As a sensor providing long-range perception and precise measurements, Light Detection and Ranging (LiDAR) can assist the development of smart cities in various aspects, which raises the rising interest in the topic of LiDAR Odometry (LO) research. Despite the popular application of LiDAR Odometry for the localization purpose, existing LiDAR Odometry methods are still limited by the estimation error in the elevation direction due to the low vertical resolution of LiDAR and the increasing storage cost with time. To address these issues, we propose Ground and Memory Optimized LiDAR Odometry (GMLO) for autonomous vehicles by integrating a joint scan-matching optimization between consecutive LiDAR frames. In addition, an efficient and robust ground extraction method, Polar Region Ground Plane Fitting (PR-GPF), is integrated into GMLO. Finally, we propose a novel scan-to-mapping scheme to eliminate accumulated errors with low storage cost and good efficiency in updating historical mapping data. To evaluate the performance of GMLO in various aspects, we have conducted extensive experiments based on the KITTI Odometry and Semantic-KITTI datasets, and the test results show that GMLO is effective with 1.03
Semantic communication (SC) is critical for the efficient and accurate transmission of data in complex environments, especially for unmanned surface vehicles (USVs) operating in marine settings characterized by strong interference and limited bandwidth. Traditional methods of transmitting large volumes of raw data are inefficient and susceptible to errors. Additionally, semantic communication systems, which depend on labeled datasets, have limited capacity to handle unfamiliar objects in dynamic marine environments. The long-tailed data distribution further complicates the recognition and processing of novel data types. To address these issues, this paper presents a novel approach that integrates zero-shot learning (ZSL) with SC, effectively enhancing image transmission and recognition tasks accuracy in various conditions. The proposed method employs semantic encoding and decoding to prioritize meaningful data, thereby enhancing transmission efficiency and accuracy. Additionally, we introduce semantic metrics for training and evaluating system performance, improving the ability to handle unfamiliar data types. Experimental results demonstrate that our algorithm significantly outperforms traditional methods in image transmission.
Current warranty systems for businesses with diverse products or extensive customer bases are bogged down by complexity. Traditional warranty management often faces challenges such as complexity, fraud, cost, customer dissatisfaction, and inefficiency. Businesses need to address these issues to effectively manage warranties, reduce costs, and improve customer satisfaction. For example, by implementing blockchain-based solutions, businesses can streamline warranty processes, reduce fraud, and enhance transparency. This research work presents an innovative, blockchain-powered solution to mitigate the above-stated issues. Each product is assigned an unique and secure warranty in the form of a Non-Fungible Token (NFT). This NFT tracks ownership and expiry directly on the blockchain, guaranteeing its validity and eliminating potential fraud. To ensure a user-friendly experience, account abstraction simplifies interaction with the system and reduces transaction fees. Finally, Gelato Ops automates the secure burning of expired NFTs upon reaching their designated date. This system streamlines warranty management for businesses, enhances trust through automation, and leverages the security and immutability of blockchain technology.
In ultra-high speed visible light communications (VLC) with light emitting diodes (LEDs) as the transmitter, analog pre-equalizers are often adopted to extend the modulation bandwidth. However, pre-equalizer may severely degrade the signal-to-noise-ratio (SNR) in the low frequency if we only focus on its bandwidth extending feature. And therefore, an important issue is that there exists a tradeoff between extending the system bandwidth and improving the SNR. In this paper, we first study the small signal characteristics of LEDs using the ABC parameters by taking into account of its internal quantum efficiency. And then, we propose an asymmetric bridged-T pre-equalizer considering an impedance matching network, based on which we study the bandwidth-SNR tradeoff problem via analytical derivations. We formulate a data rate maximization problem by considering the above tradeoff issue. Interestingly, different from common intuition, simulations results indicate that when the LED driving current and modulation bandwidth both reach maximal values, the achievable rate cannot reach the highest point. Specifically, using an OSRAM green LED, a maximum data rate of 1.15 Gbps can be achieved with a bias current of 378.63 mA and transmitter bandwidth of 163.31 MHz. The simulated data rate is 1.47 times higher than that of the literature using similar LED with 1/2 of the signal power attenuation. In the future, we will based on the optimized LED driving and equalization circuit to design PCB and test its impacts in real deployments.
Topological photonic crystals have emerged as an important branch of physics for their excellent ability to manipulate light. Furthermore, a large-area photonic crystal can provide an additional degree of freedom by inserting a photonic crystal structure without a band gap between topologically distinct photonic crystals. In this article, we propose a hybrid topological heterostructure with the potential to be applied in modern communication. The time-reversal symmetry is broken by applying an external magnetic field while the spatial-inversion symmetry is broken by rotating the dielectric components. Our system exhibits the characteristics of large-area one-way transmission that evenly distributes over the middle domain. The unidirectional excitation properties and inherent robustness are verified in a Z-shaped channel. The proposed heterostructure provides broad application prospects in photonic integrated networks and on-chip integrated communication systems.
