arXiv subjects

Yikang Wang

Publications and source records attributed to Yikang Wang.

13 recordsLinked to original sources

arXiv2026

Co-planning of Flight Corridors and Communication Infrastructure for Urban Drone Logistics Networks

Reliable wireless connectivity is essential for urban air mobility (UAM) networks in dense urban environments. It is therefore imperative to carefully plan the supporting communication infrastructure for UAM flight corridors. Most existing works optimize communication infrastructure and UAV flight paths independently, often leading to unnecessary base station (BS) deployment or excessive flight detours. This paper studies the joint optimization of BS deployment and UAV flight corridors in complex urban environments, aiming to minimize both infrastructure investment and flight distance while satisfying communication quality constraints. We propose CR-CMAB, a channel reciprocity-guided combinatorial multi-armed bandit framework. The framework constructs high-fidelity radio maps using 3D ray tracing, selects BS combinations via coverage-aware CMAB search, and dynamically expands the search space by identifying promising BS locations through channel reciprocity. Experimental results from a detailed case study demonstrate that CR-CMAB outperforms baseline methods with moderate computational time, yielding more strategically positioned BSs and shorter flight corridors. This study offers a practical planning perspective for cost-effective and communication-reliable UAM deployment in future smart cities.

cs.RO↗

arXiv2026

Seeing Before Agreeing: Aligning Multi-Agent Consensus with Visual Evidence

Vision-language models (VLMs) have achieved strong performance on visual question answering (VQA). To mitigate individual hallucinations and blind spots, aggregating diverse perspectives via multi-agent collaboration has emerged as a promising paradigm. While this approach has shown great success in textual QA, its potential in the multimodal domain remains under-explored. Existing multi-agent VQA methods predominantly adapt text-centric protocols, focusing on textual discussions while ignoring the alignment of visual information. In this work, we reveal a key insight: answer-level agreement is insufficient for reliable multi-agent VQA; \textit{aligned visual evidence} -- shared support from the image regions agents rely on -- is essential for trustworthy consensus. To leverage this insight, we propose EAGLE (\textbf{E}vidence-\textbf{A}ligned \textbf{G}rounded mu\textbf{L}ti-agent r\textbf{E}asoning), a training-free evidence-centered framework for coordinating multiple VLM agents. EAGLE explicitly exposes each agent's grounding regions as visual evidence, enables mutual verification over the evidence, and uses evidence consistency to guide final decision-making. Experiments on six VQA benchmarks show that EAGLE achieves best average performance across domains while remaining lightweight, interpretable, and practical for deployment.

cs.CV↗

arXiv2025

Astronomical image denoising by self-supervised deep learning and restoration processes

Image denoising based on deep learning has witnessed significant advancements in recent years. However, existing deep learning methods lack quantitative control of the deviation or error on denoised images. The neural networks Self2Self is designed for denoising single-image, training on it and denoising itself, during which training is costly. In this work we explore training Self2Self on an astronomical image and denoising other images of the same kind, which is suitable for quickly denoising massive images in astronomy. To address the deviation issue, the abnormal pixels whose deviation exceeds a predefined threshold are restored to their initial values. The noise reduction includes training, denoising, restoring and named TDR-method, by which the noise level of the solar magnetograms is improved from about 8 G to 2 G. Furthermore, the TDR-method is applied to galaxy images from the Hubble Space Telescope and makes weak galaxy structures become much clearer. This capability of enhancing weak signals makes the TDR-method applicable in various disciplines.

astro-ph.IM↗

arXiv2024

Enhancing Anti-spoofing Countermeasures Robustness through Joint Optimization and Transfer Learning

Current research in synthesized speech detection primarily focuses on the generalization of detection systems to unknown spoofing methods of noise-free speech. However, the performance of anti-spoofing countermeasures (CM) system is often don't work as well in more challenging scenarios, such as those involving noise and reverberation. To address the problem of enhancing the robustness of CM systems, we propose a transfer learning-based speech enhancement front-end joint optimization (TL-SEJ) method, investigating its effectiveness in improving robustness against noise and reverberation. We evaluated the proposed method's performance through a series of comparative and ablation experiments. The experimental results show that, across different signal-to-noise ratio test conditions, the proposed TL-SEJ method improves recognition accuracy by 2.7% to 15.8% compared to the baseline. Compared to conventional data augmentation methods, our system achieves an accuracy improvement ranging from 0.7% to 5.8% in various noisy conditions and from 1.7% to 2.8% under different RT60 reverberation scenarios. These experiments demonstrate that the proposed method effectively enhances system robustness in noisy and reverberant conditions.

