arXiv subjects

Tie Qiu

Publications and source records attributed to Tie Qiu.

17 recordsLinked to original sources

arXiv2026

Continuous-Variable MIMO THz Quantum Secret Sharing: Gaussian-modulation and Passive-modulation

Although quantum key distribution (QKD) enables information-theoretically secure key distribution, it is mainly designed for point-to-point communication and cannot directly support multi-user collaborative scenarios. To address this limitation, quantum secret sharing (QSS) has been proposed to enable secure multiparty key sharing. However, most existing QSS protocols rely on a single-input single-output (SISO) channel, which limits the achievable secret key rate (SKR) and transmission distance. This paper proposes a continuous-variable (CV) QSS protocol based on a multiple-input multiple-output (MIMO) architecture operating in the terahertz (THz) band. In the proposed scheme, transmit-receive beamforming decomposes the MIMO channel into multiple parallel SISO subchannels, thereby improving both the SKR and transmission distance. We describe the QSS transmission procedure and derive the SKR expressions for eight protocol variants under Gaussian collective attacks. Specifically, Gaussian modulation and passive modulation are considered at the transmitter, while homodyne and heterodyne detection are considered at the receiver. Both asymptotic and composable finite-size SKR formulas are derived to characterize the ideal upper-bound performance and the achievable performance under finite resources, respectively. Simulation results show that, under ideal assumptions including perfect channel state information, perfect phase synchronization, and ideal beamforming, the Gaussian-modulation protocol with a 32 x 32 antenna configuration and the passive-modulation protocol with a 1024 x 1024 antenna configuration achieve transmission distances of 14.99 m and 160 m in the atmospheric channel, respectively. These results provide an idealized theoretical benchmark for evaluating the potential performance gains of MIMO-assisted THz CV-QSS in indoor and short-range outdoor wireless networks.

quant-ph↗

arXiv2026

DLGStream: Dynamic Language-embedded Guassian Splatting for Open-vocabulary Enabled Free-viewpoint Video Streaming

3D Gaussian Splatting~(3DGS) has emerged as a promising paradigm for reconstructing streamable free-viewpoint video~(FVV) from multi-view videos. However, 3DGS-based FVVs typically lack user interaction and editing capabilities, which diminishes the immersive experience. Recent research has integrated language features from CLIP into 3DGS via distillation, enabling open-vocabulary queries and supporting many downstream applications. Nevertheless, the stringent requirements of FVV, low frame size and high FPS, make current language Gaussian representations unsuitable for language-embedded FVV. In this paper, we propose DLGStream, a novel language-embedded FVV representation that streams time-varying language features alongside Gaussian attributes to support 4D environment interaction, scene editing, and spatial intelligence. Specifically, we propose a dual-opacity dynamic language Gaussian representation, which maintains two opacity attributes for color and language features to deal with performance degradation that occurs when colors and features are jointly optimized. Furthermore, we introduce an interpolation-based deformation field to reduce temporal redundancy. This deformation field can also be used for 4D frame interpolation, boosting FVV sequences from low to high FPS. Experimental results demonstrate that DLGStream achieves superior performance in both on open-vocabulary segmentation and reconstruction quality with an average frame size of merely 43 KB. The code is available on \href{https://github.com/kkkzh/DLGStream}{https://github.com/kkkzh/DLGStream}.

cs.CV↗

arXiv2025

CometNet: Contextual Motif-guided Long-term Time Series Forecasting

Long-term Time Series Forecasting is crucial across numerous critical domains, yet its accuracy remains fundamentally constrained by the receptive field bottleneck in existing models. Mainstream Transformer- and Multi-layer Perceptron (MLP)-based methods mainly rely on finite look-back windows, limiting their ability to model long-term dependencies and hurting forecasting performance. Naively extending the look-back window proves ineffective, as it not only introduces prohibitive computational complexity, but also drowns vital long-term dependencies in historical noise. To address these challenges, we propose CometNet, a novel Contextual Motif-guided Long-term Time Series Forecasting framework. CometNet first introduces a Contextual Motif Extraction module that identifies recurrent, dominant contextual motifs from complex historical sequences, providing extensive temporal dependencies far exceeding limited look-back windows; Subsequently, a Motif-guided Forecasting module is proposed, which integrates the extracted dominant motifs into forecasting. By dynamically mapping the look-back window to its relevant motifs, CometNet effectively harnesses their contextual information to strengthen long-term forecasting capability. Extensive experimental results on eight real-world datasets have demonstrated that CometNet significantly outperforms current state-of-the-art (SOTA) methods, particularly on extended forecast horizons.

