt/Au contacts were evaporated on the base and collector mesas. A 5 6 m opening in the base metallization served as an optical window. Ti/Au pads were evaporated after polyimide passivation and a curing process at 300 C for 1 h. The polyimide layer covering the optical window, measured by an atomic force microscope, was 450-nm thick. The layer structure of the HBT was: a and 250-nm InP 400-nm GaInAs subcollector, a 750-nm undoped GaInAs base, a collector, a 50-nm GaInAs emitter, and a 200-nm 150-nm InP GaInAs contact layer.

0018–9480/99$10.00  1999 IEEE

BETSER et al.: INTEGRATED HETEROJUNCTION BIPOLAR TRANSISTOR CASCODE OEM

1359

DC probe with a 120-pF capacitor to provide a radio frequency (RF) ground. A distributed feedback laser emitting at 1.55 m was externally modulated by a Mach Zender modulator. The modulated light was amplified using an erbium doped fiber amplifier (EDFA). The light was focused onto the optical window with the same microscope objective used for viewing the sample. The modulation index was measured by comparing the DC and RF photo-response of the base–collector diode 6.7 (without using the EDFA), and was found to be 21% dB). In the experiment, an average optical power, typically of about 15.3 dBm, was incident on the HBT’s optical window, photons/s at a wavelength suggesting a flow of about 2.3 of 1.55 m. IV. CONVERSION GAIN AND QUANTUM EFFICIENCY Fig. 2. Optical photograph and schematic diagram of a cascode opto-electronic mixer. The emitter and base dimensions were 4 11 m2 and 9 23 m2 ; respectively. The 5 6 m2 optical window is located on the base mesa of HBT-Q1.

2

2

2

Fig. 3. Experimental setup for measuring the opto-electronic cascode mixer and single HBT mixer.

An optical micrograph and the schematic diagram of the cascode mixer are shown in Fig. 2. The emitter and base and 9 23 respectively. The dimensions were 4 11 optical window was located on the base mesa of 5 6 HBT-Q1. The light detection, mixing, and amplifying were performed by HBT-Q1 while HBT-Q2 served as a low-inputresistance unity-current-gain amplifier. Small-signal -parameters of a single device and the casand code pair were measured up to 40 GHz. The obtained of the single device were 70 and 50 GHz, respectively, mA and V. The cascode pair and at were 60 and 30 GHz at mA, V, V. and III. EXPERIMENTAL ARRANGEMENT A schematic diagram of the experimental arrangement is shown in Fig. 3. The cascode pair and the single HBT were connected in common-emitter configuration. The base of HBTQ1 was the input port, and the collector of HBT-Q2 was the output port. The measurements were carried out on-wafer using 40-GHz GSG probes. A 50- LO source and a DC voltage source were connected via bias-T to the base port. The output power was measured using a spectrum analyzer. The base of HBT-Q2 was connected to a DC voltage source using a

Both the intrinsic and extrinsic conversion gains are useful figures of merit for OEM’s [6], [8]. The intrinsic conversion is defined as the ratio of the output power gain for down conversion and for up conversion) to the primary photo-detected RF power. is the photo induced RF electrical power detected by the base–collector junction without amplification. It was measured by shorting the base–emitter junction of the photodetector transistor. The is defined as the ratio of the extrinsic conversion gain output power of the up- or down-converted signal to the that would have been equivalent electrical RF power detected by an ideal photo diode with equal load resistance. The relation between the incident peak modulated component and the equivalent electrical input of the optical power [8], where is the power is is the photon energy, and electron charge, and are related by the external quantum efof the base–collector photo diode, thus ficiency The external quantum efficiency was measured from the DC photoresponse of the base–collector junction and was or 10.8 dB. This result agrees with a calculation assuming an absorption depth of 1.5 m and 30% reflection. V. EXPERIMENTAL RESULTS The electrical responses of the cascode pair and the single HBT to the modulated optical signal were first measured as a function of the frequency modulation of the optical signal. The electrical frequency response of the base–collector PIN photodiode to the modulated optical signal served as a reference for calculating the intrinsic signal gain, excluding the effects of the external quantum efficiency of the HBT. The results shown in Fig. 4 clearly demonstrate the superior frequency response of the cascode pair, compared to the single HBT in a 50- configuration. The reason for the wider bandwidth of the cascode pair is the elimination of the parasitic time-constant associated with the base–collector capacitance and the 50- load resistance of the spectrum analyzer. The use of a cascode stage is, therefore, of great importance for 50broadband applications. The deviation of the measured data from the 20 dB/decade slope was probably due to inaccurate calibration.

