Xicheng Lu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Xicheng Lu has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Haotian Liu
School of Electrical and Electronic Engineering, The University of Sheffield, Sheffield, U.K.
2 shared publications - Kaiyuan Yang
University of Sheffield
2 shared publications - Tiantai Deng
University of Sheffield
2 shared publications - Akram Alomainy
School of Electronic Engineering and Computer Science, Queen Mary University of London, London, U.K.
1 shared publication - Haihang Xia
School of Electrical and Electronic Engineering, The University of Sheffield, Sheffield, U.K.
1 shared publication Kai Li
School of Microelectronics, Southern University of Science and Technology, Shenzhen, CHINA
1 shared publication
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 34 (11) , 3629-3633
A Low-Bit-First Multiprecision Processing Element for Energy-Efficient FPGA Implementation
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 34 (11) , 3629-3633
A Low-Bit-First Multiprecision Processing Element for Energy-Efficient FPGA Implementation
- 5 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 73 (1053587X) , 4364-4378
A 1.5-Bit Quantization Scheme and Its Application to Direction Estimation
In massive multiple-input multiple-output (MIMO) systems, the balance between cost and performance has made low-bit, especially 1-bit, analog-to-digital converters (ADCs) an indispensable part of the solution. In this paper, a three-level 1.5-bit ADC quantization scheme is proposed, which requires only one additional comparator beyond the 1-bit quantizer. Leveraging the Price theorem and Mehler’s formula, we derive the 1.5-bit correlation estimator and analyze the approximation error using a first-order Taylor expansion. Our findings reveal that, at low signal-to-noise ratios (SNRs), the eigenvalues of the 1.5-bit covariance matrix are nearly identical to those of the unquantized covariance matrix. This allows direct parameter estimation without the need to reconstruct the unquantized covariance. Moreover, we show that the approximation error for 1.5-bit measurements is much smaller than that of 1-bit quantization in high SNR conditions. Based on the derived correlation estimator, an algorithm is proposed for recovering the unquantized covariance matrix using a gradient descent method. Simulation results obtained by applying our proposed algorithm to DOA estimation show that, the 1.5-bit scheme is robust to the choice of the threshold value, and significantly outperforms 1-bit quantization without much increase in cost.
- 5 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 73 (1053587X) , 4364-4378
A 1.5-Bit Quantization Scheme and Its Application to Direction Estimation
In massive multiple-input multiple-output (MIMO) systems, the balance between cost and performance has made low-bit, especially 1-bit, analog-to-digital converters (ADCs) an indispensable part of the solution. In this paper, a three-level 1.5-bit ADC quantization scheme is proposed, which requires only one additional comparator beyond the 1-bit quantizer. Leveraging the Price theorem and Mehler’s formula, we derive the 1.5-bit correlation estimator and analyze the approximation error using a first-order Taylor expansion. Our findings reveal that, at low signal-to-noise ratios (SNRs), the eigenvalues of the 1.5-bit covariance matrix are nearly identical to those of the unquantized covariance matrix. This allows direct parameter estimation without the need to reconstruct the unquantized covariance. Moreover, we show that the approximation error for 1.5-bit measurements is much smaller than that of 1-bit quantization in high SNR conditions. Based on the derived correlation estimator, an algorithm is proposed for recovering the unquantized covariance matrix using a gradient descent method. Simulation results obtained by applying our proposed algorithm to DOA estimation show that, the 1.5-bit scheme is robust to the choice of the threshold value, and significantly outperforms 1-bit quantization without much increase in cost.
