arXiv subjects

Runhua Li

Publications and source records attributed to Runhua Li.

5 recordsLinked to original sources

arXiv2025

Adaptive Implicit-Based Deep Learning Channel Estimation for 6G Communications

With the widespread deployment of fifth-generation (5G) wireless networks, research on sixth-generation (6G) technology is gaining momentum. Artificial Intelligence (AI) is anticipated to play a significant role in 6G, particularly through integration with the physical layer for tasks such as channel estimation. Considering resource limitations in real systems, the AI algorithm should be designed to have the ability to balance the accuracy and resource consumption according to the scenarios dynamically. However, conventional explicit multilayer-stacked Deep Learning (DL) models struggle to adapt due to their heavy reliance on the structure of deep neural networks. This article proposes an adaptive Implicit-layer DL Channel Estimation Network (ICENet) with a lightweight framework for vehicle-to-everything communications. This novel approach balances computational complexity and channel estimation accuracy by dynamically adjusting computational resources based on input data conditions, such as channel quality. Unlike explicit multilayer-stacked DL-based channel estimation models, ICENet offers a flexible framework, where specific requirements can be achieved by adaptively changing the number of iterations of the iterative layer. Meanwhile, ICENet requires less memory while maintaining high performance. The article concludes by highlighting open research challenges and promising future research directions.

cs.IT↗

arXiv2022

Anomalous spontaneous emission dynamics at chiral exceptional points

An open quantum system operated at the spectral singularities where dimensionality reduces, known as exceptional points (EPs), demonstrates distinguishing behavior from the Hermitian counterpart. Here, we present an analytical description of local density of states (LDOS) for microcavity featuring chiral EPs, and unveil the anomalous spontaneous emission dynamics from a quantum emitter (QE) due to the non-Lorentzian response of EPs. Specifically, we reveal that a square Lorentzian term of LDOS contributed by chiral EPs can destructively interfere with the linear Lorentzian profile, resulting in the null Purcell enhancement to a QE with special transition frequency, which we call {\it{EP induced transparency}}. While for the case of constructive interference, the square Lorentzian term can narrow the linewidth of Rabi splitting even below that of bare components, and thus significantly suppresses the decay of Rabi oscillation. Interestingly, we further find that an open microcavity with chiral EPs supports atom-photon bound states for population trapping and decay suppression in long-time dynamics. As applications, we demonstrate the advantages of microcavity operated at chiral EPs in achieving high-fidelity entanglement generation and high-efficiency single-photon generation. Our work unveils the exotic cavity quantum electrodynamics unique to chiral EPs, which opens the door for controlling light-matter interaction at the quantum level through non-Hermiticity, and holds great potential in building high-performance quantum-optics devices.

quant-ph↗

arXiv2022

Fano effect induced giant and robust enhancement of photon correlations in cavity QED systems

Fano effect arising from interference between two dissipation channels to the radiation continuum enables to tune the photon statistics. Understanding the role of Fano effect and exploiting to achieve strong photon correlations are of both fundamental and applied significance. We present an analytical description of Fano-enhanced photon correlations based on cavity quantum electrodynamics to show that, the Fano effect in atom-cavity systems can improve the degree of antibunching by over four orders of magnitude. The enhancement factors and the optimal conditions are explicitly given, and found related to the Fano parameter $q$. Remarkably, the Fano enhancement manifests robustness against the decoherence processes, and can survive in the weak coupling regime. We expect our work to provide insight in tuning the photon statistics through Fano effect, which offers a new route to enhance the photon correlations, as well as the possibility of generating nonclassical light in a wider diversity of systems without the need of strong light-matter interaction.

quant-ph↗

arXiv2021

Single-photon blockade in quasichiral atom-photon interaction: Simultaneous high purity and high efficiency

We investigate the single-photon blockade (1PB) in the quasichiral regime of atom-photon interaction that mediates via dissipative environment, where the effective atom-photon interaction is asymmetrical but achiral. The synthetic magnetic current in the closed-loop coupling breaks down the reciprocity of atom-photon interaction, resulting in an asymmetrical and even completely unidirectional effective coupling between two selected quantum states. As an example, we couple the single-atom cavity-QED (cQED) system to a strongly dissipative auxiliary cavity. We find that in the quasichiral regime, the unconventional photon blockade (UPB) always incorporates with the conventional photon blockade (CPB) in the condition of maximum chirality. Furthermore, we show that 1PB in the quasichiral regime combines the advantages of UPB and CPB, demonstrating the perfect single-photon purity, higher efficiency, smooth time dynamics as well as lower requirement of modes coupling to achieve UPB. Our work paves the way for 1PB towards practical applications and reveals the intriguing quantum-optics phenomena in the quasichiral light-matter interaction.

physics.optics↗

arXiv2019

CMSN: Continuous Multi-stage Network and Variable Margin Cosine Loss for Temporal Action Proposal Generation

Accurately locating the start and end time of an action in untrimmed videos is a challenging task. One of the important reasons is the boundary of action is not highly distinguishable, and the features around the boundary are difficult to discriminate. To address this problem, we propose a novel framework for temporal action proposal generation, namely Continuous Multi-stage Network (CMSN), which divides a video that contains a complete action instance into six stages, namely Backgroud, Ready, Start, Confirm, End, Follow. To distinguish between Ready and Start, End and Follow more accurately, we propose a novel loss function, Variable Margin Cosine Loss (VMCL), which allows for different margins between different categories. Our experiments on THUMOS14 show that the proposed method for temporal proposal generation performs better than the state-of-the-art methods using the same network architecture and training dataset.

cs.CV↗