cs.CE↗

arXiv2025

StreamSTGS: Streaming Spatial and Temporal Gaussian Grids for Real-Time Free-Viewpoint Video

Streaming free-viewpoint video~(FVV) in real-time still faces significant challenges, particularly in training, rendering, and transmission efficiency. Harnessing superior performance of 3D Gaussian Splatting~(3DGS), recent 3DGS-based FVV methods have achieved notable breakthroughs in both training and rendering. However, the storage requirements of these methods can reach up to $10$MB per frame, making stream FVV in real-time impossible. To address this problem, we propose a novel FVV representation, dubbed StreamSTGS, designed for real-time streaming. StreamSTGS represents a dynamic scene using canonical 3D Gaussians, temporal features, and a deformation field. For high compression efficiency, we encode canonical Gaussian attributes as 2D images and temporal features as a video. This design not only enables real-time streaming, but also inherently supports adaptive bitrate control based on network condition without any extra training. Moreover, we propose a sliding window scheme to aggregate adjacent temporal features to learn local motions, and then introduce a transformer-guided auxiliary training module to learn global motions. On diverse FVV benchmarks, StreamSTGS demonstrates competitive performance on all metrics compared to state-of-the-art methods. Notably, StreamSTGS increases the PSNR by an average of $1$dB while reducing the average frame size to just $170$KB. The code is publicly available on https://github.com/kkkzh/StreamSTGS.

cs.CV↗

arXiv2025

GSVR: 2D Gaussian-based Video Representation for 800+ FPS with Hybrid Deformation Field

Implicit neural representations for video have been recognized as a novel and promising form of video representation. Existing works pay more attention to improving video reconstruction quality but little attention to the decoding speed. However, the high computation of convolutional network used in existing methods leads to low decoding speed. Moreover, these convolution-based video representation methods also suffer from long training time, about 14 seconds per frame to achieve 35+ PSNR on Bunny. To solve the above problems, we propose GSVR, a novel 2D Gaussian-based video representation, which achieves 800+ FPS and 35+ PSNR on Bunny, only needing a training time of $2$ seconds per frame. Specifically, we propose a hybrid deformation field to model the dynamics of the video, which combines two motion patterns, namely the tri-plane motion and the polynomial motion, to deal with the coupling of camera motion and object motion in the video. Furthermore, we propose a Dynamic-aware Time Slicing strategy to adaptively divide the video into multiple groups of pictures(GOP) based on the dynamic level of the video in order to handle large camera motion and non-rigid movements. Finally, we propose quantization-aware fine-tuning to avoid performance reduction after quantization and utilize image codecs to compress Gaussians to achieve a compact representation. Experiments on the Bunny and UVG datasets confirm that our method converges much faster than existing methods and also has 10x faster decoding speed compared to other methods. Our method has comparable performance in the video interpolation task to SOTA and attains better video compression performance than NeRV.

cs.CV↗

arXiv2024

TimeFormer: Capturing Temporal Relationships of Deformable 3D Gaussians for Robust Reconstruction

Dynamic scene reconstruction is a long-term challenge in 3D vision. Recent methods extend 3D Gaussian Splatting to dynamic scenes via additional deformation fields and apply explicit constraints like motion flow to guide the deformation. However, they learn motion changes from individual timestamps independently, making it challenging to reconstruct complex scenes, particularly when dealing with violent movement, extreme-shaped geometries, or reflective surfaces. To address the above issue, we design a plug-and-play module called TimeFormer to enable existing deformable 3D Gaussians reconstruction methods with the ability to implicitly model motion patterns from a learning perspective. Specifically, TimeFormer includes a Cross-Temporal Transformer Encoder, which adaptively learns the temporal relationships of deformable 3D Gaussians. Furthermore, we propose a two-stream optimization strategy that transfers the motion knowledge learned from TimeFormer to the base stream during the training phase. This allows us to remove TimeFormer during inference, thereby preserving the original rendering speed. Extensive experiments in the multi-view and monocular dynamic scenes validate qualitative and quantitative improvement brought by TimeFormer. Project Page: https://patrickddj.github.io/TimeFormer/

cs.CV↗

arXiv2024

DS-NeRV: Implicit Neural Video Representation with Decomposed Static and Dynamic Codes