Large Kernel Attention (LKA) has demonstrated superior performance in image classification/segmentation and surpassing transformer-based visual tasks. However, using 1 × 1 Conv for full-channel interaction in Large Kernel Separable Attention (LSKA) results in quadratic growth in computational complexity and memory usage as the number of channels increases. To alleviate this issue, this study proposes a Local-channel Large Separable Kernel Attention (LLSKA) module where the group convolution is employed to focus module solely on local-channel information to reduce the model complexity while maintaining system performance. The number of groups is determined by an adaptive strategy to determine the local-channel receptive ranges.
Omni-digital reconfigurable intelligent surfaces (Omni-DRIS) has been proposed to enhance the coverage of visible light communication (VLC) systems in recent years. Compared with traditional optical RIS using only reflections, Omni-DRIS is capable of extending the communication range of VLC even further, using both refection and refraction. Although RIS VLC system’s security performance inside one room has been studied in the literature, the security communication data rate is not directly applicable to Omni-DRIS systems, which cover two rooms simultaneously, using bi-directions transmissions. In this paper, we focus on the system where two rooms are interconnected using one Omni-DRIS in the middle. Alice node’s information is transmitted using both signal light emitting diodes (LEDs) and noise LEDs on the ceilings. The Eve node can appear in any room to eavesdropping Bob users. However, we argue that the minimum-secrecy-rate maximization problem in the Omni-DRIS setting is non-deterministic polynomial-time (NP) hard. To solve the optimization problem, we first make use of an alternating optimization (AO) algorithm to transform the original problem into two sub-problems, and then we adopt the successive convex approximation (SCA) method to solve the sub-problems. Simulation results show that, when the number of user is 2 and the LED transmitting power is 25 dBm, the minimum secrecy rate of the Omni-DRIS system is approximately 105
Traditional fuel vehicles emit large amounts of harmful gases, especially carbon dioxide. These gases will pose a huge threat to the ecological environment. With the emergence of electric vehicles (EVs), this problem has been alleviated and EVs have received more and more attention. Digital twin (DT) in EVs serve a pivotal role by creating a virtual model of a vehicle that mirrors the real-world vehicle. This technology enables real-time DT monitoring and diagnostics in all phases of EV production and operation. It permits the collection of data and analysis on how specific vehicles perform in different circumstances without physical testing. Therefore, the introduction of DT will bring significant potential for the development of EV. This paper explores the critical role of DT technology in EVs, specifically focusing on battery management systems, vehicle health monitoring, and propulsion drive systems. This will give readers a basic understanding of the DT technology and its potential future application in EVs.
The application of ground mobile robots in complex outdoor environments, particularly in rugged terrains, has been widely investigated to improve the capability. The terrain features significant variations in elevation, with dramatic surface undulations and diverse slope changes. In such terrain, path planning for mobile robots must balance both the distance of the path and safety considerations, and there are many safety issues that still need to be addressed. In unstructured environments, the determination of terrain traversability is crucial for generating global paths. This paper proposes a terrain traversability analysis based on Plane Fitting Rapidly-Exploring Random Trees* (PF-RRT*) to enhance the safety of ground mobile robots in outdoor terrains. The traversability metrics include sparsity and flatness of the planes containing the points along the PF-RRT* path, as well as the predicted pitch and roll angles of the mobile robot at those points. These metrics are incorporated into a cost function to assess the traversability of points along the path. The simulation results demonstrate that the proposed method could obtain safer paths with smoother velocities and has better performance compared with traditional strategies.
As the adoption of electric vehicles (EVs) continues to increase, integrating vehicle-to-grid (V2G) technology while maintaining mobility and optimizing traffic flow becomes crucial. In this paper, we formulate a mixed-autonomy traffic and wireless charging problem where both autonomous and human-driven EVs coexist on the road. We aim to mitigate traffic congestion and improve V2G functionality, and therefore we propose a framework that adopts reinforcement learning, specifically the Soft Actor-Critic algorithm, to jointly enhance traffic throughput, balance the state-of-charge (SOC) among EVs, and maximize wireless charging energy in the mixed-autonomy ring-shaped road with wireless charging facilities. Experimental results demonstrate the effectiveness of our approach in stabilizing traffic, achieving SOC balance, and increasing the energy charged to EVs. This study showcases the potential for solving mixed-autonomy traffic and wireless charging problems in future wireless charging transportation systems.