cs.SD↗

arXiv2023

Pretraining Conformer with ASR or ASV for Anti-Spoofing Countermeasure

Finding synthetic artifacts of spoofing data will help the anti-spoofing countermeasures (CMs) system discriminate between spoofed and real speech. The Conformer combines the best of convolutional neural network and the Transformer, allowing it to aggregate global and local information. This may benefit the CM system to capture the synthetic artifacts hidden both locally and globally. In this paper, we present the transfer learning based MFA-Conformer structure for CM systems. By pre-training the Conformer encoder with different tasks, the robustness of the CM system is enhanced. The proposed method is evaluated on both Chinese and English spoofing detection databases. In the FAD clean set, proposed method achieves an EER of 0.04%, which dramatically outperforms the baseline. Our system is also comparable to the pre-training methods base on Wav2Vec 2.0. Moreover, we also provide a detailed analysis of the robustness of different models.

cs.SD↗

arXiv2023

The DKU-DUKEECE System for the Manipulation Region Location Task of ADD 2023

This paper introduces our system designed for Track 2, which focuses on locating manipulated regions, in the second Audio Deepfake Detection Challenge (ADD 2023). Our approach involves the utilization of multiple detection systems to identify splicing regions and determine their authenticity. Specifically, we train and integrate two frame-level systems: one for boundary detection and the other for deepfake detection. Additionally, we employ a third VAE model trained exclusively on genuine data to determine the authenticity of a given audio clip. Through the fusion of these three systems, our top-performing solution for the ADD challenge achieves an impressive 82.23% sentence accuracy and an F1 score of 60.66%. This results in a final ADD score of 0.6713, securing the first rank in Track 2 of ADD 2023.

eess.AS↗

arXiv2022

Low Pass Filtering and Bandwidth Extension for Robust Anti-spoofing Countermeasure Against Codec Variabilities

A reliable voice anti-spoofing countermeasure system needs to robustly protect automatic speaker verification (ASV) systems in various kinds of spoofing scenarios. However, the performance of countermeasure systems could be degraded by channel effects and codecs. In this paper, we show that using the low-frequency subbands of signals as input can mitigate the negative impact introduced by codecs on the countermeasure systems. To validate this, two types of low-pass filters with different cut-off frequencies are applied to countermeasure systems, and the equal error rate (EER) is reduced by up to 25% relatively. In addition, we propose a deep learning based bandwidth extension approach to further improve the detection accuracy. Recent studies show that the error rate of countermeasure systems increase dramatically when the silence part is removed by Voice Activity Detection (VAD), our experimental results show that the filtering and bandwidth extension approaches are also effective under the codec condition when VAD is applied.

cs.SD↗

arXiv2022

The DKU-OPPO System for the 2022 Spoofing-Aware Speaker Verification Challenge

This paper describes our DKU-OPPO system for the 2022 Spoofing-Aware Speaker Verification (SASV) Challenge. First, we split the joint task into speaker verification (SV) and spoofing countermeasure (CM), these two tasks which are optimized separately. For ASV systems, four state-of-the-art methods are employed. For CM systems, we propose two methods on top of the challenge baseline to further improve the performance, namely Embedding Random Sampling Augmentation (ERSA) and One-Class Confusion Loss(OCCL). Second, we also explore whether SV embedding could help improve CM system performance. We observe a dramatic performance degradation of existing CM systems on the domain-mismatched Voxceleb2 dataset. Third, we compare different fusion strategies, including parallel score fusion and sequential cascaded systems. Compared to the 1.71% SASV-EER baseline, our submitted cascaded system obtains a 0.21% SASV-EER on the challenge official evaluation set.

eess.AS↗

arXiv2022

Combination of Time-domain, Frequency-domain, and Cepstral-domain Acoustic Features for Speech Commands Classification

In speech-related classification tasks, frequency-domain acoustic features such as logarithmic Mel-filter bank coefficients (FBANK) and cepstral-domain acoustic features such as Mel-frequency cepstral coefficients (MFCC) are often used. However, time-domain features perform more effectively in some sound classification tasks which contain non-vocal or weakly speech-related sounds. We previously proposed a feature called bit sequence representation (BSR), which is a time-domain binary acoustic feature based on the raw waveform. Compared with MFCC, BSR performed better in environmental sound detection and showed comparable accuracy performance in limited-vocabulary speech recognition tasks. In this paper, we propose a novel improvement BSR feature called BSR-float16 to represent floating-point values more precisely. We experimentally demonstrated the complementarity among time-domain, frequency-domain, and cepstral-domain features using a dataset called Speech Commands proposed by Google. Therefore, we used a simple back-end score fusion method to improve the final classification accuracy. The fusion results also showed better noise robustness.