Implicit neural representations for video (NeRV) have recently become a novel way for high-quality video representation. However, existing works employ a single network to represent the entire video, which implicitly confuse static and dynamic information. This leads to an inability to effectively compress the redundant static information and lack the explicitly modeling of global temporal-coherent dynamic details. To solve above problems, we propose DS-NeRV, which decomposes videos into sparse learnable static codes and dynamic codes without the need for explicit optical flow or residual supervision. By setting different sampling rates for two codes and applying weighted sum and interpolation sampling methods, DS-NeRV efficiently utilizes redundant static information while maintaining high-frequency details. Additionally, we design a cross-channel attention-based (CCA) fusion module to efficiently fuse these two codes for frame decoding. Our approach achieves a high quality reconstruction of 31.2 PSNR with only 0.35M parameters thanks to separate static and dynamic codes representation and outperforms existing NeRV methods in many downstream tasks. Our project website is at https://haoyan14.github.io/DS-NeRV.

cs.CV↗

arXiv2024

Machina Economicus: A New Paradigm for Prosumers in the Energy Internet of Smart Cities

Energy Internet (EI) is emerging as new share economy platform for flexible local energy supplies in smart cities. Empowered by the Internet-of-Things (IoT) and Artificial Intelligence (AI), EI aims to unlock peer-to-peer energy trading and sharing among prosumers, who can adeptly switch roles between providers and consumers in localized energy markets with rooftop photovoltaic panels, vehicle-to-everything technologies, packetized energy management, etc. The integration of prosumers in EI, however, will encounter many challenges in modelling, analyzing, and designing an efficient, economic, and social-optimal platform for energy sharing, calling for advanced AI/IoT-based solutions to resource optimization, information exchange, and interaction protocols in the context of the share economy. In this study, we aim to introduce a recently emerged paradigm, Machina Economicus, to investigate the economic rationality in modelling, analysis, and optimization of AI/IoT-based EI prosumer behaviors. The new paradigm, built upon the theory of machine learning and mechanism design, will offer new angles to investigate the selfishness of AI through a game-theoretic perspective, revealing potential competition and collaborations resulting from the self-adaptive learning and decision-making capacity. This study will focus on how the introduction of AI will reshape prosumer behaviors on the EI, and how this paradigm will reveal new research questions and directions when AI meets the share economy. With an extensive case analysis in the literature, we will also shed light on potential solutions for advancements of AI in future smart cities.

cs.CY↗

arXiv2022

TrustGNN: Graph Neural Network based Trust Evaluation via Learnable Propagative and Composable Nature

Trust evaluation is critical for many applications such as cyber security, social communication and recommender systems. Users and trust relationships among them can be seen as a graph. Graph neural networks (GNNs) show their powerful ability for analyzing graph-structural data. Very recently, existing work attempted to introduce the attributes and asymmetry of edges into GNNs for trust evaluation, while failed to capture some essential properties (e.g., the propagative and composable nature) of trust graphs. In this work, we propose a new GNN based trust evaluation method named TrustGNN, which integrates smartly the propagative and composable nature of trust graphs into a GNN framework for better trust evaluation. Specifically, TrustGNN designs specific propagative patterns for different propagative processes of trust, and distinguishes the contribution of different propagative processes to create new trust. Thus, TrustGNN can learn comprehensive node embeddings and predict trust relationships based on these embeddings. Experiments on some widely-used real-world datasets indicate that TrustGNN significantly outperforms the state-of-the-art methods. We further perform analytical experiments to demonstrate the effectiveness of the key designs in TrustGNN.

cs.LG↗

arXiv2021

High-Performance Permanent Magnet Array Design by a Fast Genetic Algorithm (GA)-based Optimization for Low-Field Portable MRI

