Forum for Electromagnetic Research Methods and Application Technologies (FERMAT) Electromagnetic Analysis and Circuit Modeling of Ultra-Wideband SPST Switch in 65-nm CMOS Technology Anak Agung Alit Apriyana(1), Jin He(2), and Yue Ping Zhang(3) Virtus IC Design Centre, Nanyang Technological University (1) (Email: E090055@ e.ntu.edu.sg) (2) School of Physics and Technology, Wuhan University (Email: jin.he@whu.edu.cn) (3) School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore (Email: EYPZhang@ntu.edu.sg) the distributive and coupling effects within the structure. Section III describes how to extract extrinsic parasitic inductances, capacitances and resistances of the switch structure for circuit modeling. Section IV compares the simulated and modeled with the measured results of the SPST switch, validating the electromagnetic analysis and circuit modeling. Finally, Section V concludes the paper. Abstract—This paper focuses on electromagnetic analysis and circuit modeling of an ultra-wideband single-pole single-throw (SPST) switch in a 65-nm CMOS technology up to millimeter-wave frequencies. We begin by carrying out the electromagnetic analysis using the HFSS, which is a three-dimensional full-wave field solver. A novel technique to simulate an on-state switching transistor is proposed and highlighted. Then, the circuit modeling is performed by including all extracted extrinsic parasitic inductances, capacitances, and resistances, especially the drain-to-source parasitic capacitances introduced by the overlapped multi-finger metalization of the transistors. Next, the electromagnetic analysis and circuit modeling are validated with the measurements of the SPST switch over the frequency ranges from 1 to 110 GHz and 140 to 170 GHz, respectively. Finally, we point out that although the electromagnetic analysis can provide a deep physical insight into the switch behavior, it has an inherent limitation, namely it cannot handle the devices for the electron transfer phenomena such as the hot carrier effect, well proximity effect, negative bias temperature instability, and etc. II. ELECTROMAGNETIC ANALYSIS A single-pole-double-throw (SPDT) switch that consists of 2 transistors, 2 resistors, and 4 transmission-line sections was simulated in [13] by using the Sonnet, a two-dimensional (2D) full-wave electromagnetic field solver. Unfortunately, no electromagnetic analysis was carried out to gain physical insight into the switch behavior [13]. In this Section, the SPST switch that consists of 2 transistors, 2 resistors, 1 inductor, 8 transmission-line sections, and 7 testing pads is simulated using the HFSS, a three-dimensional (3D) full-wave electromagnetic field solver. More importantly, the distributive and coupling effects within the SPST structure are discussed for in-depth understanding of the switch characteristics. Index Terms—65-nm CMOS, 60 GHz, millimeter wave, switch. A. The SPST Switch Fig. 1 shows the schematic and layout of the SPST switch in a 65-nm CMOS technology. The transistors M1 and M2 are used to turn-on or turn-off the switch, depending on the state of the control voltage (VCTRL). When VCTRL is low, both transistors M1 and M2 are off. Hence, the signal at Port 1 can be passed to Port 2 through the inductor L1. On the other hand, when the VCTRL is high, both transistors M1 and M2 are on and thereby shorting both ports to the ground and thus ideally there is no signal passed through from Port 1 to Port 2. The poly resistors (RG1 and RG2) are used to bootstrap the gate at RF. By bootstrapping the gate, it will be floating at RF and hence its gate-to-source and gate-to-drain potential will be relatively constant. The inductor L1 is used to improve the input and output matching by compensating the shunt capacitance effects that are introduced by the transistors M1 and M2 during their off-state. I. INTRODUCTION M ILLIMETER-WAVE technologies are very promising for multi-gigabit wireless communications at 60 GHz, sensing and detection at 77 GHz, high resolution imaging at 94 GHz, and wireless chip area networks above 100 GHz. The interest in these applications has led to extensive research in millimeter-wave