1360


Fig. 4. Optical response as a function of frequency of a cascode pair HBT’s : mA and VCE : and a single HBT. For the single HBT: IC V. For the cascode pair: IC mA, VB 2 : V, and VC : : V. The PIN base–collector junction served as a reference photodetector for calculating the intrinsic power gain.

= 12 9

= 16 7 = 22

= 15 = 28

The intrinsic down-conversion gain for an intermediate frequency (IF) of 0.5 GHz of the cascode pair mixer and the single HBT mixer at RF frequencies of 3 and 9.5 GHz is plotted in Fig. 5 as a function of the base–emitter voltage. The LO power was adjusted for optimum performance at an optimum value of base–emitter voltage. The IF power was then measured as a function of the base–emitter voltage, keeping the LO power constant at that optimal value. For the 3-GHz RF signal, the optimal LO power was 10 dBm for both devices. For the 9.5-GHz RF signal the optimal LO power was 3 and 3 dBm for the cascode pair and the single HBT, respectively. For the 3-GHz RF signal, the performances of the cascode pair and the single device were comparable. The optimum down-conversion gains were 12.7 and 11.2 dB for the cascode pair and the single device, respectively. Both devices exhibited an optimum with respect to the base–emitter voltage and LO power as reported in [6]. The cascode mixer behavior in the 0, was similar to that of a saturation regime, i.e., at VBC single HBT mixer also consistent with [6]. For an RF signal of 9.5 GHz [Fig. 5(b)], the downconversion gain of the cascode-pair mixer was superior to of 8.9 dB, that of the single HBT, with an optimum compared with 4.4 dB. The consumption of LO power was also lower. The improved performance of the cascode-pair OEM was more pronounced in up-conversion experiments. The intrinsic up-conversion gains of the cascode-pair mixer and the single HBT mixer are plotted in Fig. 6 as a function of base–emitter voltage. The RF frequencies and LO power levels were the same as in the down-conversion experiments. At an RF frequency of 3 GHz [Fig. 6(a)], the cascode mixer had 2.4dB higher optimum up-conversion gain than the single HBT OEM. At an RF frequency of 9.5 GHz [Fig. 6(b)], the cascode mixer had an optimum up-conversion gain higher by 7.6 dB than the single HBT mixer. In the experiment described above, the isolation between the RF current source, representing the response of the HBT to the modulated optical signal, and the 50- LO source was poor. Some of the RF power (or RF current) leaked from the base terminal into the 50 LO source, degrading the mixing

(a)

(b) Fig. 5. Intrinsic down-conversion gain versus base–emitter voltage, measured for a cascode opto-electronic mixer and a single HBT opto-electronic mixer. (a) Optical modulation frequency was 3 GHz and LO frequency was 3.5 GHz with LO power of 10 dBm for both devices. (b) Optical modulation frequency was 9.5 GHz and LO frequency was 10 GHz, with LO power of 3 dBm for the cascode mixer and 3 dBm for the single HBT mixer. For the single HBT mixer: VCE : V. For the cascode pair: VB 2 V and VC : V.