A permanent magnet array (PMA) is a preferred source of magnetic field for body-part-dedicated low-field (<0.5 T) portable magnetic resonance imaging (MRI) because it has a small footprint, no power consumption, and no need for a cooling system. The current popular PMA is limited by the transversal field below 100 mT, where advanced technologies developed for the long-bore MRI systems (e.g., multi-channel techniques) cannot be applied. In this paper, a sparse high-performance PMA is proposed based on inward-outward ring pairs and using magnet blocks that can be bought off the shelf, targeting on portable head imaging. Through a fast genetic algorithm (GA)-based optimization, the proposed PMA has a longitudinal magnetic field with an average field strength of 111.40 mT and a monotonic field pattern with inhomogeneity of 10.57 mT (an RF bandwidth of <10%) within a Field of View (FoV) of 20 cm in diameter and 4.5 cm in length. The resulting field was validated using analytic calculations and numerical simulations. The encoding capability of the designed PMA was examined by checking the quality of the simulated reconstructed images. The proposed field outperforms a linear pattern for encoding. The PMA is 57.91 cm wide, 38 cm long, has a 5-gauss range of 87x87x104 cm^3, allowing an operation in a small space. It weights 126.08kg, comprising of a stationary main array that supplies a strong homogeneous field, and a light rotatable sub-gradient-array (5.12kg) that supplies a monotonic field for signal encoding. It has a magnetic field generation efficiency of 0.88mT/kg, the highest among sparse PMAs that offer a monotonic field pattern. The force experienced by each magnet was calculated to validate the feasibility of physical implementation. The proposed PMA can be a promising alternative to supply the main and gradient fields combined for dedicated portable MRI.

physics.med-ph↗

arXiv2021

Measuring Network Robustness by Average Network Flow

Infrastructure networks such as the Internet backbone and power grids are essential for our everyday lives. With the prevalence of cyber-attacks on them, measuring their robustness has become an important issue. To date, many robustness metrics have been proposed. It is desirable for a robustness metric to possess the following three properties: considering global network topologies, strictly increasing upon link additions, and having a quadratic complexity in terms of the number of nodes on sparse networks. This paper proposes to use Average Network Flow (ANF) as a robustness metric, and proves that it increases strictly, and gives an algorithm to compute ANF with a quadratic complexity by leveraging Gomory-Hu trees. Thus, with ANF intrinsically considering global network topologies, ANF is unveiled to be a new robustness metric satisfying those three properties. Moreover, this paper compares ANF with seven existing representative metrics, showing that each metric has its own characteristics, so there is no silver bullet in measuring network robustness and it is recommended to apply several metrics together to gain a comprehensive view. Finally, by experimenting on the scenarios in which network topologies preserve the same numbers of nodes and links, some interesting behaviors of robustness metrics are reported.

cs.SI↗

arXiv2016

A Fast Ellipse Detector Using Projective Invariant Pruning

Detecting elliptical objects from an image is a central task in robot navigation and industrial diagnosis where the detection time is always a critical issue. Existing methods are hardly applicable to these real-time scenarios of limited hardware resource due to the huge number of fragment candidates (edges or arcs) for fitting ellipse equations. In this paper, we present a fast algorithm detecting ellipses with high accuracy. The algorithm leverage a newly developed projective invariant to significantly prune the undesired candidates and to pick out elliptical ones. The invariant is able to reflect the intrinsic geometry of a planar curve, giving the value of -1 on any three collinear points and +1 for any six points on an ellipse. Thus, we apply the pruning and picking by simply comparing these binary values. Moreover, the calculation of the invariant only involves the determinant of a 3*3 matrix. Extensive experiments on three challenging data sets with 650 images demonstrate that our detector runs 20%-50% faster than the state-of-the-art algorithms with the comparable or higher precision.

cs.CV↗

arXiv2013

A Coverage Strategy for Wireless Sensor Networks in a Three-dimensional Environment

Coverage is one of the fundamental issues in wireless sensor networks (WSNs). It reflects the ability of WSNs to detect the fields of interest. In a real sensor networks application, the detection area is always non-ideal and the terrain of the detection area is often more complex in applications of three-dimensional sensor networks. Consequently, many of the existing coverage strategies cannot be directly applied to three-dimensional spaces. This paper presents a new coverage strategy for the three-dimensional sensor networks. Sensor nodes are uniformly distributed. The cost factor is utilized to construct the perceived probability and the classical watershed algorithm after the transformation of points from the three-dimensional space to the two-dimensional plane using the dimensionality reduction method, which can maintain the topology characteristic of the non-linear terrain. The detection probability in the optimal breath path is used as the measure to evaluate the coverage. Simulation results indicate that the proposed strategy can determine the coverage with fewer nodes, while achieving the coverage requirements of the networks.

