arXiv subjects

Lei Wan

Publications and source records attributed to Lei Wan.

18 recordsLinked to original sources

arXiv2026

Towards Collaborative Joint Perception and Prediction: Framework, Baseline Evaluation, and Deployment Perspectives

Connected Autonomous Vehicles (CAVs) increasingly exploit Vehicle-to-Everything (V2X) communication to exchange multi-source sensor information, enabling advanced Collaborative Perception (CP) capabilities. Extending beyond these capabilities, this work focuses on Collaborative Joint Perception and Prediction (Co-P&P), a paradigm that unifies CP with motion prediction to mitigate two persistent challenges: the accumulation of perception errors and visual occlusions. We present a conceptual framework for Collaborative Joint Perception and Prediction (Co-P&P) that improves motion prediction of surrounding road users, thereby enhancing situational awareness in complex and dynamic traffic environments. Building upon our preliminary study, this extended version compares the performance of different fusion strategies and establishes baseline performance for a modular design of perception and prediction. Experimental results show that prediction-level fusion leads to a decline in overall system performance compared to detection-level or tracking-level fusion. We further implement a minimal end-to-end Co-P&P prototype that couples collaborative point-cloud sharing via the RENO neural codec with joint detection-forecasting via FutureDet, showing that collaboration improves forecasting accuracy while neural compression preserves this benefit at roughly 34x lower communication bandwidth.

cs.CV↗

arXiv2026

Superimposed-Pilot OTFS Under Fractional Doppler: Modular End-to-End Learning

Orthogonal time frequency space (OTFS) modulation has emerged as a promising candidate to overcome the performance degradation of orthogonal frequency division multiplexing (OFDM), which are commonly encountered in high-mobility wireless communication scenarios. However, conventional OTFS transceivers rely on multiple separately designed signal-processing modules, whose isolated optimization often limits global optimal performance. To overcome limitations, this paper proposes a modular deep learning (DL) based end-to-end OTFS transceiver framework that consists of trainable and interchangeable neural network (NN) modules, including constellation mapping/demapping, superimposed pilot placement, inverse Zak (IZak)/Zak transforms, and a U-Net-enhanced NN tailored for joint channel estimation and detection (JCED), while explicitly accounting for the impact of the cyclic prefix. This physics-informed modular architecture provides flexibility for integration with conventional OTFS systems and adaptability to different communication configurations. Simulations demonstrate that the proposed design significantly outperforms baseline methods in terms of both normalized mean squared error (NMSE) and detection reliability, maintaining robustness under integer and fractional Doppler conditions. The results highlight the potential of DL-based end-to-end optimization to enable practical and high-performance OTFS transceivers for next-generation high-mobility networks.

eess.SP↗

arXiv2025

Breaking the 800 mV open-circuit voltage barrier in antimony sulfide photovoltaics

Sb2S3 is a promising material for low-toxicity, high-stability next-generation photovoltaics. Despite high optical limits in efficiency, progress in improving its device performance has been limited by severe voltage losses. Recent spectroscopic investigations suggest that self-trapping occurs in Sb2S3, limiting the open-circuit voltage (Voc) to a maximum of approximately 800 mV, which is the level the field has asymptotically approached. In this work, we surpass this voltage barrier through reductions in the defect density in Sb2S3 thin films by modulating the growth mechanism in chemical bath deposition using citrate ligand additives. Deep level transient spectroscopy identifies two deep traps 0.4-0.7 eV above the valence band maximum, and, through first-principles calculations, we identify these to likely be S vacancies, or Sb on S anti-sites. The concentrations of these traps are lowered by decreasing the grain boundary density from 1114+/-52 nm/um2 to 585+/-10 nm/um2, and we achieve a Voc of 824 mV, the record for Sb2S3 solar cells. This work addresses the debate in the field around whether Sb2S3 is limited by defects or self-trapping, showing that it is possible to improve the performance towards the radiative limit through careful defect engineering.