CMOS circuits and systems [1-6]. This paper reports a SPST switch, a basic circuit to build single-pole-multiple-throw (SPMT) switches for time-domain-duplexing transceivers, switched-beam arrays, and multiple-input and multiple-output systems [7-15]. The SPST switch designed in a 65-nm CMOS exhibits ultra-wideband characteristics up to millimeter-wave frequencies. Section II of the paper addresses how to build and simulate the SPST switch in the HFSS and more importantly discusses 1 Forum for Electromagnetic Research Methods and Application Technologies (FERMAT) rectangles/boxes. Lump-port excitation is used at the input and output. Radiation boundaries surround the SPST switch with horizontal separations of λ0/8 (λ0 is the wavelength at 60GHz) from each edges and vertical separations of λ0/8 from the top and bottom surface. Since the ground probing planes of the input or output are connected, we can apply two-port excitation as shown by the arrow directing from the ground to Port 1 and Port 2. L1 Port 1 Port 2 96 pH M1 RG1 RG2 M2 10 k 10 k VCTRL 0 / 1.2V (a) GND GND (G) GND GND (G) L1 Port 1 L1 Port 2 (S) + Port 1 (S) + Port 2 M1 GND M2 RG2 RG1 GND VCTRL GND (G) VCTRL GND (G) (b) Fig. 1. The SPST switch: (a) schematic and (b) layout. B. HFSS Simulation The 65-nm CMOS process is a fully developed technology. There are 29 different dielectric layers in the process. The bottom dielectric layer is the silicon substrate and the active region. The thickness of the silicon substrate can be chosen as 250 or 350 or 450 µm. The dielectric constant and resistivity of the silicon substrate are around 11.9 and 10 Ω·cm, respectively. On the silicon substrate is the dielectric contact layer of thickness 400 nm and dielectric constant 4.2. On the contact layer are the 24 silicon-oxide layers whose thicknesses are 20~790 nm and dielectric constants are 2.9~8.1. On top of the silicon-oxide layers are the 3 passivation layers whose thicknesses are 40~600 nm and dielectric constant 4.2~8.1 to protect the circuits. There are also 7 copper and one aluminum metal layers in the process. They are sandwiched in between the silicon-oxide layers and can be connected with inter metal vias. The sheet resistance of these metal layers are 19~225 mΩ/sq, depending on the layer thickness. Dummy fills need to be inserted to respect metal density rules and guarantee component integrity. (a) (b) Fig. 2. The HFSS model of the SPST switch: (a) Top view and (b) 3D view showing the radiation boundaries enclosing the chip. Fig. 2 shows the HFSS model of the SPST switch. It is known that the electromagnetic simulation of such a multi-scale structure is quite challenging and almost impossible with our computer resource of 8G RAM. To speed up computation and to reduce required memory, the number of dummies has been greatly reduced and series/arrays of vias are simplified as The HFSS or other field solvers have an inherent limitation, that is, they cannot handle the devices in their active states. Hence, the electromagnetic simulation of the isolation performance of this SPST switch seems to be infeasible because the transistors M1 and M2 are in the on- state. The limitation is overcome by using a T-junction to control the amount of 2 Forum for Electromagnetic Research Methods and Application Technologies (FERMAT) (a) currents that are flowing from the input port to the inductor and the drain of the control transistor, and by using a thin conductive layer that has a resistivity of 4~6 Ω placed underneath the source and drain metallization to represent the conducting channel during the on-state of the control transistors. Fig. 3(a) shows the current distribution when the control voltage is high (1.2V) and both M1 and M2 are turned-on. For this particular case, the current x (µA) flows to the drain of the transistor and y (µA) flows to the inductor. To mimic this transistor action in the HFSS simulation, the T-junction as illustrated in Fig. 3(b) should have the same current distribution as that in the circuit simulation. Hence, the equivalent resistance of metal interconnection toward the inductor (RB) is set (x/y) times higher than the equivalent resistance of