PAPawan Agarwal

Pawan Agarwal

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Pawan Agarwal has a total of 28 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Improving Linearity in CMOS Phase Interpolators

    by Amit Kumar Mishra, Yifei Li, Pawan Agarwal, Sudip Shekhar

    We compare the prior art in phase interpolators (PIs), classifying them as current-mode, voltage-mode, and integrating-mode PI. Next, we present an integrating-mode PI where the voltage slopes with high phase linearity are generated through the integration of phase-shifted weighted current sources. The constant and variable voltage slopes are generated by current sources/sinks created using stacked devices in a 0.75-V 5-nm finFET technology. This PI technique supports the high-speed and low-power operation and achieves dual-edge interpolation with improved duty-cycle distortion characteristics. The PI generates an output clock with 9 bits of resolution and a small peak-to-peak integral nonlinearity (INL $_{\text{pp}}$ ) and peak-to-peak differential nonlinearity (DNL $_{\text{pp}}$ ) of 2.4and 1.4, respectively, at 13.3 GHz with just quadrature clock inputs. The PI has a 71-fs $_{\text{rms}}$ random jitter (integrated from 3 MHz to 3 GHz) and occupies an active area of 0.006 mm 2 while consuming 6-mW power at 14 GHz. An integrated rotation spur of $-$ 42.6 dBc for 256-ppm modulation at 13.3-GHz operating frequency is achieved for 1-GHz update rate for the dynamic linearity measurements.


  • A New Boosted Active-Capacitor With Negative-G m for Wide Tuning Range VCOs

    by Pawan Agarwal, Mohammad Chahardori, Deukhyoun Heo

    A new boosted active-capacitor (BAC) architecture using LC-tuned active-impedance-conversion is presented. In the chosen frequency range, the proposed BAC acts as a tunable capacitance multiplier and negative-transconductance, but where the positive-transconductance is not multiplied. The BAC is used as a low-loss tunable capacitor bank element for designing a wide tuning range voltage-controlled oscillator (VCO). A current-reuse technique is used to lower power consumption and suppress potential parasitic oscillation modes. A zero-capacitance-variation biasing is proposed for a BAC-VCO robust against process, voltage, and temperature variations. The BAC designed in a 65 nm bulk CMOS process achieves a capacitance tuning ratio of 7.8 with a quality factor of higher than 20 across VCO's tuning range. The BAC-VCO prototype has a tuning range of 8.3-14.3 GHz using less than 5 mW core power and achieves FoMT of up to -202.4 dBc/Hz.


  • A Ka-Band Dual-Band Digitally Controlled Oscillator With −195.1-dBc/Hz FoM${_T}$ Based on a Compact High-$Q$ Dual-Path Phase-Switched Inductor

    by Joe Baylon, Pawan Agarwal, Luke Renaud, Sheikh Nijam Ali, Deukhyoun Heo

    In this paper, a compact, low-loss, dual-path phase-switched inductor suitable for millimeter-wave (mmWave) transceiver subblocks is proposed. The signal path through the structure is determined by the relative polarity of two differential excitation signals applied to the four-port structure. Because inductance is varied by altering signal path through the inductor, signal-path MOSFET switches are avoided. The inductor, therefore, demonstrates high, mode-invariant quality factor, facilitating the design of high-performance dual-band mmWave transceiver subblocks. Based on the proposed inductor, a dual-band Ka-band oscillator is designed in a 65-nm CMOS technology which covers two frequency bands from 14.8 to 18.7 GHz and from 20.8 to 26.6 GHz. The proposed inductor facilitates a widely tunable, dual-band frequency range while maintaining the state-of-the-art phase noise, power consumption, and figure of merit (FoM). The oscillator achieves measured fractional tuning ranges of 23.3% and 24.5% and phase noise of -115.1 dBc/Hz at 2-MHz offset while consuming only 4.8-mW dc power. A peak tuning-range FoM of -195.1 and -194.9 dBc/Hz is demonstrated across the low- and high-frequency bands.


  • Transformer-Based Predistortion Linearizer for High Linearity and High Modulation Efficiency in mm-Wave 5G CMOS Power Amplifiers