cond-mat.mtrl-sci↗

arXiv2025

Octupole-driven spin torque switching of antiferromagnetic tunnel junctions

Magnetic tunnel junctions (MTJs) based on ferromagnets are canonical devices in spintronics, with wide-ranging applications in data storage, computing, and sensing. They simultaneously exhibit mechanisms for electrical detection and control of magnetic order through the tunneling magnetoresistance (TMR) and spin-transfer torque (STT) effects, respectively. It was long assumed that neither of these effects could be sizeable in all-antiferromagnetic tunnel junctions (AATJs), since they exhibit no net magnetization. Recently, however, it was shown that AATJs based on chiral antiferromagnets do exhibit TMR due to their non-relativistic momentum-dependent spin polarization and cluster magnetic octupole moment (CMO), which are manifestations of their spin-split band structure. However, the reciprocal effect, i.e., the antiferromagnetic counterpart of STT, has been assumed non-existent due to the total electric current being spin-neutral. Here, we report nanoscale AATJs exhibiting this reciprocal effect, which we term octupole-driven spin-transfer torque (OTT). We demonstrate current-induced OTT switching of PtMn3|MgO|PtMn3 AATJs, exhibiting a TMR value of 363% at room temperature and switching current densities of the order of 10 MA/cm2. Our theoretical modeling explains the origin of OTT in terms of the imbalance between intra- and inter-sublattice spin currents across the AATJ, and equivalently, in terms of the non-zero net cluster octupole polarization of each PtMn3 layer. This work establishes a new materials platform for antiferromagnetic spintronics and provides a pathway towards deeply scaled magnetic memory and room-temperature terahertz technologies.

cond-mat.mtrl-sci↗

arXiv2025

VALISENS: A Validated Innovative Multi-Sensor System for Cooperative Automated Driving

Reliable perception remains a key challenge for Connected Automated Vehicles (CAVs) in complex real-world environments, where varying lighting conditions and adverse weather degrade sensing performance. While existing multi-sensor solutions improve local robustness, they remain constrained by limited sensing range, line-of-sight occlusions, and sensor failures on individual vehicles. This paper introduces VALISENS, a validated cooperative perception system that extends multi-sensor fusion beyond a single vehicle through Vehicle-to-Everything (V2X)-enabled collaboration between Connected Automated Vehicles (CAVs) and intelligent infrastructure. VALISENS integrates onboard and roadside LiDARs, radars, RGB cameras, and thermal cameras within a unified multi-agent perception framework. Thermal cameras enhances the detection of Vulnerable Road Users (VRUs) under challenging lighting conditions, while roadside sensors reduce occlusions and expand the effective perception range. In addition, an integrated sensor monitoring module continuously assesses sensor health and detects anomalies before system degradation occurs. The proposed system is implemented and evaluated in a dedicated real-world testbed. Experimental results show that VALISENS improves pedestrian situational awareness by up to 18% compared with vehicle-only sensing, while the sensor monitoring module achieves over 97% accuracy, demonstrating its effectiveness and its potential to support future Cooperative Intelligent Transport Systems (C-ITS) applications.

cs.RO↗

arXiv2025

Systematic Literature Review on Vehicular Collaborative Perception -- A Computer Vision Perspective

The effectiveness of autonomous vehicles relies on reliable perception capabilities. Despite significant advancements in artificial intelligence and sensor fusion technologies, current single-vehicle perception systems continue to encounter limitations, notably visual occlusions and limited long-range detection capabilities. Collaborative Perception (CP), enabled by Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communication, has emerged as a promising solution to mitigate these issues and enhance the reliability of autonomous systems. Beyond advancements in communication, the computer vision community is increasingly focusing on improving vehicular perception through collaborative approaches. However, a systematic literature review that thoroughly examines existing work and reduces subjective bias is still lacking. Such a systematic approach helps identify research gaps, recognize common trends across studies, and inform future research directions. In response, this study follows the PRISMA 2020 guidelines and includes 106 peer-reviewed articles. These publications are analyzed based on modalities, collaboration schemes, and key perception tasks. Through a comparative analysis, this review illustrates how different methods address practical issues such as pose errors, temporal latency, communication constraints, domain shifts, heterogeneity, and adversarial attacks. Furthermore, it critically examines evaluation methodologies, highlighting a misalignment between current metrics and CP's fundamental objectives. By delving into all relevant topics in-depth, this review offers valuable insights into challenges, opportunities, and risks, serving as a reference for advancing research in vehicular collaborative perception.

