Linxiao Shen
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
Research Subjects
Research Interests
Linxiao Shen has a total of 110 co-authors. Here are the top co-authors by number of shared publications:
- Xiyuan Tang
Institute for Artificial Intelligence and the School of Integrated Circuits, Peking University, Beijing, China
14 shared publications - Le Ye
Peking University
10 shared publications - Nan Sun
Tsinghua University
10 shared publications - Jihang Gao
School of Integrated Circuits, Peking University, Beijing, China
9 shared publications Ru Huang
School of Integrated Circuits, Peking University, Beijing, China
9 shared publications- Wenda Zhao
Department of Electrical and Computer Engineering, The University of Texas at Austin, Austin, TX, USA
9 shared publications
Latest Publications
- 22 October 2026
- Published by AIP Publishing in Journal of Applied Physics
- Vol. 140 (16)
Field-free switching of perpendicular magnetization using orthogonal spin currents for all 16 Boolean logic functions
All-electrical magnetization control by spin currents is a critical step toward low-power, nonvolatile, and programmable spin logic. Here, using magnetization dynamics simulations, we demonstrate a field-free switching scheme driven by two orthogonal spin-current pulses with a temporal offset. Deterministic perpendicular magnetization switching is achieved when both field-like and damping-like spin–orbit torques are present. The first pulse tilts the magnetization into the plane, while the second breaks torque symmetry to select the final state. The switching polarity is governed solely by the relative orientations of the two pulses and is independent of the initial magnetization state. Employing orthogonal currents as input variables, all 16 Boolean logic functions are implemented within a single device. This field-free switching mechanism and Boolean logic functionality are fully compatible with industry-standard magnetic tunnel junctions, paving a practical route toward all-electrical spin-based in-memory computing.
- 22 October 2026
- Published by AIP Publishing in Journal of Applied Physics
- Vol. 140 (16)
Field-free switching of perpendicular magnetization using orthogonal spin currents for all 16 Boolean logic functions
All-electrical magnetization control by spin currents is a critical step toward low-power, nonvolatile, and programmable spin logic. Here, using magnetization dynamics simulations, we demonstrate a field-free switching scheme driven by two orthogonal spin-current pulses with a temporal offset. Deterministic perpendicular magnetization switching is achieved when both field-like and damping-like spin–orbit torques are present. The first pulse tilts the magnetization into the plane, while the second breaks torque symmetry to select the final state. The switching polarity is governed solely by the relative orientations of the two pulses and is independent of the initial magnetization state. Employing orthogonal currents as input variables, all 16 Boolean logic functions are implemented within a single device. This field-free switching mechanism and Boolean logic functionality are fully compatible with industry-standard magnetic tunnel junctions, paving a practical route toward all-electrical spin-based in-memory computing.
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Microelectromechanical Systems
- Vol. 35 (5) , 771-779
A Broadband and High-Sensitivity Ultrasonic Receiver for Miniaturized Photoacoustic Systems
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Microelectromechanical Systems
- Vol. 35 (5) , 771-779
A Broadband and High-Sensitivity Ultrasonic Receiver for Miniaturized Photoacoustic Systems
- 21 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. PP (00189200) , 1-11
An Energy-Efficient Pipelined-SAR ADC With Cascode Capacitively Degenerated Dynamic Amplifier and MSB Pre-Conversion Technique
This article presents an energy-efficient fully dynamic pipelined successive approximation register (SAR) analog-to-digital converter (ADC), which utilizes a cascode capacitively degenerated dynamic amplifier (CD-DA) as the residue amplifier. The proposed amplifier achieves inherent linearity together with a boosted gain, while an output swing enhancement technique further extends its swing range. To guarantee stable linearity across wide temperature variations, a PTAT bias current is employed. In addition, the most significant bit (MSB) pre-conversion technique is introduced, which elegantly integrates the output swing enhancement method with the second-stage SAR ADC, enabling simultaneous swing extension and faster residue conversion without additional hardware. The prototype ADC is fabricated in a 22-nm CMOS process and achieves 73.9-dB signal-to-noise-and-distortion ratio (SNDR) and 89.4-dB spurious free dynamic range (SFDR) at Nyquist input while consuming only 0.36 mW. This corresponds to a Walden figure of merit (FoM) of 1.78 fJ/conversion-step and Schreier FoM of 182.3 dB. The measurement shows that the SNDR variation is below 2 dB at $- 40\sim 80~^{\circ } $ C and over ±10% supply variation.