    by Sheikh Nijam Ali, Pawan Agarwal, Srinivasan Gopal, Deukhyoun Heo

    This paper presents a new predistortion linearization technique for high linearity and high modulation efficiency in millimeter-wave (mm-wave) CMOS power amplifiers (PA) for fifth-generation (5G) mobile communications. Our proposed linearizer adopts a transformer-based (i.e., inductive) self-compensated predistortion network at the input of the PA whose amplitude-modulation to phase-modulation (AM-PM) response is opposite compared with the AM-PM response of a CMOS PA, resulting in an AM-PM cancellation effect. This proposed inductive linearization method mitigates the large gain reduction problem in traditional capacitor-based linearization approaches while consuming no extra dc power or without introducing additional control circuitry. As a result, a significant improvement in power-efficiency and linearity is achieved with high-order complex modulation signals. To validate the proposed linearization method, a PA prototype in 65-nm CMOS technology was fabricated and tested, and it exhibited <1° of | AM-PM| distortion at Pₒ,1, dB over 4 GHz of bandwidth (27-31 GHz). At 28 GHz, the measured saturated Pₒ and peak power-added-efficiency (PAE) was 15.6 dBm and 41%, respectively, while achieving a 6-dB Pₒ back-off PAE of 25%. To assess PA's large-signal performance for 5G communications, the prototype was measured with the 64-quadratic-amplitude modulation (QAM) signal at 2-Gb/s data rates at 28 and 30 GHz, and the PA achieves modulated-PAE of 18.2%/17.6% and average-Pₒ of 9.8 dBm/10 dBm, respectively, while maintaining <-30 dBc of adjacent-channel-power-ratio and <-25.5 dB of error-vector-magnitude. The achieved modulated-PAE at 28 and 30 GHz shows more than 2x improvement in comparison with the recently reported 28-GHz linear CMOS PAs. Also, the PA occupies a compact active-area of 0.24 mm².


  • High-Performance and Small-Form Factor Near-Field Inductive Coupling for 3-D NoC

    by Srinivasan Gopal, Sourav Das, Pawan Agarwal, Sheikh Nijam Ali, Deukhyoun Heo, Partha Pratim Pande

    Wireless interconnects using near-field inductive coupling (NFIC) enables contactless vertical communications necessary for the design of energy efficient and robust 3-D manycore systems. However, the achievable performance, energy efficiency, bandwidth, and associated area overhead of NFICs are intertwined imposing significant design challenges and tradeoffs to explore the optimum link configuration. To address these challenges, in this paper, we propose a holistic design approach for exploring energy-efficient NFICs and target to exploit the benefits of the NFICs in the context of efficient and reliable network-on-chip (NoC) design. The proposed design framework employs statistical link analysis to select optimum NFIC-link configuration and is significantly more efficient in terms of energy efficiency and area overhead compared to the state-of-the-art counterparts. We demonstrate that 3-D NoCs incorporating NFIC-enabled links outperform through-silicon-via (TSV) counterparts. In addition, the overall reliability of TSV- and NFIC-enabled hybrid 3-D NoC is significantly better than only TSV-based NoCs in order to counteract the electromigration and workload-induced stress challenges.


  • A 25–35 GHz Neutralized Continuous Class-F CMOS Power Amplifier for 5G Mobile Communications Achieving 26% Modulation PAE at 1.5 Gb/s and 46.4% Peak PAE

    by Sheikh Nijam Ali, Pawan Agarwal, Srinivasan Gopal, Shahriar Mirabbasi, Deukhyoun Heo

    This paper presents a high-efficiency neutralized continuous class-F (CCF) CMOS power amplifier (PA) design technique for millimeter-wave (mmW) 5G mobile communications. A parasitic-aware tuned-load with a high-order harmonic-resonance network is proposed to shape the current and voltage waveforms for the CCF PA. At mmW frequencies, the gate-drain capacitance (Cgd) creates adverse capacitive loading on harmonic-tuned output load networks. As a result, a CCF PA suffers from compromised power efficiency. To address this, we integrate a transformer with a tunable coupling coefficient between gate-drain of the power-device. This proposed technique allows accurate neutralization of Cgd, reducing detrimental loading effect on harmonic-tuned load while enhancing power efficiency and stability. We fabricated a CCF PA in 65-nm CMOS technology achieving >40% power-added efficiency (PAE) over 33.3% fractional bandwidth (25-35 GHz) and 46.4% peak PAE at 29 GHz. The measured peak saturated output power (Po,sat) is 14.8 dBm at 30 GHz. The PA is tested with 64-quadrature-amplitude-modulated signal at a data rate of 1.5 Gb/s. Under this test setup, the PA achieves modulated PAE of 26%/24%/21.4% and average output power (Po,avg) of 9.2/8.8/8.6 dBm at 28, 32, and 34 GHz, respectively, while maintaining better than -25 dB of error vector magnitude and -27 dBc of adjacent channel leakage ratio. To the authors' knowledge, this design presents one of the highest reported PAEs among mmW CMOS PAs.


  • A Spatial Multi-Bit Sub-1-V Time-Domain Matrix Multiplier Interface for Approximate Computing in 65-nm CMOS

    by Srinivasan Gopal, Pawan Agarwal, Joe Baylon, Luke Renaud, Sheikh Nijam Ali, Partha Pratim Pande, Deukhyoun Heo

    Large scale parallel implementation of matrix multiply and accumulate (MAC) core poses significant energy and area constraints in analog voltage domain under reduced supply voltage. A spatial multi-bit sub-1V time-domain matrix multiplier interface is presented using multi-bit back-gate-driven delay elements as a scalable alternative for various approximate computing applications. A single-chip solution is demonstrated for two application modes: a high throughput digitally-driven mode for acceleration and a low energy analog front-end mode for sensing. In accelerate-mode, the system achieves an aggregate throughput of 21.6 GMAC/s with 9 TOPS/W energy efficiency. In sense-mode, the system exhibits an energy efficiency of 55.3 TOPS/W for classification purpose. The proposed architecture utilizes 16-parallel 6-bit input vectors to perform matrix MAC computations using time-domain signal processing with 3-bit resistive weights at a sub-1V supply of 0.7 V. An integrated speculative time-to-digital converter (TDC) is employed for 6-bit time-domain quantization with an on-chip mismatch calibration scheme. The prototype is fabricated in 65 nm CMOS technology and occupies an active area of 0.04 mm2. The system performs image recognition of handwritten digits using a machine learning scheme and demonstrates an average classification accuracy of 84.3% on the MNIST dataset. The resultant energy per MAC computation in the proposed spatial architecture is about 15× lower than a digital CMOS combinational logic based parallel-tree MAC.