cs.CV↗

arXiv2025

R-LiViT: A LiDAR-Visual-Thermal Dataset Enabling Vulnerable Road User Focused Roadside Perception

In autonomous driving, the integration of roadside perception systems is essential for overcoming occlusion challenges and enhancing the safety of Vulnerable Road Users(VRUs). While LiDAR and visual (RGB) sensors are commonly used, thermal imaging remains underrepresented in datasets, despite its acknowledged advantages for VRU detection in extreme lighting conditions. In this paper, we present R-LiViT, the first dataset to combine LiDAR, RGB, and thermal imaging from a roadside perspective, with a strong focus on VRUs. R-LiViT captures three intersections during both day and night, ensuring a diverse dataset. It includes 10,000 LiDAR frames and 2,400 temporally and spatially aligned RGB and thermal images across 150 traffic scenarios, with 7 and 8 annotated classes respectively, providing a comprehensive resource for tasks such as object detection and tracking. The dataset and the code for reproducing our evaluation results are made publicly available.

cs.CV↗

arXiv2025

The Components of Collaborative Joint Perception and Prediction -- A Conceptual Framework

Connected Autonomous Vehicles (CAVs) benefit from Vehicle-to-Everything (V2X) communication, which enables the exchange of sensor data to achieve Collaborative Perception (CP). To reduce cumulative errors in perception modules and mitigate the visual occlusion, this paper introduces a new task, Collaborative Joint Perception and Prediction (Co-P&P), and provides a conceptual framework for its implementation to improve motion prediction of surrounding objects, thereby enhancing vehicle awareness in complex traffic scenarios. The framework consists of two decoupled core modules, Collaborative Scene Completion (CSC) and Joint Perception and Prediction (P&P) module, which simplify practical deployment and enhance scalability. Additionally, we outline the challenges in Co-P&P and discuss future directions for this research area.

cs.CV↗

arXiv2024

A Method for Fabricating CMOS Back-End-of-Line-Compatible Solid-State Nanopore Devices

Solid-state nanopores, nm-sized holes in thin, freestanding membranes, are powerful single-molecule sensors capable of interrogating a wide range of target analytes, from small molecules to large polymers. Interestingly, due to their high spatial resolution, nanopores can also identify tags on long polymers, making them an attractive option as the reading element for molecular information storage strategies. To fully leverage the compact and robust nature of solid-state nanopores, however, they will need to be packaged in a highly parallelized manner with on-chip electronic signal processing capabilities to rapidly and accurately handle the data generated. Additionally, the membrane itself must have specific physical, chemical, and electrical properties to ensure sufficient signal-to-noise ratios are achieved, with the traditional membrane material being SiNX . Unfortunately, the typical method of deposition, low-pressure vapour deposition, requires temperatures beyond the thermal budget of CMOS back-end-of-line integration processes, limiting the potential to generate an on-chip solution. To this end, we explore various lower-temperature deposition techniques that are BEOL-compatible to generate SiNx membranes for solid-state nanopore use, and successfully demonstrate the ability for these alternative methods to generate low-noise nanopores that are capable of performing single-molecule experiments.

physics.app-ph↗

arXiv2024

Additive engineering for Sb$_2$S$_3$ indoor photovoltaics with efficiency exceeding 17%