- 21 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. PP (00189200) , 1-11
An Energy-Efficient Pipelined-SAR ADC With Cascode Capacitively Degenerated Dynamic Amplifier and MSB Pre-Conversion Technique
This article presents an energy-efficient fully dynamic pipelined successive approximation register (SAR) analog-to-digital converter (ADC), which utilizes a cascode capacitively degenerated dynamic amplifier (CD-DA) as the residue amplifier. The proposed amplifier achieves inherent linearity together with a boosted gain, while an output swing enhancement technique further extends its swing range. To guarantee stable linearity across wide temperature variations, a PTAT bias current is employed. In addition, the most significant bit (MSB) pre-conversion technique is introduced, which elegantly integrates the output swing enhancement method with the second-stage SAR ADC, enabling simultaneous swing extension and faster residue conversion without additional hardware. The prototype ADC is fabricated in a 22-nm CMOS process and achieves 73.9-dB signal-to-noise-and-distortion ratio (SNDR) and 89.4-dB spurious free dynamic range (SFDR) at Nyquist input while consuming only 0.36 mW. This corresponds to a Walden figure of merit (FoM) of 1.78 fJ/conversion-step and Schreier FoM of 182.3 dB. The measurement shows that the SNDR variation is below 2 dB at $- 40\sim 80~^{\circ } $ C and over ±10% supply variation.
- 10 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 60 (12) , 4490-4502
A Calibration-Free Pipelined-SAR ADC With Cross-Stage Gain-Mismatch Error Shaping and Inherent Noise Shaping
This article presents a calibration-free pipelined-successive-approximation-register (SAR) analog-to-digital converter (ADC) based on the proposed cross-stage gain-mismatch-error shaping (CS-GMES) mechanism. The CS-GMES is realized by including the entire 2nd stage into MES operation to unify the gain error and the 2nd-stage mismatch error. A feedback capacitor provides cross-stage connection and mismatch reference, and prevents saturation issues. Besides, low-cost noise shaping (NS) is achieved by the inherent residue preservation of the 2nd stage. A negative-R-assisted residue integrator instead of an amplifier is proposed to realize high-efficiency residue summation and 1st-order error-feedback NS. With the proposed techniques, the prototype fabricated in a 55-nm CMOS process achieves a 93.3-dB signal-to-noise-and-distortion-ratio (SNDR) within a bandwidth of 156.25 kHz without any off-chip calibration. It consumes 306.88 uW from a 1.2-V supply and exhibits a 180.4-dB Schreier figure of merits (FoM). Among a gain error range of −33% to +50%, the SNDR drops less than 3 dB.
- 10 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 60 (12) , 4490-4502
A Calibration-Free Pipelined-SAR ADC With Cross-Stage Gain-Mismatch Error Shaping and Inherent Noise Shaping
This article presents a calibration-free pipelined-successive-approximation-register (SAR) analog-to-digital converter (ADC) based on the proposed cross-stage gain-mismatch-error shaping (CS-GMES) mechanism. The CS-GMES is realized by including the entire 2nd stage into MES operation to unify the gain error and the 2nd-stage mismatch error. A feedback capacitor provides cross-stage connection and mismatch reference, and prevents saturation issues. Besides, low-cost noise shaping (NS) is achieved by the inherent residue preservation of the 2nd stage. A negative-R-assisted residue integrator instead of an amplifier is proposed to realize high-efficiency residue summation and 1st-order error-feedback NS. With the proposed techniques, the prototype fabricated in a 55-nm CMOS process achieves a 93.3-dB signal-to-noise-and-distortion-ratio (SNDR) within a bandwidth of 156.25 kHz without any off-chip calibration. It consumes 306.88 uW from a 1.2-V supply and exhibits a 180.4-dB Schreier figure of merits (FoM). Among a gain error range of −33% to +50%, the SNDR drops less than 3 dB.
