arXiv subjects

Hongqing Liu

Publications and source records attributed to Hongqing Liu.

7 recordsLinked to original sources

arXiv2026

DriveAnchor: Progressive Anchor-based Flow Learning for Autonomous Driving Planning

We present DriveAnchor, a three-stage framework for autonomous driving planning that achieves behavioral diversity, controllability, and safety in a composable pipeline. Demonstration Flow Pretraining replaces the unstructured Gaussian prior with a vocabulary of 2,398 trajectory shapes constructed by farthest-point sampling, structurally grounding behavioral diversity in vocabulary coverage. Guided Flow Post-training jointly post-trains an Energy Field module with flow matching (FM), conditioning the Energy Field on static road geometry alone, to relocate anchors toward user-specified corridor polygons before flow generation, adding controllability without differentiable guidance; after Stage 2, new corridor presets require only Energy Field updates, not FM retraining. Reward-Refined Flow Fine-tuning applies zeroth-order reinforcement learning to align each anchor's output with collision-avoidance objectives: because the flow-matching model is a deterministic feedforward network in single-step mode, each anchor uniquely determines the output trajectory, reducing reward optimization to a direction search in anchor space without log-likelihood computation or ODE-to-SDE conversion. Evaluated on approximately 2 million held-out driving scenarios, DriveAnchor reduces near-range collision rates by 89% and improves mean reward by 32% without degradation in imitation accuracy, with 2.06 ms inference on NVIDIA Drive Orin. DriveAnchor has been validated through real-world vehicle testing, confirming its practicality for production deployment.

cs.RO↗

arXiv2025

FPGA-Enabled Modulo ADC with x100 Dynamic-Range Expansion: Hardware Design and Performance Evaluation

Conventional analog-to-digital converters (ADCs) fail to capture high-dynamic-range (HDR) signals due to clipping. Modulo ADCs circumvent this limitation by folding the input prior to quantization and algorithmically reconstructing the original waveform. This work presents a field-programmable gate array (FPGA)-based modulo ADC platform for systematic HDR performance evaluation. The mixed-signal architecture integrates a precision analog front end with a 200-MHz FPGA control loop that incorporates multi-bit updates and digital under-compensation calibration, ensuring stable folding and accurate feedback generation. The system achieves more than a hundred-fold dynamic-range expansion within a 400-kHz bandwidth while maintaining fidelity comparable to that of a conventional ADC. A system-on-chip (SoC)-like implementation enables on-board real-time recovery and supports benchmarking of state-of-the-art reconstruction algorithms, providing a compact and practical framework for HDR signal acquisition and evaluation.

eess.SP↗

arXiv2025

Moduli Selection in Robust Chinese Remainder Theorem: Closed-Form Solutions and Layered Design

We study the fundamental problem of \emph{moduli selection} in the Robust Chinese Remainder Theorem (RCRT), where each residue may be perturbed by a bounded error. Consider $L$ moduli of the form $m_i = Γ_i m$ ($1 \le i \le L$), where $Γ_i$ are pairwise coprime integers and $m \in \mathbb{R}^+$ is a common scaling factor. For small $L$ ($L = 2, 3, 4$), we obtain exact solutions that maximize the robustness margin under dynamic-range and modulus-bound constraints. We also introduce a Fibonacci-inspired \emph{layered} construction (for $L = 2$) that produces exactly $K$ robust decoding layers, enabling predictable trade-offs between error tolerance and dynamic range. We further analyze how robustness and range evolve across layers and provide a closed-form expression to estimate the success probability under common data and noise models. The results are promising for various applications, such as sub-Nyquist sampling, phase unwrapping, range estimation, modulo analog-to-digital converters (ADCs), and robust residue-number-system (RNS)-based accelerators for deep learning. Our framework thus establishes a general theory of moduli design for RCRT, complementing prior algorithmic work and underscoring the broad relevance of robust moduli design across diverse information-processing domains.

eess.SP↗

arXiv2023

Super Denoise Net: Speech Super Resolution with Noise Cancellation in Low Sampling Rate Noisy Environments