Indoor photovoltaics (IPVs) have attracted increasing attention for sustainably powering Internet of Things (IoT) electronics. Sb$_2$S$_3$ is a promising IPV candidate material with a bandgap of ~1.75 eV, which is near the optimal value for indoor energy harvesting. However, the performance of Sb$_2$S$_3$ solar cells is limited by nonradiative recombination, closely associated with the poor-quality absorber films. Additive engineering is an effective strategy to improved the properties of solution-processed films. This work shows that the addition of monoethanolamine (MEA) into the precursor solution allows the nucleation and growth of Sb$_2$S$_3$ films to be controlled, enabling the deposition of high-quality Sb$_2$S$_3$ absorbers with reduced grain boundary density, optimized band positions and increased carrier concentration. Complemented with computations, it is revealed that the incorporation of MEA leads to a more efficient and energetically favorable deposition for enhanced heterogeneous nucleation on the substrate, which increases the grain size and accelerates the deposition rate of Sb$_2$S$_3$ films. Due to suppressed carrier recombination and improved charge-carrier transport in Sb$_2$S$_3$ absorber films, the MEA-modulated Sb$_2$S$_3$ solar cell yields a power conversion efficiency (PCE) of 7.22% under AM1.5G illumination, and an IPV PCE of 17.55% under 1000 lux white light emitting diode (WLED) illumination, which is the highest yet reported for Sb$_2$S$_3$ IPVs. Furthermore, we construct high performance large-area Sb$_2$S$_3$ IPV modules to power IoT wireless sensors, and realize the long-term continuous recording of environmental parameters under WLED illumination in an office. This work highlights the great prospect of Sb$_2$S$_3$ photovoltaics for indoor energy harvesting.

cond-mat.mtrl-sci↗

arXiv2024

Hybrid thin-film lithium niobate micro-ring acousto-optic modulator for microwave-to-optical conversion

Highly efficient acousto-optic modulation plays a vital role in the microwave-to-optical conversion. Herein, we demonstrate a hybrid thin-film lithium niobate (TFLN) racetrack micro-ring acousto-optic modulator (AOM) implemented with low-loss chalcogenide (ChG) waveguide. By engineering the electrode configuration of the interdigital transducer, the double-arm micro-ring acousto-optic modulation is experimentally confirmed in nonsuspended ChG loaded TFLN waveguide platform. Varying the position of blue-detuned bias point, the half-wave-voltage-length product VpaiL of the hybrid TFLN micro-ring AOM is as small as 9 mVcm. Accordingly, the acousto-optic coupling strength is estimated to be 0.48 Hz s1/2 at acoustic frequency of 0.84 GHz. By analyzing the generation of phonon number from the piezoelectric transducer, the microwave-to-optical conversion efficiency is calculated to be 0.05%, approximately one order of magnitude larger than that of the state-of-the-art suspended counterpart. Efficient microwave-to-optical conversion thus provides new opportunities for low-power-consumption quantum information transduction using the TFLN-ChG hybrid piezo-optomechanical devices.

physics.optics↗

arXiv2024

Weaver: Foundation Models for Creative Writing

This work introduces Weaver, our first family of large language models (LLMs) dedicated to content creation. Weaver is pre-trained on a carefully selected corpus that focuses on improving the writing capabilities of large language models. We then fine-tune Weaver for creative and professional writing purposes and align it to the preference of professional writers using a suit of novel methods for instruction data synthesis and LLM alignment, making it able to produce more human-like texts and follow more diverse instructions for content creation. The Weaver family consists of models of Weaver Mini (1.8B), Weaver Base (6B), Weaver Pro (14B), and Weaver Ultra (34B) sizes, suitable for different applications and can be dynamically dispatched by a routing agent according to query complexity to balance response quality and computation cost. Evaluation on a carefully curated benchmark for assessing the writing capabilities of LLMs shows Weaver models of all sizes outperform generalist LLMs several times larger than them. Notably, our most-capable Weaver Ultra model surpasses GPT-4, a state-of-the-art generalist LLM, on various writing scenarios, demonstrating the advantage of training specialized LLMs for writing purposes. Moreover, Weaver natively supports retrieval-augmented generation (RAG) and function calling (tool usage). We present various use cases of these abilities for improving AI-assisted writing systems, including integration of external knowledge bases, tools, or APIs, and providing personalized writing assistance. Furthermore, we discuss and summarize a guideline and best practices for pre-training and fine-tuning domain-specific LLMs.