0

0

= 25

+ = 22

=2

efficiency of the mixers [6]. In order to eliminate some of this parasitic effect, a three-stub tuner was inserted between the LO source and the base input port. By a proper adjustment of the three-stub tuner, the down-conversion gain of the 3GHz signal increased by 6.2 and 5.5 dB for the single HBT mixer and the cascode mixer, respectively. The LO output power changed by less than 0.4 dB. Thus, it was verified that the increase in the measured conversion gain was not due to a better matching of the LO source but due to a better isolation between RF and LO signals. This was further verified by measuring the reflection coefficient of the three-stub tuner at the optimum condition, showing a 1.3-dB reflection at a 3-GHz modulation. The obtained intrinsic down-conversion gain of the cascode-pair mixer was thus 18.2 dB and the extrinsic down-conversion gain 7.4 dB. This result compares and of 10.4 favorably with previously reported and 5.5 dB, respectively, for a single three-terminal HBT of 7 dB for a 2-terminal HBT OEM [8]. OEM [6] and


1361

(a)

(a)

(b) Fig. 7. (a) HBT large-signal model. (b) Schematic model of the HBT OEM measurement setup for large-signal simulations.

(b) Fig. 6. Intrinsic up-conversion gain versus base–emitter voltage, measured for a cascode opto-electronic mixer and a single HBT opto-electronic mixer. (a) Optical modulation frequency was 3 GHz and LO frequency was 3.5 GHz with LO power of 10 dBm for both devices. (b) Optical modulation frequency was 9.5 GHz and LO frequency was 10 GHz, with LO power of 3 dBm for the cascode mixer and 3 dBm for the single HBT mixer. For the single HBT mixer: VCE : V. For the cascode pair: VB 2 V and VC : V.

0

0

= 25

+ = 22

=2

Note that a three-stub tuner is certainly not an appropriate solution for wide-band applications. However, it was used in this experiment in order to demonstrate the intrinsic efficiency of the HBT-based mixers. The above results indicate that more sophisticated circuits, such as the Gilbert cell [9] configuration, are required for optimal performance of OEM’s. VI. MODELING Previously, a PSPICE-based large-signal model was employed to model the mixing performance of a single HBT OEM [6]. Two main nonlinear mechanisms were identified: the emitter–current dependence of the dynamic emitter resistance, and the transition from linear mode operation into the saturation mode. The model predicted very well the base–emitter dependence of the mixing efficiency. Here, we make use of the measured ratio between the down-conversion and up-conversion gains of the devices to obtain a better understanding of the prominent nonlinear effects in the OEM operating in the linear mode. We illustrate our findings using

a large-signal -model. A schematic diagram of the model is shown in Fig. 7(a). This model is similar to the conventional charge control PSPICE large-signal model in the active region. Neglecting the base–collector capacitance, the OEM represented by this model is composed of a nonideal input mixing stage and an ideal amplifier stage with current amplification. The nonlinear input capacitance is where is the emitter to the saturation current, the thermal collector delay time, the time-dependent base–emitter voltage. voltage, and The base–collector and base–emitter junction capacitance, and are assumed to be constant. This model does not take into account saturation effects due to a voltage drop on the 50- load at high collector currents, which results in forward biasing of the base–collector junction. Neglecting the base–collector capacitance, the frequency response of the HBT is determined only by the input impedance. A first-order approximation is that the only frequency components present are the harmonics of the LO and RF signals. In in this case, the down-conversion and up-conversion currents that flow through the diode have exactly the same amplitude. These at the output, and therefore the currents are multiplied by output power levels of both signals, predicted by this model with the above assumptions, are the same. Thus, an important conclusion is that up-conversion efficiency is expected to be comparable with down-conversion efficiency, which makes the HBT OEM an attractive device for applications where up-conversion is needed. A better approximation is that the down-conversion and up-conversion frequency components The amplitudes and phases of are also present in

