Hong Kong University of Science and Technology
Hong Kong University of Science and Technology Institutional RepositoryNot a member yet
162821 research outputs found
Sort by
Noise and Interference Co-Suppression Technique in SPAD-Based Direct Time-of-Flight (d-ToF) Pulsed LiDAR Sensors
No full textSingle-photon avalanche diode (SPAD)-based direct time-of-flight (d-ToF) light detection and ranging (LiDAR) sensors are highly vulnerable to ambient noise, especially sunlight, and interference from nearby LiDAR systems. Simultaneously addressing both challenges remains difficult in real-world, densely populated environments. This paper presents a technique based on laser repetition frequency modulation for simultaneous suppression of background noise and strong interference. The method is analytically modeled for multi-target scenarios and experimentally validated on two platforms: a mechanically-scanning LiDAR with linear-mode detection and a flash LiDAR using a custom-fabricated SPAD sensor chip. Measurements show interference suppression of 15.85 dB with 4-bit pseudo-random modulation with the SPAD-based sensor, demonstrating the method's effectiveness in mitigating both noise and interference. Additionally, depth image of the scene is captured at 87-fps frame rate with 2.34−cm depth resolution
Accelerated dissolution mechanisms of rare earth elements in waste permanent magnet with oxygen vacancy
No full textSlow dissolution rate of magnet oxides is an important barrier to the separation of rare earth elements. We can combine mechanochemical processing with extraction using deep eutectic solvents (DES) to achieve an accelerated dissolution of iron (Fe) and neodymium (Nd) in permanent magnet oxides. The dissolution rate of Nd was found to increase to a greater extent compared to that of Fe, which was partly because iron oxide (Fe2O3) possessed a higher internal energy reserve than neodymium oxide (Nd2O3). The emergence of oxygen vacancies played a primary role for the enhanced dissolution and rate differentiation of Fe and Nd. Density functional theory calculations indicated that Nd2O3 was more likely to develop oxygen vacancies than Fe2O3. The process of Nd2O3 dissolution in the DES system (guanidine hydrochloride and lactic acid) occurred through a two-step reaction mechanism involving the adsorption of lactic acid molecules on oxygen vacancies and electronic interaction with lactic acid molecules. The defective oxygen vacancies in Nd2O3 (001) exhibited greater adsorption and binding affinity towards hydrogen ions and lactic acid molecules when compared to Fe2O3 (001). Our work provides a mechanistic understanding of the accelerated dissolution of rare earth elements in waste permanent magnets in DES systems.</p
Towards energy-positive and eco-profitable future: a perspective on building-to-vehicle-to-building ecosystems
No full textGlobally, sustainable urban development increasingly prioritizes energy-positive and eco-profitable systems. This study examines the emerging building-to-vehicle-to-building (V2B2) ecosystem—an integrated framework extending beyond traditional vehicle-to-grid (V2G), vehicle-to-home (V2H), and vehicle-to-building (V2B) technologies. Given limited direct V2B2 empirical evidence, this review develops the first systematic V2X-to-V2B2 extrapolation framework incorporating five quantitative assessment criteria with explicit confidence classifications. The analysis examines how V2B2 enhances energy system integration, reduces carbon emissions, enhances economic viability, and improves occupant comfort across four sustainability domains. Energy system integration enhancements range from 35-65 % (95 % CI: 41–55 %) compared to conventional grid-tied buildings, with grid electricity usage reductions of 52–77 % (95 % CI: 58–71 %). Economic projections indicate payback periods of 4.2–8.4 years (95 % CI: 5.4–7.2 years), while carbon reduction potential ranges from 38-83 % (95 % CI: 42–63 %) depending on regional grid carbon intensity and renewable energy penetration. The review explores challenges and opportunities associated with scaling V2B2 implementation, examining technological advancements, policy frameworks, infrastructure requirements, and consumer demands. Critical limitations include predominance of simulation-based findings (78 % of studies) and limited geographic scope. This study establishes replicable methodology for emerging technology assessment under data scarcity, providing transparent framework for evidence-based decision-making while maintaining scientific rigor through explicit uncertainty quantification.</p
Molecular Changes of a Chemosensory Component in Males that Facilitate the Mating System Transition from Dioecy to Androdioecy in Caenorhabditis Species