cs.CL↗

arXiv2021

Highly efficient acousto-optic modulation using nonsuspended thin-film lithium niobate-chalcogenide hybrid waveguides

A highly efficient on-chip acousto-optic modulator, as a key component, occupies an exceptional position in microwave-to-optical conversion. Homogeneous thin-film lithium niobate is preferentially employed to build the suspended configuration forming the acoustic resonant cavity to improve the modulation efficiency of the device. However, the limited cavity length and complex fabrication recipe of the suspended prototype restrain further breakthrough in the modulation efficiency and impose challenges for waveguide fabrication. In this work, based on a nonsuspended thin-film lithium niobate-chalcogenide glass hybrid Mach-Zehnder interferometer waveguide platform, we propose and demonstrate a built-in push-pull acousto-optic modulator with a half-wave-voltage-length product as low as 0.03 V cm, presenting a modulation efficiency comparable to that of the state-of-the-art suspended counterpart. Based on the advantage of low power consumption, a microwave modulation link is demonstrated using our developed built-in push-pull acousto-optic modulator. The nontrivial acousto-optic modulation performance benefits from the superior photoelastic property of the chalcogenide membrane and the completely bidirectional participation of the antisymmetric Rayleigh surface acoustic wave mode excited by the impedance-matched interdigital transducer, overcoming the issue of amplitude differences of surface acoustic waves applied to the Mach-Zehnder interferometer two arms in traditional push-pull acousto-optic modulators.

physics.app-ph↗

arXiv2021

Implementation of a Binary Neural Network on a Passive Array of Magnetic Tunnel Junctions

The increasing scale of neural networks and their growing application space have produced demand for more energy- and memory-efficient artificial-intelligence-specific hardware. Avenues to mitigate the main issue, the von Neumann bottleneck, include in-memory and near-memory architectures, as well as algorithmic approaches. Here we leverage the low-power and the inherently binary operation of magnetic tunnel junctions (MTJs) to demonstrate neural network hardware inference based on passive arrays of MTJs. In general, transferring a trained network model to hardware for inference is confronted by degradation in performance due to device-to-device variations, write errors, parasitic resistance, and nonidealities in the substrate. To quantify the effect of these hardware realities, we benchmark 300 unique weight matrix solutions of a 2-layer perceptron to classify the Wine dataset for both classification accuracy and write fidelity. Despite device imperfections, we achieve software-equivalent accuracy of up to 95.3 % with proper tuning of network parameters in 15 x 15 MTJ arrays having a range of device sizes. The success of this tuning process shows that new metrics are needed to characterize the performance and quality of networks reproduced in mixed signal hardware.

cs.ET↗

arXiv2021

Joint CFO, Gridless Channel Estimation and Data Detection for Underwater Acoustic OFDM Systems

In this paper, we propose an iterative receiver based on gridless variational Bayesian line spectra estimation (VALSE) named JCCD-VALSE that \emph{j}ointly estimates the \emph{c}arrier frequency offset (CFO), the \emph{c}hannel with high resolution and carries out \emph{d}ata decoding. Based on a modularized point of view and motivated by the high resolution and low complexity gridless VALSE algorithm, three modules named the VALSE module, the minimum mean squared error (MMSE) module and the decoder module are built. Soft information is exchanged between the modules to progressively improve the channel estimation and data decoding accuracy. Since the delays of multipaths of the channel are treated as continuous parameters, instead of on a grid, the leakage effect is avoided. Besides, the proposed approach is a more complete Bayesian approach as all the nuisance parameters such as the noise variance, the parameters of the prior distribution of the channel, the number of paths are automatically estimated. Numerical simulations and sea test data are utilized to demonstrate that the proposed approach performs significantly better than the existing grid-based generalized approximate message passing (GAMP) based \emph{j}oint \emph{c}hannel and \emph{d}ata decoding approach (JCD-GAMP). Furthermore, it is also verified that joint processing including CFO estimation provides performance gain.