1362


TABLE I VALUES OF THE EQUIVALENT CIRCUIT COMPONENTS USED IN THE CALCULATIONS

conversion gain. The calculations were performed for RF modulation frequency of 9.5 GHz and LO frequency of 10 GHz, where the measured ratio between down-conversion gain and up-conversion gain was sufficiently large. First, and were set to be zero, so that the diode remained the was only nonlinear element in the circuit. The value of set to 0.6 pF in order to maintain the correct small-signal frequency response of the HBT. The calculated result was that the down-conversion and up-conversion currents at the HBT output were equal to within 0.2 dB. We therefore conclude that the observed ratio between the up-conversion and downconversion gains was due to the nonlinear component of Setting ps and fF, the down-conversion gain was found to be about 4 dB larger than the up-conversion was set to 43 fF, the ratio between gain. When, in addition, down-conversion and up-conversion output powers increased can be understood to almost 6 dB. The nonlinear effect of is multiplied by the noting that due to the Miller effect voltage gain between collector and base terminals, which is is bias dependent. In the cascode OEM, the effect of reduced, as is found experimentally. VII. CONCLUSION

Fig. 8. Down-conversion and up-conversion gain of the single HBT mixer as a function of collector current. Comparison between experimental results and a simulation based on the -model. The optical modulation frequency was 3 GHz, the LO frequency 3.5 GHz, LO power 10 dBm, and VCE 2:2 V.

0

=

the down-conversion and up-conversion components of may not be the same due to the effect of the input network, and thus, some difference between down-conversion and upconversion efficiencies is expected. The differential equations (given in the Appendix) that represent the schematic circuit diagram shown in Fig. 7(b) were solved numerically using the MATLAB/SIMULINK software. The down-conversion and up-conversion gains of the single HBT OEM were calculated for RF modulation frequency of 3 GHz and LO frequency of 3.5 GHz. The LO power was 10 dBm. The parameters used in the simulations are listed in Table I. They were extracted from DC and -parameter measurements of the device [10]. The calculated and experimental results are shown in Fig. 8 as a function of collector current. The simulated results correspond well to the measured data for low to moderate values of collector currents. The measured ratio between down-conversion gain and up-conversion gain is, however, slightly larger than the simulated one. For high collector currents, the model overestimates the experimental data because it does not take into account the transition into the saturation region of the transistor characteristics. We solved the differential equations describing the above model for different sets of parameters in order to identify the main nonlinear components in the equivalent circuit that account for the measured ratio between the up- and down-

A cascode HBT opto-electronic mixer was demonstrated for the first time. The down-conversion and up-conversion gain was up to 7.6-dB higher than that for a comparable single HBT opto-electronic mixer. The maximum intrinsic and extrinsic down-conversion gains were 18.2 and 7.4 dB, respectively, for an RF optical-intensity modulation frequency of 3 GHz and LO frequency of 3.5 GHz. The intrinsic up-conversion gain varied from 6.9 to 5.7 dB for output frequencies in the range of 6.5–19.5 GHz, respectively. A simulation of the mixing performance was carried out by numerically solving the nonlinear differential equations describing the opto-electronic mixer. The principal nonlinear elements of the equivalent circuit were identified to be the nonlinear input impedance of the phototransistor. APPENDIX In order to calculate the performance of a single HBT OEM, the following set of differential equations was solved using the MATLAB/SIMULINK software: (A1)

(A2)

(A3) (A4) The above differential equations describe the schematic circuit diagram shown in Fig. 7(b). To simplify the differential


equations, it was assumed in the simulations that the spectrum analyzer is connected directly to the collector port and not via a bias-T, as shown in Fig. 7(b). This does not affect the calculations because the DC bias at the collector has no effect on the model, as it does not take into account saturation mF and mH are the capacitance effects. and inductance of the bias-T connected to the base port, 50 is the input resistance of the respectively, and is the voltage at LO source and the spectrum analyzer. the the output of the LO source (after the 50- resistor), the base DC bias, the voltage at the base port, the voltage at the collector port and base DC current, the collector DC bias. The LO voltage is given The optical generated RF current, by: is given by: The values of the model components taken in the calculations are shown in Table I.