No full textThe nematode Caenorhabditis elegans has transitioned from a dioecious ancestry to an androdioecious reproductive system, wherein hermaphrodites possess the genetic capacity for self-sperm production, activation, and reduced dependence on mating. While the molecular adaptations enabling hermaphroditism are well characterized, the evolutionary trajectory of males during this shift remains less understood. Through comparative analyses across Caenorhabditis species, we reveal that androdioecious hermaphrodites display attenuated sex pheromone potency, whereas their male counterparts exhibit robust sex pheromone habituation and reduced mate-searching behaviour. Substitution of the SRD-1 receptor in androdioecious males with its dioecious orthologs restores ancestral dioecious male-like behaviour, implicating the receptor's C-terminal cytoplasmic domain in this functional divergence. We propose that C. elegans males have accumulated mutations in the sex pheromone specific receptor, altering their chemosensory perception of the opposite sex, which confers a selective advantage favouring hermaphroditism. These findings underscore the male-specific traits, that have been modulated by shifts in chemosensory signalling. These findings highlight how variations in sensory perception can reshape ecological interactions, ultimately contributing to the evolution and persistence of hermaphroditism.</p
Tensor completion via tensor train based low-rank quotient geometry under a preconditioned metric
No full textThe low-rank tensor completion problem is about recovering a tensor from partially observed entries. We consider this problem in the tensor train format and extend the preconditioned metric from the matrix case to the tensor case. The first-order and second-order quotient geometry of the manifold of fixed tensor train rank tensors under this metric is studied in detail. Algorithms, including Riemannian gradient descent, Riemannian conjugate gradient, and Riemannian Gauss–Newton, have been proposed for the tensor completion problem based on the quotient geometry. It has also been shown that the Riemannian Gauss–Newton method on the quotient geometry is equivalent to the Riemannian Gauss–Newton method on the embedded geometry with a specific retraction. Empirical evaluations on random instances as well as on function-related tensors show that the proposed algorithms are competitive with other existing algorithms in terms of recovery ability, convergence performance, and reconstruction quality.</p
Differential Organ Distribution of 7-(Deoxyadenosin-N6-yl)-aristolactam I in Mice Exposed to Aristolochic Acid I: Insights from Acute and Chronic Exposure Studies
No full textThe distribution of 7-(deoxyadenosin-N6-yl)-aristolactam I (ALI-dA) in mice treated with the same amount of aristolochic acid I (AA-I) at different dosage rates was investigated. The results showed a distinct organ distribution pattern of ALI-dA, with the highest adduct levels observed in the bladders of mice that received chronic doses of AA-I. In contrast, in mice that received AA-I acutely, the highest levels were found in the kidneys and were over ten times higher than those observed with chronic exposure. These results indicate that, in addition to the cumulative dose, the rate at which humans are exposed to AA-I is an under-recognized risk factor in the development of aristolochic acid nephropathy.</p
Robust Transmit Beamforming for Integrating Communication, Sensing and Power Transfer Systems
No full textIntegrating communication, sensing, and power transfer (ICSPT) is an emerging network paradigm for the sixth-generation (6G) systems, which is able to provide concurrent communication and sensing functions while simultaneously wirelessly powering low-power Internet of Things (IoT) devices with shared spectrum and hardware resources. To enhance the performance of ICSPT in fading channels, the outage probability (OP)constrained robust transmit beamforming design (OP-RTBD) is proposed, and a transmit power minimization problem is formulated with imperfect channel state information (CSI) by jointly optimizing information, sensing, and energy beam vectors at the base station (BS), subject to OP constraints on the communication rate, sensing Cramér-Rao bound, and energy transfer. To solve the non-convex problem, we propose a Bernstein-type inequality (BTI)-based method to conservatively approximate the probabilistic constraints to handle the CSI uncertainty. Then, a semi-positive definite relaxation-based method is proposed to solve the approximated problem. Simulation results show that the proposed OP-RTBD achieves near-optimal performance compared to the exhaustive search method with only less than 4% deviation, and it also significantly reduces the transmit power compared to baselines. Moreover, OP-RTBD exhibits strong robustness, achieving performance very close to that in perfect CSI scenarios, with a deviation of only less than 10%. Besides, the simulation results indicate that the BS’s transmit power should be allocated with priority to communication requirements over sensing and power transfer demands. Additionally, they further demonstrate that to simultaneously meet communication, sensing, and power transfer requirements, our proposed OP-RTBD in ICSPT is more energy-efficient, reducing energy consumption by approximately 10% and 20% compared to SWIPT and ISAC, respectively.</p