metal interconnection toward drain of M1 or M2 (RA). y (uA) L1 M2 M1 (b) L1 Port 1 Port 2 96 pH M2 x (uA) RG2 RG1 M1 80/0.06 M2 80/0.06 (c) (a) x RB = y RA Fig. 4 shows the current density distribution across the chip at 60 GHz when the switch is on (VCTRL = 0 V). The current distribution is shown with high density within inductor (L1). When the switch is on, most of the current should flow from Port 1 to Port 2 through the inductor to achieve low insertion loss. However, as shown in the magnified area around the transistor, some current is leaked to the ground through the off-state control transistor. As shown in Fig. 4(d), the current densities at the multi-finger drain and source metalization are very high, which indicate that a strong electromagnetic coupling exists between the source and drain metallization. This strong electromagnetic coupling introduces very large parasitic capacitances that affect the insertion loss and return loss of the switch. The coupling strength is found to increase with frequency as illustrated in Fig. 5. As shown in Fig. 5(a), at 5 GHz the leakage current due to electromagnetic coupling across the drain-source metalization is smaller. As shown in Fig. 5(b), 5(c) and 5(d) respectively for 20 GHz, 40 GHz and 50 GHz, the leakage currents increase with frequency. Based on this finding, the insertion loss is expected to increase as the frequency increases due to higher leakage. RA (b) Fig. 3. Illustration of T-junctions used in isolation simulation with the HFSS: (a) current distribution when both control transistors are turned-on and (b) the equivalent resistance of T-junction to mimic the current distribution as in (a). GND GND L1 Port 1 Port 2 + + GND (d) Fig. 4. Current density distribution when the switch is on (VCTRL = 0 V) at 60 GHz: (a) Top view of the current density distribution, (b) Zoom-in 3D view, (c) Zoom-in at the rectangle area as in (b) showing the current leakages to the ground via M2, and (d) Zoom-in at the dashed rectangle area as in (c) showing the coupling strength across the transistor metallization. VCTRL = 1.2 V Port 1 M2 GND VCTRL 3 Forum for Electromagnetic Research Methods and Application Technologies (FERMAT) Port 2 L1 Port 1 (a) (b) (a) Port 2 L1 Id2 (c) (d) Fig.5. Current density distribution when the switch is on (VCTRL = 0 V) at difference frequencies: (a) at 5 GHz, (b) at 20 GHz, (c) at 40 GHz, (d) at 50 GHz. Ip1 Ip2 Figure 6 shows the current density distribution, when the switch is off (VCTRL = 1.2 V) at 135 GHz. When the control transistors are on, most of the current will be sinked to the ground and less current will flow from Port 1 to Port 2 through inductor. This is shown clearly that the current density within the inductor is now low. To achieve good isolation performance, the current that flows from Port 1 to Port 2 should be minimized when VCTRL is high. However, the current at Port 1 may reach Port 2 due to electromagnetic coupling within the control transistor. As illustrated in Fig. 6(b), the drain current of the of M1 (Id) can be coupled to the source through the drain-to-source electromagnetic coupling (Ip1), and can be coupled to the gate through the drain-to-gate electromagnetic coupling (Ip2). The currents Ip1 and Ip2 are then coupled to the drain and eventually reach Port 2 via a similar mechanism. Id1 Port 1 (b) M2 RG2 (c) Fig. 6. Current density distribution when the switch is off (VCTRL = 1.2 V) at 135 GHz: (a) Full 3D view, (b) Zoom inside the rectangle as in (a), and (c) Zoom inside the dashed rectangle as in (b) around the transistor area. III. CIRCUIT MODELING HFSS simulation is known for its accuracy in predicting the electromagnetic behavior of a passive circuit. However, designing a circuit with active devices based on HFSS simulation alone will be infeasible and very time consuming. Usually, the parasitic parameters are extracted from the HFSS 4 Forum for Electromagnetic Research Methods and Application Technologies (FERMAT) simulations and are used to enhance the circuit model to improve the accuracy of circuit simulation. R & L GND Metalizations PORT 1 A. Parasitic Extraction CPORT Fig. 7 illustrates the source-to-ground