Speech super-resolution (SSR) aims to predict a high resolution (HR) speech signal from its low resolution (LR) corresponding part. Most neural SSR models focus on producing the final result in a noise-free environment by recovering the spectrogram of high-frequency part of the signal and concatenating it with the original low-frequency part. Although these methods achieve high accuracy, they become less effective when facing the real-world scenario, where unavoidable noise is present. To address this problem, we propose a Super Denoise Net (SDNet), a neural network for a joint task of super-resolution and noise reduction from a low sampling rate signal. To that end, we design gated convolution and lattice convolution blocks to enhance the repair capability and capture information in the time-frequency axis, respectively. The experiments show our method outperforms baseline speech denoising and SSR models on DNS 2020 no-reverb test set with higher objective and subjective scores.

eess.AS↗

arXiv2023

Multi-Loss Convolutional Network with Time-Frequency Attention for Speech Enhancement

The Dual-Path Convolution Recurrent Network (DPCRN) was proposed to effectively exploit time-frequency domain information. By combining the DPRNN module with Convolution Recurrent Network (CRN), the DPCRN obtained a promising performance in speech separation with a limited model size. In this paper, we explore self-attention in the DPCRN module and design a model called Multi-Loss Convolutional Network with Time-Frequency Attention(MNTFA) for speech enhancement. We use self-attention modules to exploit the long-time information, where the intra-chunk self-attentions are used to model the spectrum pattern and the inter-chunk self-attention are used to model the dependence between consecutive frames. Compared to DPRNN, axial self-attention greatly reduces the need for memory and computation, which is more suitable for long sequences of speech signals. In addition, we propose a joint training method of a multi-resolution STFT loss and a WavLM loss using a pre-trained WavLM network. Experiments show that with only 0.23M parameters, the proposed model achieves a better performance than DPCRN.

cs.SD↗

arXiv2023

mdctGAN: Taming transformer-based GAN for speech super-resolution with Modified DCT spectra

Speech super-resolution (SSR) aims to recover a high resolution (HR) speech from its corresponding low resolution (LR) counterpart. Recent SSR methods focus more on the reconstruction of the magnitude spectrogram, ignoring the importance of phase reconstruction, thereby limiting the recovery quality. To address this issue, we propose mdctGAN, a novel SSR framework based on modified discrete cosine transform (MDCT). By adversarial learning in the MDCT domain, our method reconstructs HR speeches in a phase-aware manner without vocoders or additional post-processing. Furthermore, by learning frequency consistent features with self-attentive mechanism, mdctGAN guarantees a high quality speech reconstruction. For VCTK corpus dataset, the experiment results show that our model produces natural auditory quality with high MOS and PESQ scores. It also achieves the state-of-the-art log-spectral-distance (LSD) performance on 48 kHz target resolution from various input rates. Code is available from https://github.com/neoncloud/mdctGAN

eess.AS↗

arXiv2014

Performance and analysis of Quadratic Residue Codes of lengths less than 100

In this paper, the performance of quadratic residue (QR) codes of lengths within 100 is given and analyzed when the hard decoding, soft decoding, and linear programming decoding algorithms are utilized. We develop a simple method to estimate the soft decoding performance, which avoids extensive simulations. Also, a simulation-based algorithm is proposed to obtain the maximum likelihood decoding performance of QR codes of lengths within 100. Moreover, four important theorems are proposed to predict the performance of the hard decoding and the maximum-likelihood decoding in which they can explore some internal properties of QR codes. It is shown that such four theorems can be applied to the QR codes with lengths less than 100 for predicting the decoding performance. In contrast, they can be straightforwardly generalized to longer QR codes. The result is never seen in the literature, to our knowledge. Simulation results show that the estimated hard decoding performance is very accurate in the whole signal-to-noise ratio (SNR) regimes, whereas the derived upper bounds of the maximum likelihood decoding are only tight for moderate to high SNR regions. For each of the considered QR codes, the soft decoding is approximately 1.5 dB better than the hard decoding. By using powerful redundant parity-check cuts, the linear programming-based decoding algorithm, i.e., the ACG-ALP decoding algorithm performs very well for any QR code. Sometimes, it is even superior to the Chase-based soft decoding algorithm significantly, and hence is only a few tenths of dB away from the maximum likelihood decoding.

cs.IT↗