PlanScope: Learning to Plan Within Decision Scope for Urban Autonomous Driving
No full textIn the context of urban autonomous driving, imitation learning-based methods have shown remarkable effectiveness, with a typical practice to minimize the discrepancy between expert driving logs and predictive decision sequences. As expert driving logs natively contain future short-term decisions with respect to events, such as sudden obstacles or rapidly changing traffic signals. We believe that unpredictable future events and corresponding expert reactions can introduce reasoning disturbances, negatively affecting the convergence efficiency of planning models. At the same time, long-term decision information, such as maintaining a reference lane or avoiding stationary obstacles, is essential for guiding short-term decisions. Our preliminary experiments on shortening the planning horizon show a rise-andfall trend in driving performance, supporting these hypotheses. Based on these insights, we present PlanScope, a sequentialdecision- learning framework with novel techniques for separating short-term and long-term decisions in decision logs. To identify and extract each decision component, the Wavelet Transform on trajectory profiles is proposed. After that, to enhance the detailgenerating ability of Neural Networks, extra Detail Decoders are proposed. Finally, to enable in-scope decision supervision across detail levels, Multi-Scope Supervision strategies are adopted during training. The proposed methods, especially the timedependent normalization, outperform baseline models in closedloop evaluations on the nuPlan dataset, offering a plug-and-play solution to enhance existing planning models.</p
V2X-Assisted Distributed Computing and Control Framework for Connected and Automated CAVs under Ramp Merging Scenario
No full textThis paper presents a mobile computing-based framework for distributed computing and cooperative control of connected and automated vehicles (CAVs) in ramp merging scenarios under intelligent transportation systems (ITS). A centralized trajectory planning problem is first formulated to optimize merging efficiency and safety. To eliminate reliance on a central controller, a distributed solution is developed using ADMM algorithm based on V2X communication, enabling CAVs to collaboratively compute trajectories in parallel by leveraging their onboard computing power. Building on this, a multi-vehicle model predictive control (MPC) problem is proposed to enhance system stability under strict constraints. To solve it efficiently, a Distributed Cooperative Iterative MPC (DCIMPC) method is introduced, which decomposes and reformulates the problem for real-time distributed execution across CAVs. Together, these methods form a mobile edge computing-driven control framework. Simulations and experiments demonstrate significant improvements in computational efficiency and system performance, highlighting the potential of mobile computing in cooperative CAV control.</p
Toward Intelligent Design and Measurement of MEMS Acoustic Wave Resonators
No full textSurface acoustic wave (SAW) resonators are a cornerstone in 5G/6G radio frequency (RF) front ends owing to their compact form factor, strong electromechanical coupling, and compatibility with massive wafer-level manufacturability. Nevertheless, conventional design–test workflows remain constrained by extensive reliance on full-wave multiphysics simulations, iterative tape-out cycles, and exhaustive frequency-domain sweeps. Such processes incur high cost, long iteration cycles, and limited scalability, while spurious modes and parasitic effects are difficult to capture accurately. To overcome these challenges, this work establishes a data-driven design–test prediction framework that unifies modeling, validation, and optimization of SAW resonators within a single computational hierarchy. Training and evaluation use 14 883 industry-verified devices, 10 000 simulated structures, and 283 fabricated samples under a fixed split protocol. The system attains R2 > 0.99 on key metrics and 10-3-level mse for full admittance spectrum reconstruction from structure via ensemble regressors. Furthermore, a tailored masked U-Net-1D performs sparse-spectrum recovery from 16 uniformly spaced points (a 98% reduction from 1024 points) while preserving R2 > 0.98 on the 25% held-out test cohort. Beyond prediction, a physics-guided surrogate library enables inverse design. Given targets, the engine synthesizes previously unseen, spurious-suppressed resonators whose predicted spectrum aligns with measurements. Collectively, these elements bridge design and fabrication, jump out of physics and mathematical equations, sharply reduce simulation and test burden, and provide a scalable, physics-aware pathway to AI-driven automation of such microelectromechanical systems (MEMS) acoustic devices for next-generation RF applications.</p