1363

Jacob Lasri was born in Haifa, Israel, on February 22, 1971. He received the B.A. degree in physics, and the M.Sc. degree in electrical engineering, both from the Technion-Israel Institute of Technology, Haifa, Israel, in 1995 and 1998, respectively, where he is currently working toward the Ph.D. degree in the Electrical Engineering Department. His current interests are in the field of microwave photonics and optical communication.

Victor Sidorov received the M.Sc.(Hons.) degree in chemistry from Voronezh University, Russia, in 1985. He is currently working toward the D.Sc. degree in chemical engineering at the Technion-Israel Institute of Technology, Haifa, Israel. Since 1995, he has been with Microelectronics Research Center, the Technion, as a Senior Research Assistant, with primary responsibilities involving microwave and optoelectronic devices and circuit manufacturing based on III–V semiconductors processing. Prior to this, he was an Engineer Researcher with Mizur Micromechanics Technologies Limited, responsible for microsensors manufacturing and micromachinning. His current research interest is in III–V semiconductor surface coatings and passivation.

REFERENCES [1] T. E. Darcie, “Subcarrier multiplexing for multiple-access lightwave networks,” J. Lightwave Technol., vol. LT-5, pp. 1103–1110, Aug. 1987. [2] R. Olshansky, V. A. Lanzisera, and P. M. Hill, “Subcarrier multiplexed lightwave systems for broad-band distribution,” J. Lightwave Technol., vol. 7, pp. 1329–1342, Sept. 1989. [3] S. Chandrasekhar, L. M. Lunardi, A. H. Gnauck, R..A. Hamm, and G. J. Qua, “High speed monolithic pin/HBT HPT/HBT photoreceivers implemented with simple phototransistor structure,” IEEE Photon. Technol. Lett., vol. 5, pp. 1316–1318, 1993. [4] C. Rauscher and K. J. Williams, “Heterodyne reception of millimeterwave-modulated optical signals with an InP-based transistor,” IEEE Trans. Microwave Theory Tech., vol. 42, pp. 2027–2034, Nov. 1994. [5] L. E. M. jr. de Barros, A. Paolella, M. Y. Frankel, M. J. Romero, P. R. Herczfeld, and A. Madjar, “Photoresponse of microwave transistors to high-frequency modulated lightwave carrier signal,” IEEE Trans. Microwave Theory Tech., vol. 45, pp. 1368–1374, Aug. 1997. [6] Y. Betser, D. Ritter, C. P. Liu, A. J. Seeds, and A. Madjar, “A single-stage three-terminal heterojunction bipolar transistor optoelectronic mixer,” J. Lightwave Technol., vol. 16, pp. 1–5, Mar. 1998. [7] R. A. Hamm, D. Ritter, and H. Temkin, “A compact MOMBE growth system” J. Vac. Sci. Technol., vol. A12, pp. 2790–2794, 1994. [8] C. P. Liu, A. J. Seeds, and D. Wake, “ Two-terminal edge-coupled InP/InGaAs heterojunction phototansistor optoelectronic mixer,” IEEE Microwave Guided Wave Lett. , vol. 7, pp. 72–74, Mar. 1997. [9] P. Weger, G. Schultes, L. Treitinger, E. Bertagnolli, and K. Ehinger, “Gilbert multiplier as an active mixer with conversion gain bandwidth of up to 17 GHz,” Electron. Lett., vol. 27, no. 7, pp. 570–571, 1991. [10] S. J. Spiegel, D. Ritter, R. A. Hamm, A. Feygenson, and P. R. Smith, “Extraction of the InP/GaInAs heterojunction bipolar transistor small signal equivalent circuit,” IEEE Trans. Electron. Devices, vol. 42, pp. 1059–1064, June 1995.