- 7 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 73 (4) , 2345-2356
An Area-Efficient Noise-Shaping SAR ADC Utilizing Dynamic Common-Gate Amplifier With Charge-Boosted Amplification and Dynamic-Bulk-Switching
This paper presents an area-efficient $2{^{\text {nd}}}$ -order noise-shaping (NS) SAR ADC leveraging a dynamic common-gate amplifier. By reconfiguring a simple switch into a dynamic common-gate (DCG) amplifier through adjusting the pulse height applied to a transistor’s gate, voltage gain is achieved prior to the passive loop filter, thereby enhancing noise transfer function (NTF) while maintaining area- and power-efficient loop filtering. To implement a $2{^{\text {nd}}}$ -order loop filter, two key techniques are proposed: First, a charge-boosted amplification scheme doubles the charges transferred to the residue capacitors, enabling realization of $2{^{\text {nd}}}$ -order noise shaping; Second, a dynamic bulk-switching mechanism triggers a second charge transfer by switching the amplifier’s bulk, eliminating the need for additional amplification stages. Furthermore, to ensure robust noise shaping across process-voltage-temperature (PVT) variations, a PVT-tracking pulse generator is introduced to maintain stable amplifier oper ation. With these techniques, the prototype ADC achieves a 76-dB signal-to-noise-and-distortion ratio (SNDR) with a compact active area of 0.0045 mm2. Operating at 5MS/s sample rate with a 312.5-kHz bandwidth, it consumes 31.5uW, yielding a Schreier Figure-of-Merit (FoM) of 176 dB and a Walden FoM of 9.8fJ/conv.step.
- 7 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 73 (4) , 2345-2356
An Area-Efficient Noise-Shaping SAR ADC Utilizing Dynamic Common-Gate Amplifier With Charge-Boosted Amplification and Dynamic-Bulk-Switching
This paper presents an area-efficient $2{^{\text {nd}}}$ -order noise-shaping (NS) SAR ADC leveraging a dynamic common-gate amplifier. By reconfiguring a simple switch into a dynamic common-gate (DCG) amplifier through adjusting the pulse height applied to a transistor’s gate, voltage gain is achieved prior to the passive loop filter, thereby enhancing noise transfer function (NTF) while maintaining area- and power-efficient loop filtering. To implement a $2{^{\text {nd}}}$ -order loop filter, two key techniques are proposed: First, a charge-boosted amplification scheme doubles the charges transferred to the residue capacitors, enabling realization of $2{^{\text {nd}}}$ -order noise shaping; Second, a dynamic bulk-switching mechanism triggers a second charge transfer by switching the amplifier’s bulk, eliminating the need for additional amplification stages. Furthermore, to ensure robust noise shaping across process-voltage-temperature (PVT) variations, a PVT-tracking pulse generator is introduced to maintain stable amplifier oper ation. With these techniques, the prototype ADC achieves a 76-dB signal-to-noise-and-distortion ratio (SNDR) with a compact active area of 0.0045 mm2. Operating at 5MS/s sample rate with a 312.5-kHz bandwidth, it consumes 31.5uW, yielding a Schreier Figure-of-Merit (FoM) of 176 dB and a Walden FoM of 9.8fJ/conv.step.
- 10 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 61 (3) , 1153-1163
An 8b 1-GS/s SAR ADC With Comparison Time Quantizer, Metastability Reduction, and On-Chip Self-Tuning Delay Achieving 47.2-dB SNDR at Nyquist Input
This article presents a 1-GS/s 8-bit SAR ADC incorporating a comparison time quantizer ( $t_{\text {comp}}$ quantizer) and an on-chip self-tuning delay to mitigate metastability while enhancing resolution and speed. The ( $t_{\text {comp}}$ quantizer) extracts additional bit information by quantizing the comparator delay, improving resolution without sacrificing speed. The metastability from ( $t_{\text {comp}}$ quantizer) is also analyzed and effectively suppressed. The self-tuning delay dynamically adjusts time thresholds based on hit ratio detection, ensuring robust operation across process, voltage, and temperature (PVT) variations. Fabricated in 22-nm CMOS, the prototype occupies only 0.0032 mm2 and achieves 47.2-dB signal-to-noise-and-distortion ratio (SNDR) at 1 GS/s with a Nyquist input, consuming 4.15 mW. Measurements confirm the effectiveness of metastability suppression, with no metastability observed in $2\times 10^{11}$ data points.