Blockchain technology, exemplified by Ethereum, relies on P2P networks for transaction and block propagation. Ethereum uses a hybrid transaction propagation protocol to conserve bandwidth, where nodes forward either full transactions or transaction hashes to their neighbors. While this approach saves bandwidth, it can introduce delays in transaction propagation, especially when nodes must request full transactions using GetData messages. In this paper, we introduce FLTP, a fast and low-bandwidth transaction propagation protocol designed to improve bandwidth efficiency and reduce propagation delays. Our analysis of Ethereum’s transaction propagation process reveals that nodes with higher connectivity tend to receive transactions faster due to fewer GetData requests. Based on these insights, in FLTP, each node maintains a table to track the frequency of GetData requests from each neighbor. This allows nodes to prioritize propagation transactions to those with a history of needing more data. Experiments conducted in a simulated Ethereum environment demonstrate that, compared to transaction propagation protocol in Ethereum, FLTP reduces data by 14.36
The widespread adoption of the Internet of Things (IoT) has led to a surge in image generation, with many being outsourced to the cloud to reduce storage pressures. To prevent from privacy leakage, it is wise to upload the encrypted form of images to the cloud server, however, encryption often leads to difficulty to image retrieval. Thus, the field of encrypted image retrieval has garnered significant interest for researchers. In this paper, a new encrypted JPEG image retrieval scheme using new adaptive images encryption algorithm and neural network is proposed. Specifically, the histograms of DCT coefficients are extracted from the cipher images as features vectors. These features vectors are then sent into a highly lightweight self-attention based neural network in retrieval process. Experiments results reveal that our scheme can attain enhanced retrieval precision with less computational cost compared to previous neural network based schemes. The lightweight self-attention networks (LSAN) we proposed can directly run with CPU, the GPU accelerated computing is not necessarily required. We upload our code at https://github.com/FrankZanyar/LSAN .
Recently, the rapid increase in electric vehicles (EVs) has significantly raised their integration into the power grid. This overlap between EV charging times and daily routines has created 'peaks upon peaks,' exerting substantial pressure on distribution networks. Therefore, shifting EV charging loads to off-peak periods is essential. The existing time-of-use (TOU) static pricing is no longer suitable for current demands, leading to the introduction of a dynamic pricing mechanism. EV charging demand is highly adjustable yet unpredictable, necessitating the proposal of a compliance incentive coefficient. Based on this, a dual-objective function is established, aiming to minimize charging costs and the grid load peak-valley difference, thereby enhancing the predictability and controllability of the charging load. The existing Particle Swarm Optimization (PSO) algorithm faces issues with global search capability and convergence speed. To address these shortcomings, the Grey Wolf Optimization (GWO) algorithm is proposed. This algorithm can achieve an orderly charging strategy for EVs under dual constraints, effectively reducing user charging costs and minimizing the grid’s peak-valley difference, successfully achieving the goal of peak shaving and valley filling.
This paper proposes an automated design method for microstrip filters which comprises topology exploration and parameter optimization stages. In the topology exploration stage, a novel grid-like topology is utilized in conjunction with an efficient generation and deduplication algorithm based on VF2++ isomorphism detection, significantly expanding the available design freedom of RF filter. The parameter optimization depends on a hybrid mode combining simulated annealing and gradient optimization, reducing computation time and simplifying the design process. To validate the method's effectiveness and practicality, a dual-band bandpass filter for 6G communication (1.8–2.7GHz and 3.4–3.7GHz) was successfully designed and implemented within 5 h (4 h of computation and 1 h of manual layout design). The electromagnetic (EM) simulation results show that the filter, within a compact size of 0.275λ_g× 0.285λ_g , achieves frequency selectivity with an insertion loss below 0.6 dB in the passband, roll-off rates of 71/100 dB/GHz and 180/87 dB/GHz for the first and second frequency passbands respectively, and effective out-of-band suppression.
In the context of escalating electricity demand and the critical challenge of electricity theft in smart grids, this paper presents a novel dual-attention fusion transformer model, termed DaFT, for improved electricity theft detection. By effectively integrating both horizontal and vertical attention mechanisms, our approach captures the multidimensional periodicity inherent in electricity consumption data, with a particular focus on weekly and day-specific patterns. Utilizing a comprehensive dataset from the State Grid Corporation of China, we rigorously analyze consumption behaviors to discriminate between legitimate users and electricity thieves. Experimental results reveal that DaFT outperforms state-of-the-art methods in terms of Mean Average Precision (MAP) and Area Under the Curve (AUC) metrics, achieving a MAP@100 of 0.988 and an AUC of 0.827 under optimal training conditions. Our results highlight the eligibility of the scheme to address the pressing issue of electricity theft, thereby contributing to the security and sustainability of smart grid systems. This work not only advances the field of electricity theft detection, but also opens avenues for applying similar methodologies to other time-series data analysis.