- 19 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 72 (1) , 265-276
A 3-D Multi-Precision Scalable Systolic FMA Architecture
Artificial Intelligence (AI) has almost become the default approach in a wide range of applications, such as computer vision, chatbots, and natural language processing. These AI-based applications require computing large-scale data with sufficient precision, typically in floating-point numbers, within a limited time window. A primary target for AI acceleration is matrix multiplication, mainly involving dot products through Multiply-Accumulate (MAC) operations. Current research employs the Fused Multiply-Add (FMA) operation, based on IEEE-754 Floating Point (FP) standard, to meet these requirements. However, current research focuses more on simplifying the internal digital circuits of the Processing Elements (PEs) performing FMA operations, rather than optimizing the FMA process specifically for MAC tasks. Current PE arrays often use a two-dimensional (2-D) systolic array design, without specific optimization for MAC operations, thus their parallelism is not fully utilized. Additionally, these designs lack reconfigurability and flexibility, leading to suboptimal performance on Field-Programmable Gate Arrays (FPGAs). Moreover, some designs adopt lower precision computing in AI inference for higher performance. However, some AI models still rely on high-precision computing to maintain the accuracy. Thus, multi-precision computing is commonly used in AI accelerators. To address these challenges, this paper proposes a novel Multi-Fused Multiply-Accumulate (MFMA) scheme and a corresponding three-dimensional (3-D) scalable systolic FP computing architecture. The MFMA scheme addresses the problem of the classical FMA scheme. It optimizes FMA for MAC operations with the Fused Multiply-Accumulate (FMAC) operation. Also, it combines multi-precision and mixed-precision FP computing methods for higher accuracy and lower overflow error. The proposed architecture integrates two 2-D systolic arrays into the PE for a 3-D systolic array, achieving higher parallelism and flexibility. The proposed scalable architecture can be customized to suit various FMAC operations. Compared with existing state-of-the-art FP architectures on FPGAs, our proposed architecture achieves 47%, 10%, and 159% energy efficiency improvements in FP32, FP16, and INT8 operations, respectively. Furthermore, our proposed architecture achieves energy efficiency improvements of 105%, 54%, and 262% under efficiency saturation conditions, outperforming the existing state-of-the-art design.
- 19 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 72 (1) , 265-276
A 3-D Multi-Precision Scalable Systolic FMA Architecture
Artificial Intelligence (AI) has almost become the default approach in a wide range of applications, such as computer vision, chatbots, and natural language processing. These AI-based applications require computing large-scale data with sufficient precision, typically in floating-point numbers, within a limited time window. A primary target for AI acceleration is matrix multiplication, mainly involving dot products through Multiply-Accumulate (MAC) operations. Current research employs the Fused Multiply-Add (FMA) operation, based on IEEE-754 Floating Point (FP) standard, to meet these requirements. However, current research focuses more on simplifying the internal digital circuits of the Processing Elements (PEs) performing FMA operations, rather than optimizing the FMA process specifically for MAC tasks. Current PE arrays often use a two-dimensional (2-D) systolic array design, without specific optimization for MAC operations, thus their parallelism is not fully utilized. Additionally, these designs lack reconfigurability and flexibility, leading to suboptimal performance on Field-Programmable Gate Arrays (FPGAs). Moreover, some designs adopt lower precision computing in AI inference for higher performance. However, some AI models still rely on high-precision computing to maintain the accuracy. Thus, multi-precision computing is commonly used in AI accelerators. To address these challenges, this paper proposes a novel Multi-Fused Multiply-Accumulate (MFMA) scheme and a corresponding three-dimensional (3-D) scalable systolic FP computing architecture. The MFMA scheme addresses the problem of the classical FMA scheme. It optimizes FMA for MAC operations with the Fused Multiply-Accumulate (FMAC) operation. Also, it combines multi-precision and mixed-precision FP computing methods for higher accuracy and lower overflow error. The proposed architecture integrates two 2-D systolic arrays into the PE for a 3-D systolic array, achieving higher parallelism and flexibility. The proposed scalable architecture can be customized to suit various FMAC operations. Compared with existing state-of-the-art FP architectures on FPGAs, our proposed architecture achieves 47%, 10%, and 159% energy efficiency improvements in FP32, FP16, and INT8 operations, respectively. Furthermore, our proposed architecture achieves energy efficiency improvements of 105%, 54%, and 262% under efficiency saturation conditions, outperforming the existing state-of-the-art design.