- 10 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 61 (3) , 1153-1163
An 8b 1-GS/s SAR ADC With Comparison Time Quantizer, Metastability Reduction, and On-Chip Self-Tuning Delay Achieving 47.2-dB SNDR at Nyquist Input
This article presents a 1-GS/s 8-bit SAR ADC incorporating a comparison time quantizer ( $t_{\text {comp}}$ quantizer) and an on-chip self-tuning delay to mitigate metastability while enhancing resolution and speed. The ( $t_{\text {comp}}$ quantizer) extracts additional bit information by quantizing the comparator delay, improving resolution without sacrificing speed. The metastability from ( $t_{\text {comp}}$ quantizer) is also analyzed and effectively suppressed. The self-tuning delay dynamically adjusts time thresholds based on hit ratio detection, ensuring robust operation across process, voltage, and temperature (PVT) variations. Fabricated in 22-nm CMOS, the prototype occupies only 0.0032 mm2 and achieves 47.2-dB signal-to-noise-and-distortion ratio (SNDR) at 1 GS/s with a Nyquist input, consuming 4.15 mW. Measurements confirm the effectiveness of metastability suppression, with no metastability observed in $2\times 10^{11}$ data points.
- 6 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 61 (2) , 603-614
A Highly Power-Efficient Input-Boosted First Stage for Capacitively Coupled Chopper Instrumentation Amplifiers
This article presents a highly power-efficient capacitively coupled chopper instrumentation amplifier (CCIA) featuring an input-boosted first stage and leakage compensation circuits. To address the limitations of low-noise front-end amplifier design imposed by the physical constraints of $g_{\mathrm {m}}/I_{\mathrm {d}}$ , we proposed an input-boosted, inverter-based operational transconductance amplifier (OTA) scheme. This approach introduces a new perspective by doubling the signal-dependent voltage, effectively halving the input-referred noise (IRN) with the same supply current and voltage. Two prototypes were implemented in 180 and 22 nm CMOS processes, consuming 24 and 1.6 nW, respectively. To mitigate leakage current issues in the ultralow-power 22 nm prototype, compensation circuits are incorporated to address reverse-biased diode leakage and bulk leakage current. The 180 nm prototype achieves an IRN of $5.09~{\mathrm {\mu V_{\rm rms}}}$ within a 1.6 kHz bandwidth (BW), yielding a power efficiency factor (PEF) of 0.58. The 22 nm prototype achieves an IRN of $8.05~{\mathrm {\mu V_{\rm rms}}}$ within a 230 Hz BW, resulting in a PEF of 0.69. Measurements under varying temperatures validate the effectiveness of the leakage compensation circuits. The proposed input-boosted CCIA achieves the highest PEF among continuous-time amplifiers (CTAs), offering a general solution for power-efficient amplifier design.
- 6 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 61 (2) , 603-614
A Highly Power-Efficient Input-Boosted First Stage for Capacitively Coupled Chopper Instrumentation Amplifiers
This article presents a highly power-efficient capacitively coupled chopper instrumentation amplifier (CCIA) featuring an input-boosted first stage and leakage compensation circuits. To address the limitations of low-noise front-end amplifier design imposed by the physical constraints of $g_{\mathrm {m}}/I_{\mathrm {d}}$ , we proposed an input-boosted, inverter-based operational transconductance amplifier (OTA) scheme. This approach introduces a new perspective by doubling the signal-dependent voltage, effectively halving the input-referred noise (IRN) with the same supply current and voltage. Two prototypes were implemented in 180 and 22 nm CMOS processes, consuming 24 and 1.6 nW, respectively. To mitigate leakage current issues in the ultralow-power 22 nm prototype, compensation circuits are incorporated to address reverse-biased diode leakage and bulk leakage current. The 180 nm prototype achieves an IRN of $5.09~{\mathrm {\mu V_{\rm rms}}}$ within a 1.6 kHz bandwidth (BW), yielding a power efficiency factor (PEF) of 0.58. The 22 nm prototype achieves an IRN of $8.05~{\mathrm {\mu V_{\rm rms}}}$ within a 230 Hz BW, resulting in a PEF of 0.69. Measurements under varying temperatures validate the effectiveness of the leakage compensation circuits. The proposed input-boosted CCIA achieves the highest PEF among continuous-time amplifiers (CTAs), offering a general solution for power-efficient amplifier design.