Yoram Betser was born in Israel in 1965. He received the B.Sc., M.Sc., and Ph.D. degrees in electrical engineering from the Technion-Israel Institute of Technology, Haifa, Israel, in 1992, 1994, and 1998, respectively. His doctoral research involved device physics and microwave opto-electronic modeling and applications of InP/GaInAs HBT’s. He recently joined the Faculty of Electrical and Computer Engineering, University of California, Santa Barbara, CA, for conducting post-doctoral research. His current research interests include the area of fast LSI mixed digital/analog circuits in InP/InGaAs substrate-transferred technology and high-frequency device modeling.

Shimon Cohen, photograph and biography not available at the time of publication.

Dan Ritter, photograph and biography not available at the time of publication.

Meir Orenstein (M’95–A’96), photograph and biography not available at the time of publication.

Gadi Eisenstein (S’80–M’80–SM’90–F’99) received the B.Sc. degree from the University of Santa Clara, Santa Clara, CA, in 1975, and the M.Sc. and Ph.D. degrees from the University of Minnesota at Minneapolis in 1978 and 1980, respectively. In 1980, he joined the Photonic Circuits Research Department of AT&T Bell Laboratories as a member of Technical Staff, where his research involved diode laser dynamics, high-speed optoelectronic devices, optical amplification, optical communication systems, and thin-film technology. In 1989, he joined the faculty of the Technion-Israel Institute of Technology, Haifa, Israel, where he is currently Professor of Electrical Engineering and Head of the Barbara and Norman Seiden Advanced Optoelectronics Center. His current research activities are in the fields of fiber-optics systems and components for such systems, dynamics of quantum-well lasers, nonlinear semiconductor optical amplifiers, and compact short-pulse generators. He has published over 200 journal and conference papers, lectures regularly in all major fiber optics and diode laser conferences, and serves on numerous technical program committees.

A. J. Seeds (M’81–SM’92–F’97), for photograph and biography, see this issue, p. 1264.

1364


Asher Madjar (S’77–M’79–SM’83–F’97) received the B.Sc. and M.Sc. degrees from the TechnionIsrael Institute of Technology, Haifa, Israel, in 1967 and 1969, respectively, and the D.Sc. degree from Washington University, St. Louis, MO, in 1979. Since 1969, he has been with RAFAEL, Haifa, Israel, and with the Technion. At RAFAEL, he performed research in the areas of passive and active microwave devices. He headed the MIC Group from 1973 to 1976, served as a microwave Chief Engineer in the Communications Department from 1979 to 1982, and as Chief Scientist of the microwave Department from 1982 to 1989, with direct responsibility for the MMIC group from 1987 to 1989. From 1989 to 1991, he was a visiting Professor at Drexel University, Philadelphia, PA, where he performed research on optical control of microwave devices and developed a comprehensive model for the optical response of the MESFET and also participated in graduate student instruction and taught a course on microwave devices. Presently, he is a Research Fellow involved in microwave optoelectronics activity, MMIC, monolithic circuits combining microwave and optical devices, and microwave modules. Additionally, at the Technion and the Ort Braude College, he teaches courses on microwaves, passive microwave devices, active microwave devices, and transmission and receiption techniques, and serves as an instructor for graduate students. He is the author or co-author of over 100 papers in the areas of microwave components and devices, MIC, MMIC, linear and nonlinear microwave circuits (harmonic balance, APFT, etc.), microwave device modeling (including optical effects), optical links at microwave frequencies, and more. Dr. Madjar served as the Israel S-AP/MTT chapter Chairman for several years, organizing 13 symposia. From 1985 to 1989, he served as Secretary of the Israel Section of IEEE. He served on the technical committees for MELECON for the 14th–19th conventions of Electrical and Electronics Engineers in Israel. He has served as a member of the management committee of the European Microwave Conference since 1990 and as a member of the technical program committee for the European Microwave Conferences since 1993. In April 1996, he was elected to the newly created Steering Committee of the European Microwave Conference. He served as the chairman of the 27th European Microwave Conference held in Jerusalem, Israel, in September 1997. In 1998, he was awarded the RAFAEL Best Researcher Award.