With the promotion and application of renewable energy sources such as wind and solar power, which effectively reduce carbon emissions, the intermittency and volatility of these renewable sources impact power quality and stability. Energy storage systems, with their rapid charge and discharge capabilities and ease of construction, can smooth out fluctuations in renewable energy output, reduce wind and solar curtailment, improve supply reliability, and support the achievement of carbon neutrality goals. Therefore, a multi-objective optimization scheme for energy storage capacity is proposed in this paper. For proposed scheme, a multi-objective optimization model for energy storage capacity under low-carbon constraints is first set up, then, the Non-Dominated Sorting Artificial Cooperative Search (NSACS) algorithm is integrated with Pareto evaluation to generate the Pareto front, moreover, the composite weighted technique for order preference by similarity to ideal solution (CW-TOPSIS) is proposed to determine the optimal solution on the Pareto front. At last, case studies demonstrate that the proposed scheme is able to automatically determine the optimal energy storage capacity effectively and achieve synergistic optimization of energy storage economic efficiency, reliability, and carbon reduction.
In modern maritime communication, although UAVs provide better line-of-sight and data transmission capabilities, ensuring secure and reliable transmission has become increasingly difficult due to unstable channel conditions and the ever-present threat of eavesdropping. This work considers a scenario in which a formation of UAVs communicates with USVs on the sea surface in the presence of eavesdroppers (EDs). To prevent eavesdropping and ensure the safety of the UAV formation, a distributed control strategy based on Control Barrier Functions (CBFs) is introduced. Unlike previous methods, the proposed CBFs are suitable for dynamic maritime environments and can easily adapt to various control constraints, such as collision avoidance and eavesdropping prevention, as discussed in this paper. The simulation results demonstrate the stability of the control strategy, providing new insights for the use of UAV-based systems in critical applications such as military operations, rescue missions, and marine transport.
Triple phase shift (TPS) is commonly utilized to enhance the efficiency of dual active bridge (DAB) DC/DC converters; however, the formulation of optimal phase-shift ratios poses a significant challenge given the versatility of the DAB converter in four distinct operational scenarios, each comprising five specific cases. Prior research has often sought the most effective local operation phase ratio set by analyzing parameters in every case, neglecting the importance of locating the global solution. Additionally, the expressions for circuit states such as peak or root-mean-square current value, and transmitted power in DAB fluctuate across different cases, further complicating the pursuit of global optimization. Consequently, it is crucial to examine the location of the optimal solution set when analyzing the optimal solution for a multi-case scenario. That is the pivotal task in this paper. A simulation is conducted to validate the case simplification analysis. The discussion space for general optimization is focused on cases 3 and 4, streamlining the analytical process and reducing the time required for comprehensive optimization.
Direction of arrival (DOA) estimation is a significant technology in navigation and positioning systems. With the development of machine learning, DOA estimation methods based on machine learning have shown great potential in indoor and outdoor navigation. This paper reviews the theoretical background and models of DOA and gives an overview of machine learning methods applied to DOA estimation. It also discusses DOA estimation methods based on machine learning and their applications in navigation. Machine learning can be divided into traditional machine learning methods, deep learning methods, and reinforcement learning methods. Traditional machine learning methods are support vector machines, k-nearest neighbor classification, etc. Deep learning methods consist of deep neural networks, convolutional neural networks, etc. By analyzing the challenges faced by current applications and future development directions, potential strategies for enhancing the performance of DOA estimation are proposed. This study aims to provide a comprehensive technical framework and a reference for future research for researchers in related fields.
In recent years autonomous driving and efficient traffic management have become pivotal areas of focus, amplifying the research on Vehicle-to-Everything (V2X) communication technology. However, most research on V2X has relied heavily on computer simulations due to the scarcity and high costs of practical hardware test-beds, often failing to capture real-world complexities. This research presents a novel, low-cost V2X test-bed developed using commodity hardware components, offering a platform for studying V2X communication. The proposed test-bed, which integrates On-Board Units (OBUs) and Roadside Units (RSUs), demonstrates its ability to achieve latency and packet loss that meet relevant standards through extensive experiments under various V2V (Vehicle-to-Vehicle) and V2I (Vehicle-to-Infrastructure) scenarios.