cs.NI↗

arXiv2013

A Localization Strategy Based on N-times Trilateral Centroid with Weight

Localization based on received signal strength indication (RSSI) is a low cost and low complexity technology, and it is widely applied in distance-based localization of wireless sensor networks (WSNs). Error of existed localization technologies is significant. This paper presents the N-times trilateral centroid weighted localization algorithm (NTCWLA), which can reduce the error considerably. Considering the instability of RSSI, we use the weighted average of many RSSIs as current RSSI. To improve the accuracy we select a number of (no less than three) reliable beacon nodes to increase the localization times. Then we calculate the distances between reliable beacon nodes and the mobile node using an empirical formula. The mobile node is located N times using the trilateral centroid algorithm. Finally, we take the weighted average of the filtered reference coordinates as the mobile node's coordinates. We conduct experiments with the STM32W108 chip which supports IEEE 802.15.4. The results show that the proposed algorithm performs better than the trilateral centroid algorithm.

cs.NI↗

arXiv2013

Evaluating IEEE 802.15.4 for Cyber-Physical Systems

With rapid advancements in sensing, networking, and computing technologies, recent years have witnessed the emergence of cyber-physical systems (CPS) in a broad range of application domains. CPS is a new class of engineered systems that features the integration of computation, communications, and control. In contrast to general-purpose computing systems, many cyber-physical applications are safety-cricial. These applications impose considerable requirements on quality of service (QoS) of the employed networking infrastruture. Since IEEE 802.15.4 has been widely considered as a suitable protocol for CPS over wireless sensor and actuator networks, it is of vital importance to evaluate its performance extensively. Serving for this purpose, this paper will analyze the performance of IEEE 802.15.4 standard operating in different modes respectively. Extensive simulations have been conducted to examine how network QoS will be impacted by some critical parameters. The results are presented and analyzed, which provide some useful insights for network parameter configuration and optimization for CPS design.

cs.NI↗

arXiv2013

A Search Strategy of Level-Based Flooding for the Internet of Things

This paper deals with the query problem in the Internet of Things (IoT). Flooding is an important query strategy. However, original flooding is prone to cause heavy network loads. To address this problem, we propose a variant of flooding, called Level-Based Flooding (LBF). With LBF, the whole network is divided into several levels according to the distances (i.e., hops) between the sensor nodes and the sink node. The sink node knows the level information of each node. Query packets are broadcast in the network according to the levels of nodes. Upon receiving a query packet, sensor nodes decide how to process it according to the percentage of neighbors that have processed it. When the target node receives the query packet, it sends its data back to the sink node via random walk. We show by extensive simulations that the performance of LBF in terms of cost and latency is much better than that of original flooding, and LBF can be used in IoT of different scales.

cs.NI↗

arXiv2011

A Failure Self-recovery Strategy with Balanced Energy Consumption for Wireless Ad Hoc Networks

In energy constrained wireless sensor networks, it is significant to make full use of the limited energy and maximize the network lifetime even when facing some unexpected situation. In this paper, all sensor nodes are grouped into clusters, and for each cluster, it has a mobile cluster head to manage the whole cluster. We consider an emergent situation that one of the mobile cluster heads is broken down, and hence the whole cluster is consequently out of work. An efficient approach is proposed for recovering the failure cluster by selecting multiple static sensor nodes as the cluster heads to collect packets and transmit them to the sink node. Improved simulated annealing algorithm is utilized to achieve the uniform deployment of the cluster heads. The new cluster heads are dynamically changed in order to keep balanced energy consumption. Among the new cluster heads, packets are transmitted through multi-hop forwarding path which is cost-lowest path found by Dijkstra's algorithm. A balanced energy consumption model is provided to help find the cost-lowest path and prolong the lifetime of the network. The forwarding path is updated dynamically according to the cost of the path and residual energy of the node in that path. The experimental results show that the failure cluster is recovered and the lifetime of the cluster is prolonged.

cs.NI↗