parasitic resistance (RP1) and inductance (LP1) that are contributed by: (1) Metals and vias of the source-to-ground metallization as illustrated by Fig. 7(b). (2) Metal-1 ground plane (M1 GND plane) (3) The ground metals and vias of the GSG testing pads (R & L GND metallization) as illustrated by Fig. 7(c). The main contributors of RP1 and LP1 are the control transistor’s source-to-ground metallization as illustrated by Fig. 7(b). In this case, the equivalent resistance and inductance of the ground of the GSG testing pads shown in Fig. 7(c) are much smaller since there are many vias within the GSG testing pads multilayer metallization. M2 RG1 RG1 CP1 (c) Fig. 7. (a) and (b) Illustrations of the equivalent extrinsic parasitic resistance and inductance (Rp1 & Lp1) of the source-to-actual ground (ground of testing pad), and (c) AA’ cross-section to illustrate the parasitic capacitances between the signal ports to the ground (CPORT) and also source-to-ground parasitic capacitances (CP1). The GSG testing pads of Port 1 and Port 2 are used as the direct access to the circuit; hence the modeling and characterization of them are also critical to ensure that they are operable beyond the operating frequency [17]. The signal pads are aluminum on top of copper metal 7. The top of ground pads are also aluminum followed by a series of copper metallization from metal 7 down to metal 1. Metal 1 provides the ground shield to minimize the substrate losses due to fringing fields. Fig. 8 shows the equivalent circuit of the signal pad of the GSG testing pads. For this case, Rs-pad models the resistive losses due to metal and the resistivity of the vias. The inductivity of the signal pad, which is mostly introduced by the vias, is modeled by Lpad. The fringing capacitance effect between the port metallization to the metal 1 (ground plane) is modeled by CPORT. Since the actual ground resides on top of the aluminum layer of the GSG testing pads, we have series resistance and inductance that are introduced by the multilayer ground pad, as illustrated in Fig. 7(c). RP1 M1 CP1 Bulk/Substrate PORT 2 RP1 CP1 M1 VCTRL L1 PORT 1 CP1 CP1 LP1 LP1 CPORT CPORT A LP1 M1 GND Plane LP1 RG2 A' VCTRL Lpad (a) VCTRL RG1 Rs-pad CPORT M1 CPORT R & L GND metalization RP1 & LP1 S-to-GND Parasitics S to GND Multilayer Metalizations (b) M1 GND Plane (a) Fig. 8. The equivalent circuit for a port: (a) complex model and (b) simplified model. (b) Fig. 9 shows the insertion loss simulation of the GSG testing pads. The GSG testing pads are capacitive from low frequency up to 200 GHz and after that they become more inductive. The parasitic capacitance of the GSG testing pads (CPORT) is connected in parallel with the signal pad; it acts as an additional loading capacitance to the input/output port of the switch which 5 Forum for Electromagnetic Research Methods and Application Technologies (FERMAT) TABLE I COMPONENT VALUES IN THE ENAHNCED CIRCUIT MODEL L1 M1 M2 RG1 RG2 96pH 80/0.06 80/0.06 10kΩ 10kΩ RP1 RP2 CP1 CPORT CDS CGD CGS LP1 0.1Ω 0.1Ω 5fp 50fp 6fp 2fp 2fp 20pH affects the overall matching performance. The GSG testing pads introduce about 0.5-dB insertion loss at 60 GHz and about 2-dB insertion loss at 100 GHz. The parasitic capacitances associated with the multilayer metalization of the sources, drains and gates and also the parasitic capacitances for the sources, drains, gates to bulk/substrate are extracted. As expected, (CDS, CGD, CGS) are much larger than (CSB, CDB, CGB) [19]. IV. EXPERIMENTAL VALIDATION Fig. 11 shows the die micrograph of the SPST switch that has been fabricated in the 65-nm CMOS technology. The active area is only 96 µm × 140 µm and the whole chip area including all testing pads is 340 µm × 340 µm. The measurements from 1 to 100 GHz were performed using Cascade 100-µm ground-signal-ground (GSG) probes and an Anritsu 110-GHz on-wafer S-parameter measurement system with a SussTec 12 probe station. The measurements from 110 to 170 GHz were recently carried out using Cascade Microtech I170–GSG probes with 100-µm pitch and R&S vector network analyzer, which was calibrated using line-reflect-reflect-match calibration technique with WinCal XE