cs.IT↗

arXiv2019

Immunity of nanoscale magnetic tunnel junctions to ionizing radiation

Spin transfer torque magnetic random access memory (STT-MRAM) is a promising candidate for next generation memory as it is non-volatile, fast, and has unlimited endurance. Another important aspect of STT-MRAM is that its core component, the nanoscale magnetic tunneling junction (MTJ), is thought to be radiation hard, making it attractive for space and nuclear technology applications. However, studies of the effects of high doses of ionizing radiation on STT-MRAM writing process are lacking. Here we report measurements of the impact of high doses of gamma and neutron radiation on nanoscale MTJs with perpendicular magnetic anistropy used in STT-MRAM. We characterize the tunneling magnetoresistance, the magnetic field switching, and the current-induced switching before and after irradiation. Our results demonstrate that all these key properties of nanoscale MTJs relevant to STT-MRAM applications are robust against ionizing radiation. Additionally, we perform experiments on thermally driven stochastic switching in the gamma ray environment. These results indicate that nanoscale MTJs are promising building blocks for radiation-hard non-von Neumann computing.

physics.app-ph↗

arXiv2017

Information-Coupled Turbo Codes for LTE Systems

We propose a new class of information-coupled (IC) Turbo codes to improve the transport block (TB) error rate performance for long-term evolution (LTE) systems, while keeping the hybrid automatic repeat request protocol and the Turbo decoder for each code block (CB) unchanged. In the proposed codes, every two consecutive CBs in a TB are coupled together by sharing a few common information bits. We propose a feed-forward and feed-back decoding scheme and a windowed (WD) decoding scheme for decoding the whole TB by exploiting the coupled information between CBs. Both decoding schemes achieve a considerable signal-to-noise-ratio (SNR) gain compared to the LTE Turbo codes. We construct the extrinsic information transfer (EXIT) functions for the LTE Turbo codes and our proposed IC Turbo codes from the EXIT functions of underlying convolutional codes. An SNR gain upper bound of our proposed codes over the LTE Turbo codes is derived and calculated by the constructed EXIT charts. Numerical results show that the proposed codes achieve an SNR gain of 0.25 dB to 0.72 dB for various code parameters at a TB error rate level of $10^{-2}$, which complies with the derived SNR gain upper bound.

cs.IT↗

arXiv2015

Bit Patterned Magnetic Recording: Theory, Media Fabrication, and Recording Performance

Bit Patterned Media (BPM) for magnetic recording provide a route to densities $>1 Tb/in^2$ and circumvents many of the challenges associated with conventional granular media technology. Instead of recording a bit on an ensemble of random grains, BPM uses an array of lithographically defined isolated magnetic islands, each of which stores one bit. Fabrication of BPM is viewed as the greatest challenge for its commercialization. In this article we describe a BPM fabrication method which combines e-beam lithography, directed self-assembly of block copolymers, self-aligned double patterning, nanoimprint lithography, and ion milling to generate BPM based on CoCrPt alloys. This combination of fabrication technologies achieves feature sizes of $<10 nm$, significantly smaller than what conventional semiconductor nanofabrication methods can achieve. In contrast to earlier work which used hexagonal close-packed arrays of round islands, our latest approach creates BPM with rectangular bitcells, which are advantageous for integration with existing hard disk drive technology. The advantages of rectangular bits are analyzed from a theoretical and modeling point of view, and system integration requirements such as servo patterns, implementation of write synchronization, and providing for a stable head-disk interface are addressed in the context of experimental results. Optimization of magnetic alloy materials for thermal stability, writeability, and switching field distribution is discussed, and a new method for growing BPM islands on a patterned template is presented. New recording results at $1.6 Td/in^2$ (teradot/inch${}^2$, roughly equivalent to $1.3 Tb/in^2$) demonstrate a raw error rate $<10^{-2}$, which is consistent with the recording system requirements of modern hard drives. Extendibility of BPM to higher densities, and its eventual combination with energy assisted recording are explored.

cond-mat.other↗