cs.SD↗

arXiv2021

Fast magnetic wave could heat the solar low-beta chromosphere

Magnetohydrodynamic (MHD) waves are candidates for heating the solar chromosphere, although it is still unclear which mode of the wave is dominant in heating. We perform two-dimensional radiative MHD simulation to investigate the propagation of MHD waves in the quiet region of the solar chromosphere. We identify the mode of the shock waves by using the relationship between gas pressure and magnetic pressure across the shock front and calculate their corresponding heating rate through the entropy jump to obtain a quantitative understanding of the wave heating process in the chromosphere. Our result shows that the fast magnetic wave is significant in heating the low-beta chromosphere. The low-beta fast magnetic waves are generated from high-beta fast acoustic waves via mode conversion crossing the equipartition layer. Efficient mode conversion is achieved by large attacking angles between the propagation direction of the shock waves and the chromospheric magnetic field.

astro-ph.SR↗

arXiv2020

Simulation of Alfven wave propagation in magnetic chromosphere with radiative loss: effects of non-linear mode coupling on chromospheric heating

We perform magnetohydrodynamic (MHD) simulations to investigate the propagation of Alfven wave in magnetic chromosphere. We use the 1.5-dimensional expanding flux tube geometry setting and transverse perturbation at the bottom to generate the Alfven wave. Compared with previous studies, our expansion is that we include the radiative loss term introduced by Carlsson & Leenaarts (2012). We find that when an observation based transverse wave generator is applied, the spatial distribution of the time-averaged radiative loss profile in our simulation is consistent with that in the classic atmospheric model. In addition, the energy flux in the corona is larger than the required value for coronal heating in the quiet region. Our study shows that the Alfven wave driven model has the potential to explain chromospheric heating and transport enough energy to the corona simultaneously.

astro-ph.SR↗

arXiv2016

Distributions of quasar hosts on the galaxy main-sequence plane

The relation between star formation rates and stellar masses, i.e. the galaxy main sequence, is a useful diagnostic of galaxy evolution. We present the distributions relative to the main sequence of 55 optically-selected PG and 12 near-IR-selected 2MASS quasars at z <= 0.5. We estimate the quasar host stellar masses from Hubble Space Telescope or ground-based AO photometry, and the star formation rates through the mid-infrared aromatic features and far-IR photometry. We find that PG quasar hosts more or less follow the main sequence defined by normal star-forming galaxies while 2MASS quasar hosts lie systematically above the main sequence. PG and 2MASS quasars with higher nuclear luminosities seem to have higher specific SFRs (sSFRs), although there is a large scatter. No trends are seen between sSFRs and SMBH masses, Eddington ratios or even morphology types (ellipticals, spirals and mergers). Our results could be placed in an evolutionary scenario with quasars emerging during the transition from ULIRGs/mergers to ellipticals. However, combined with results at higher redshift, they suggest that quasars can be widely triggered in normal galaxies as long as they contain abundant gas and have ongoing star formation.

astro-ph.GA↗

arXiv2015

Predicting the Arrival Time of Coronal Mass Ejections with the Graduated Cylindrical Shell and Drag Force Model

Accurately predicting the arrival of coronal mass ejections (CMEs) at the Earth based on remote images is of critical significance in the study of space weather. In this paper, we make a statistical study of 21 Earth directed CMEs, exploring in particular the relationship between CME initial speeds and transit times. The initial speed of a CME is obtained by fitting the CME with the Graduated Cylindrical Shell model and is thus free of projection effects. We then use the drag force model to fit results of the transit time versus the initial speed. By adopting different drag regimes, i.e., the viscous, aerodynamics, and hybrid regimes, we get similar results, with the least mean estimation error of the hybrid model of 12.9 hours. CMEs with a propagation angle (the angle between the propagation direction and the Sun-Earth line) larger than its half angular width arrive at the Earth with an angular deviation caused by factors other than the radial solar wind drag. The drag force model cannot be well applied to such events. If we exclude these events in the sample, the prediction accuracy can be improved, i.e., the estimation error reduces to 6.8 hours. This work suggests that it is viable to predict the arrival time of CMEs at the Earth based on the initial parameters with a fairly good accuracy. Thus, it provides a method of space weather forecast of 1--5 days following the occurrence of CMEs.

astro-ph.SR↗