software and impedance standard substrate provided by Cascade Microtech. Fig. 9. HFSS simulated insertion loss of the testing pads up to 300 GHz. B. Enhanced Circuit Model The foundry-provided circuit model is added with the extracted parasitic components to improve its accuracy at frequencies higher than 50 GHz. Fig. 10 shows the enhanced circuit model of the switch. Table I lists the component values in the enhanced circuit model. L1 Port 1 CDS1 CPORT M1 CGD1 RG1 RG2 Port 2 CGD2 M2 C DS2 Fig.11. The SPST switch die micrograph. CPORT CGS2 CGS1 Fig. 12 shows the simulated, modeled, and measured results. They are in good agreement, which validates the electromagnetic analysis and circuit modeling. The isolation behavior of the SPST switch deserves a scrutiny because the transistors are in their on-state and simulated with the novel technique illustrated in Fig. 3. It is evident from Fig. 12(c) that the isolation deteriorates from the frequency of 35 GHz on up to 170 GHz. This can be understood from the pole at 35 GHz introduced by the parallel resonance due to interaction of between inductor L1 and the parasitic drain-to-source capacitances (CDS) of the transistor, the parasitic capacitances of the ports (CPORT) and the ground capacitances (CP1). From the isolation circuit simulation, we understand that the inductor L1 is not completely short-circuited to the ground from 1 to 30 GHz, even when both transistors are turned on. That is the reason why VCTRL RP2 RP1 RP1 RP2 CP1 LP1 LP1 CP1 Fig.10. The enhanced circuit model of the switch. 6 Forum for Electromagnetic Research Methods and Application Technologies (FERMAT) we obtain a finite isolation performance (about 20 dB) at low frequencies. The isolation continuously improves as the frequency is increased from 1 to 30 GHz since the impedance of the inductor continues to increase and, hence, less leakage current can flow through the inductor during the off-state. However, as the frequency further increases, the strength of the coupling within the transistor metalization becomes stronger. The electromagnetic coupling provides an alternative path for the current to flow from Port 1 to Port 2 as shown in Fig. 6. Consequently, the isolation becomes worse as the frequency is increased. (c) Fig. 12. The simulated, modeled, and measured results: (a) Return loss, (b) Insertion loss, and (c) Isolation. Clearly, the electromagnetic analysis has provided a deep physical insight into the switch behavior. However, it has an inherent limitation, that is, it cannot handle the devices in their active states for the electron transfer phenomena such as the hot carrier effect, well-proximity effect, negative-bias temperature instability, etc. Those active phenomena may lead to mobility degradation and a shift in the threshold voltage [20]. Furthermore, the analysis cannot handle the shallow trench isolation induced stress that impacts on the millimeter-wave performance of the active devices due to change in carrier mobility, saturation velocity, threshold voltage and other second-order behaviors [21]. (a) V. CONCLUSION The electromagnetic analysis and circuit modeling of the ultra-wideband SPST switch in the 65-nm CMOS technology up to millimeter-wave frequencies have been conducted and validated experimentally. A novel technique that makes it possible to conduct electromagnetic simulation of an on-state switching transistor has been proposed for the first time. It has been shown that the electromagnetic analysis provides a much deeper physical insight into the switch behavior and predicts the otherwise unforeseeable parasitic series resonance at high frequency. Also, the electromagnetic analysis improves circuit modeling and enhances circuit simulation. However, the electromagnetic analysis has an inherent limitation, that is, it cannot handle the devices for the electron transfer phenomena such as the hot carrier effect, well-proximity effect, negative-bias temperature instability, etc. Furthermore, the electromagnetic analysis cannot handle the shallow trench isolation induced stress that impacts on the millimeter-wave performance of the active devices due to change in carrier mobility, saturation