- Top Cited Papers
- 19 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 60 (8) , 3020-3032
A 22-nm Delta-Sigma Computing-In-Memory SRAM Macro With Near-Zero-Mean Outputs and LSB-First ADCs for Edge AI Processing
This work proposes a static random access memory (SRAM)-based delta–sigma computing-in-memory (ΔΣCIM) processor for edge intelligence applications in 22-nm CMOS technology. This ΔΣCIM processor effectively mitigates redundant operations associated with unchanged input features and adaptively adjusts computing-in-memory (CIM) outputs to achieve a near-zero-mean (NZM) distribution, thereby benefiting the energy-intensive analog-to-digital converters (ADCs). The proposed ΔΣCIM macro involves three main innovations at the architecture and circuit levels: first, it replaces absolute inputs with deltas computed from consecutive input values, resulting in a notable 35.8% reduction in energy consumption for the CIM array; second, the delta matrix–vector multiplications (MVMs) implemented by ΔΣCIM array can self-adaptively shift the analog CIM outputs to be NZM distributed, facilitating streamlined centralized data processing; and third, the introduction of a least significant bit first (LSB-first) ADC, whose number of conversions adjust dynamically based on instantaneous input values, allows for efficient analog output conversion of highly concentrated NZM analog data without compromising full-scale range or precision. Compared with state-of-the-art (SOTA) analog SRAM-based CIM macros, this work achieves superior area–energy efficiency product by 1.25× with up to 21.38 TOPS/W and 1.44 TOPS/mm2 at full precision of 8b input, 8b weight, and 24b output, through bit-serial inputs and shifter-and-adder.
- 19 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 60 (8) , 3020-3032
Top Cited PapersA 22-nm Delta-Sigma Computing-In-Memory SRAM Macro With Near-Zero-Mean Outputs and LSB-First ADCs for Edge AI Processing
This work proposes a static random access memory (SRAM)-based delta–sigma computing-in-memory (ΔΣCIM) processor for edge intelligence applications in 22-nm CMOS technology. This ΔΣCIM processor effectively mitigates redundant operations associated with unchanged input features and adaptively adjusts computing-in-memory (CIM) outputs to achieve a near-zero-mean (NZM) distribution, thereby benefiting the energy-intensive analog-to-digital converters (ADCs). The proposed ΔΣCIM macro involves three main innovations at the architecture and circuit levels: first, it replaces absolute inputs with deltas computed from consecutive input values, resulting in a notable 35.8% reduction in energy consumption for the CIM array; second, the delta matrix–vector multiplications (MVMs) implemented by ΔΣCIM array can self-adaptively shift the analog CIM outputs to be NZM distributed, facilitating streamlined centralized data processing; and third, the introduction of a least significant bit first (LSB-first) ADC, whose number of conversions adjust dynamically based on instantaneous input values, allows for efficient analog output conversion of highly concentrated NZM analog data without compromising full-scale range or precision. Compared with state-of-the-art (SOTA) analog SRAM-based CIM macros, this work achieves superior area–energy efficiency product by 1.25× with up to 21.38 TOPS/W and 1.44 TOPS/mm2 at full precision of 8b input, 8b weight, and 24b output, through bit-serial inputs and shifter-and-adder.
- 25 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 72 (2) , 384-388
A Flying-Capacitor-Based Reset Scheme for Low Power Dynamic Comparator
A dynamic comparator reset scheme based on the flying capacitor is proposed. The reset power used to charge the integration nodes of the comparator is nearly halved by exploiting the existent integration capacitors with only a single D-flip-flop and few switches. The proposed reset scheme improves the energy efficiency while the small CLK-OUT delay of the dynamic comparator is maintained. The prototype comparators with the conventional or the proposed reset scheme are fabricated in a 22-nm bulk CMOS process. The measurement results of the delay, input-referred noise, and energy consumption show that the flying-capacitor-based reset scheme boosts the overall performance of both the conventional StrongARM and Elzakker’s comparator by about 1.8-times.
- 25 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 72 (2) , 384-388
A Flying-Capacitor-Based Reset Scheme for Low Power Dynamic Comparator
A dynamic comparator reset scheme based on the flying capacitor is proposed. The reset power used to charge the integration nodes of the comparator is nearly halved by exploiting the existent integration capacitors with only a single D-flip-flop and few switches. The proposed reset scheme improves the energy efficiency while the small CLK-OUT delay of the dynamic comparator is maintained. The prototype comparators with the conventional or the proposed reset scheme are fabricated in a 22-nm bulk CMOS process. The measurement results of the delay, input-referred noise, and energy consumption show that the flying-capacitor-based reset scheme boosts the overall performance of both the conventional StrongARM and Elzakker’s comparator by about 1.8-times.