  • A 40% PAE Frequency-Reconfigurable CMOS Power Amplifier With Tunable Gate–Drain Neutralization for 28-GHz 5G Radios

    by Sheikh Nijam Ali, Pawan Agarwal, Luke Renaud, Reza Molavi, Shahriar Mirabbasi, Partha Pratim Pande, Deukhyoun Heo

    In this paper, a high-efficiency frequency-reconfigurable CMOS power amplifier (PA) design technique is presented at 24 and 28 GHz using integrated tunable neutralization and matching networks. To cope with the adverse effects of gate-drain capacitance (C gd ) in millimeter-wave (mm-wave) CMOS PAs in deep-submicrometer technologies, we propose a reconfigurable coupling-coefficient-based transformer. This technique provides optimum neutralization of C gd in a common-source configuration, while enabling tunable neutralization in a frequency reconfigurable PA. Furthermore, to reconfigure the PA's input and output matching networks, a low-loss frequency-reconfigurable matching topology using a switched substrate-shield inductor is proposed. The adopted matching network addresses the high loss in a conventional frequency reconfigurable matching approach, while facilitating high efficiency in the proposed PA design. Using the proposed techniques, a class-AB PA is fabricated in a 65-nm CMOS technology. This prototype achieves measured saturated power-added efficiency (PAE sat ) of 42.6%, saturated output power (P o,sat ) of 14.7 dBm at 24 GHz, and 40.1% PAE sat , and 14.4 dBm P o,sat at 28 GHz while occupying an active area of only 0.11 mm 2 . At 24 and 28 GHz, the PA is tested under 16 and 64-quadrature amplitude modulation (QAM) signals with 250 MHz of channel BW. The PA achieves modulated PAE of 15.2%/14.1%, error vector magnitude of -26.4/-26.6 dB, and adjacent channel leakage ratio of -30/-32 dBc at an average output power of 7.1/7 dBm for a 64-QAM signal with 250-MHz BW at 24 and 28 GHz, respectively. To the best of the author's knowledge, the design presents one of the highest PAEs among mm-wave CMOS PAs reported to date, while it also supports multiband operation.


  • Open Access

    Zero-Power Feed-Forward Spur Cancelation for Supply-Regulated CMOS Ring PLLs

    by Pawan Agarwal, Jong-Hoon Kim, Partha Pratim Pande, Deukhyoun Heo

    A new reference-spur cancelation technique is presented for supply-regulated ring-oscillator-based integer-N phaselocked loops (PLLs). A passive RC filter is used to implement a feed-forward (FF) spur-coupling path to perform spur cancelation at the PLL control signal. The proposed technique achieves a simulated spur cancelation of about 22 dB at the first spur harmonic. The simulated postcancelation spur value is -79 dBc for an oscillator gain of 0.1 GHz/V and -46 dBc for an oscillator gain of 6 GHz/V. Spur cancelation is also robust against large process, voltage, and temperature variations in the gain and bandwidth of the FF path. A 1-GHz integerN PLL prototype in a 65-nm CMOS process has a measured cancelation of 19.5 and 13 dB at the first and the second spur harmonic, respectively, with 320 μW of total power consumption. The PLL prototype has an oscillator gain of 1.5 GHz/V, which results in a postcancelation spur of -53 dBc. The proposed zero-power technique is suitable for low-power PLLs as it achieves a large spur cancelation without requiring any additional power consumption or calibration.


  • Open Access

    Switched Substrate-Shield-Based Low-Loss CMOS Inductors for Wide Tuning Range VCOs

    by Pawan Agarwal, Suman Prasad Sah, Reza Molavi, Shahriar Mirabbasi, Partha Pratim Pande, Seung Eel Oh, Jong-Hoon Kim, Deukhyoun Heo

    A switched substrate-shield inductor (SSI) topology in bulk CMOS is proposed which minimizes parasitic capacitance and substrate losses, while tuned magnetically induced currents facilitate inductor tunability. The high frequency behavior of the induced current is analyzed, resulting in intuitive insights and design guidelines for a high-performance SSI. An SSI prototype in 65-nm bulk CMOS achieves 34% inductance tunability with a quality factor of >10.3. A voltage-controlled oscillator (VCO) using SSI achieves 40.3% tuning range, from 21 to 31.6 GHz, and a phase noise of -119.1 ± 3.7 dBc/Hz at 10-MHz offset frequencies. The VCO core consumes 4.3 ± 0.2 mW from a 1-V supply.