velocity, threshold voltage and other second-order behaviors. (b) 7 Forum for Electromagnetic Research Methods and Application Technologies (FERMAT) REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] Alit Apriyana received the B.E. degree in Electrical and Electronic Engineering from Nanyang Technological University, Singapore, in 2003, and the M. Phil. degree in RF and telecommunication engineering from Nanyang Technological University, Singapore, in 2004. He has been with Nanyang Technological University, Singapore since 2010, working towards the Ph.D. degree. From 2004 to 2009, he worked at EPCOS Pte Ltd, Singapore as the senior engineer in RF Surface Acoustic Wave (SAW) filter design. In 2007, he was appointed to be the design leader of GSM group in EPCOS Pte Ltd. His research interests include analog/RF/millimetre-wave IC design and system architectures for wireless communications. C. H. Doan, S. Emami, A. M. Niknejad, and R. W. Brodersen, “Millimeter- wave CMOS design,” IEEE J. Solid-State Circuits, vol. 40, no. 1, pp. 144–155, 2005. Y. P. Zhang, M. Sun, L. H. Guo, “On-chip antennas for 60-GHz radios in silicon technology,” IEEE Trans. Electron Devices, vol. 52, no. 7, pp. 1664-1668, 2005. X. L. Tang, et al., “Performance improvement versus CPW loss distribution analysis of slow-wave CPW in 65 nm HR-SOI CMOS technology,” IEEE Trans. Electron Devices, vol. 59, no. 5, pp. 1279-1285, 2012. B. Leite, E. Kerherve, J. Begueret, D. Belot, “An analytical broadband model for millimeter-wave transformers in silicon technologies,” IEEE Trans. Electron Devices, vol. 59, no. 3, pp. 582-589, 2012. A. Tomkins, R. A. Aroca, T. Yamamoto, S. T. Nicolson, Y. Doi, and S. P. Voinigescu, “A zero-IF 60 GHz 65 nm CMOS transceiver with direct BPSK modulation demonstrating up to 6 Gb/s data rates over a 2 m wireless link,” IEEE J. Solid-State Circuits, vol. 44, no. 8, pp. 2085–2099, 2009. E. Laskin, et al., “Nanoscale CMOS transceiver in the 90-170-GHz range,” IEEE Trans. Microw. Theory Tech., vol. 57, no. 12, pp. 3477–3490, 2009. X. J. Li and Y. P. Zhang, “Flipping the CMOS switch,” IEEE Microw. Magazine, vol. 11, no. 1, pp. 86-96, Feb. 2010. J. He, Y. Z. Xiong, and Y. P. Zhang, “Analysis and design of 60-GHz SPDT switch in 130-nm CMOS,” IEEE Trans. Microw. Theory Tech., vol. 60, no. 10, pp. 3113-3119, 2012. Y. P. Zhang, Q. Li, W. Fan, C. H. Ang, and H. Li, "A differential CMOS T/R switch for multistandard applications," IEEE Trans. Circuits Syst. II: Express Briefs, vol. 53, no. 8, pp. 782-786, 2006. Q. Li and Y. P. Zhang, “CMOS T/R switch design: towards ultra-wideband and higher frequency,” IEEE J. Solid-State Circuits, vol. 42, no. 3, pp. 563-570, 2007. A. Poh, Y. P. Zhang, “Design and analysis of transmit/receive switch in triple-well CMOS for MIMO wireless systems,” IEEE Trans. Microw. Theory Tech., vol. 55, no. 3, pp. 458-466, 2007. Q. Li, Y. P. Zhang, K. S. Yeo, and W. M. Lim, "16.6- and 28-GHz fully integrated CMOS RF switches with improved body floating," IEEE Trans. Microw. Theory Tech., vol. 56, no. 2, pp. 339-345, 2008. M. Uzunkol, G. M. Rebeiz, A Low-Loss 50–70 GHz SPDT switch in 90 nm CMOS,” IEEE J. Solid-State Circuits, vol. 45, no. 10, pp. 2003–2007, 2010. H.-Y. Chang and C.-Y. Chan, “A low loss high isolation DC–60 GHz SPDT traveling-wave switch with a body-bias technique in 90 nm CMOS process,” IEEE Microw. Wireless Compon. Lett., vol. 20, no. 2, pp. 82–84, 2010. C. S. Kuo, H. C. Kuo, H. R. Chuang, C. Y. Chen, and T. H. Huang, “A high-isolation 60GHz CMOS transmit/receive switch,” in IEEE RFIC Symp. Dig., 5-7 Jun. 2011, Baltimore, USA. A. Tomkins, P. Garcia, and S. P. Voinigescu, “A 94GHz SPST in 65nm bulk CMOS,” in Proc. IEEE CSIC Symp., pp. 139–142, 12-15 Oct. 2008, Monteray, USA. S. Aloui, E. Kerhervé, J. B. Bégueret, R. Plana, D. Bélot, “Optimized pad design for millimeter-wave applications with a 65nm CMOS RF technology,” in Proc 39th European Microw. Conf., 29 Sep. - 1 Oct. 2009, Rome, Italy. C. K. Liang, B. Razavi, “Systematic transistor and inductor modeling for millimeter-wave design,” IEEE J. Solid-state Circuits, vol. 44, no. 2, pp. 450-457, Feb. 2009. Alit Apriyana Anak Agung, Y. P. Zhang, “Multi-fingers capacitances modeling of 65nm CMOS transistor by unit cell method,” to appear in Int. J. RF and Microw. CAE. W. Wang, V. Reddy, A. T. Krishnan, R. Vattikonda, S. Krishnan, Y. Cao, “An integrated modeling paradigm of circuit reliability for 65nm CMOS technology,” in Proc. Custom Integrated Circuits Conf., 16-19 Sep, San Jose, California. L. Yang, M. Cui, J. Ma, J. He, W. Wang, W. Wong, “Advanced Spice modeling for 65nm CMOS technology,” in Proc. Int. Conf. Solid-State Integrated-Circuit Tech, 20-23 Oct. 2008, Beijing, China. Jin He (SM’13) received the Ph.D. degree from the Nanyang Technological University, Singapore, in 2011. He has been working on analog/RF/mm-Wave in both academia and industry during 2003-2013 in China and Singapore, respectively, holding positions of Engineer, Research Associate, Senior Research Engineer, Scientist, and Project Leader. He joined in Wuhan University, China in 2013 and is now an Associate Professor of the School of Physics and Technology, Wuhan University, China. His current research interests include analog/RF/mm-Wave/THz integrated circuit design in CMOS/BiCMOS/SiGe for optical and wireless applications. Dr. He has authored and co-authored over 10 international journal and conference papers in the area of RF/mm-Wave IC design for wireless applications. He was promoted to be IEEE Senior Member and has served as a technical reviewer for the IEEE Transaction on Microwave and Theory Techniques, the IEEE Transaction on Electron Devices, the IEEE Microwave and Wireless Components Letters, the Electronics and Telecommunications Research Institute (ETRL) Journal, and the Journal of Electronic Science and Technology. Yueping Zhang (M’03-SM’07-F’10) received the B.E. and M.E. degrees from Taiyuan Polytechnic Institute and Shanxi Mining Institute of Taiyuan University of Technology, Shanxi, China, in 1982 and 1987, respectively and the Ph.D. degree from the Chinese University of Hong Kong, Hong Kong, in 1995, all in electronic engineering. From 1982 to 1984, he worked at Shanxi Electronic Industry Bureau, from 1990 to 1992, the University of Liverpool, U. K., and from 1996 to 1997, City University of Hong Kong. From 1987 to 1990, he taught at Shanxi Mining Institute and from 1997 to 1998, the University of Hong Kong. He was promoted to Professor at Taiyuan University of Technology in 1996. He is now Professor of Electronic Engineering with the School of Electrical and Electronic Engineering at Nanyang Technological University, Singapore. He has researched extensively in the field of radio science and technology. He has published numerous papers and taken various US patents. Dr. Zhang was a Guest Editor of the International Journal of RF and Microwave Computer-Aided Engineering, an Associate Editor of the International Journal of Microwave Science and Technology, and an Associate Editor of the International Journal of Electromagnetic Waves and Applications. He serves as an Editor of ETRI Journal and an Associate Editor of the IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION. He also serves on the Editorial Boards of a large number of Journals including the IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES and IEEE MICROWAVE AND WIRELESS COMPONENTS LETTERS. He was the recipient of the Sino-British Technical Collaboration Award (1990) for his contribution to the advancement of subsurface radio science and technology, the Best Paper Award of the 2nd IEEE/IET International Symposium on Communication Systems, Networks and Digital Signal Processing, Bournemouth, U.K. (July 18–20, 2000), and the Best Paper Prize of the 3rd IEEE International Workshop on Antenna Technology, Cambridge, U.K. (March 21–23, 2007). He was also the recipient of a William Mong Visiting Fellowship of the University of Hong Kong (2005) and the S. A. Schelkunoff Transactions Prize Paper Award of the IEEE Antennas and Propagation Society (2012). He was the Chair, leading the Singapore Chapter to win the Best Chapter Award of the IEEE Antennas and Propagation Society (2013). 8 Forum for Electromagnetic Research Methods and Application Technologies (FERMAT) Manuscript received October 12, 2012. 9