Electrode coating measurement is a crucial task in practical lithium-ion battery systems, where the thickness and refractive index of the electrode coating directly reflect the battery's quality, energy density, capacity, and lifespan. In this paper, we propose the design, implementation, and evaluation of BatTera, the practical system for an accurate, high-resolution, non-destructive, and safe electrode coating measurement, with the ability to simultaneously measure coating thickness and refractive index. BatTera's contributions are twofold. Firstly, we build a comprehensive mathematical model that characterizes the arrival time of echo signals from both sides of the electrode coating by thoroughly analyzing the electrode structure based on “coating-foil-coating”. This model serves as a theoretical foundation guiding the measurement of coating thickness and refractive index. Secondly, we propose a series of effective signal-processing algorithms to address the practical challenges of double-side coating misalignment and deformation interference, thus adaptive improving the signal-to-noise ratio of Terahertz signals and pushing BatTera one big step closer to real adoptions. We implement BatTera based on the commercial Terahertz device QT-TO1000 and conduct extensive experiments using five types of cathode electrode samples in three different sizes, collected from one of the world's largest new energy battery manufacturers. The results show that BatTera achieves high measurement accuracy with a mean average error of 6.106$\upmu$m for thickness and 0.230 for refractive index.

Carrier interference is a fundamental research challenge in backscatter systems. Existing solutions leverage frequency shifting or full duplex designs to mitigate carrier interference in the analog or digital domain. However, these solutions introduce extra spectrum usage, protocol overhead, and power consumption, all of which are undesirable in backscatter systems. In this paper, we revisit polarization mismatch and propose Lotus, a low-cost analog design to combat carrier interference for backscatter systems. Lotus comprises a novel antenna design and a backscatter tag design. At runtime, Lotus antenna replaces the receiver default antenna to cancel out the carrier interference, without any protocol or hardware overhead. Lotus tag mitigates the power loss caused by polarization mismatch and remains compatible with all existing backscatter radios. Experimental results show that Lotus achieves comparable cancellation gain (i.e., 42 dB) with the state-of-the-art frequency-shifting baseline. Meanwhile, Lotus outperforms the baseline by 2× and 6.5× in spectrum and power efficiency, respectively. Additionally, Lotus achieves comparable performance to the frequency shifting baseline regarding backscatter range and throughput across three backscatter technologies, including Wi-Fi, Bluetooth, and LoRa.

Traditional Chinese Medicine (TCM) has made significant progress driven by AI. Previous approaches to herbal prescription recommendation (HPR) have focused mainly on the association between symptoms and herbs in prescriptions, neglecting the TCM compatibility mechanism of the ‘monarch, minister, assistant and envoy’ (MMAE) and information on herbal dosage. Therefore, this paper proposes a novel DHHNGDP for the prediction of the dose of the TCM formulation. The framework first utilizes an improved diffusion model (DM) trained in multidimensional herb features to globalize the symptomherb interaction patterns. Subsequently, the representation of the features is enriched by a heterogeneous hierarchical network designed for symptom-herbal nodes from the dimension of herbal dose. Finally, we combine the mechanism of TCM compatibility of MMAE with the syndrome-aware prediction process through the higher-order pooling layer to guide the herbal prescription generation process. Meanwhile, we constructed a TCM knowledge graph (KG) and 30-dimensional attribute information related to herbal dosage to complement the additional TCM knowledge. Extensive experiments on public datasets compared to leading-edge algorithms demonstrate the superiority of our approach. In addition, we applied analytical methods from network pharmacology and modern medicine to comprehensively evaluate instances of the DHHNGDP. This improved HPR method can provide a new idea for the integration and innovative development of modern TCM.

Artificial intelligence (AI) technology plays a crucial role in recommending prescriptions for traditional Chinese medicine (TCM). Previous studies have made significant progress by focusing on the symptom-herb relationship in prescriptions. However, several limitations hinder model performance: (i) Insufficient attention to patient-personalized information such as age, BMI, and medical history, which hampers accurate identification of syndrome and reduces efficacy. (ii) The typical long-tailed distribution of herb data introduces training biases and affects generalization ability. (iii) The oversight of the 'monarch, minister, assistant and envoy' compatibility among herbs increases the risk of toxicity or side effects, opposing the 'treatment based on syndrome differentiation' principle in clinical TCM. Therefore, we propose a novel hierarchical structure-enhanced personalized recommendation model for TCM formulas based on knowledge graph (KG) diffusion guidance, namely TCM-HEDPR. Specifically, we pre-train symptom representations using patient-personalized prompt sequences and apply prompt-oriented contrastive learning (CL) for data augmentation. Furthermore, we employ a KG-guided homogeneous graph diffusion method integrated with a self-attention mechanism to globally capture the non-linear symptom-herb relationship. Lastly, we design a heterogeneous graph hierarchical network to integrate herbal dispensing relationships with implicit syndromes, guiding the prescription generation process at a fine-grained level and mitigating the long-tailed herb data distribution problem. Extensive experiments on two public datasets and one clinical dataset demonstrate the effectiveness of TCM-HEDPR. In addition, we incorporate insights from modern medicine and network pharmacology to evaluate the recommended prescriptions comprehensively. It can provide a new paradigm for the recommendation of modern TCM.

The complexity and diversity of clinical cases in Traditional Chinese Medicine (TCM) pose significant challenges for accurate syndrome differentiation and intelligent prescription recommendation. To address these, we propose a novel TCM intelligent prescription model based on a symptom hypergraph convolutional network (HGCN). The model constructs a symptom knowledge network from unstandardized symptom descriptions, builds a symptom hypergraph, and uses HGCN to capture higher-order semantic relationships among symptoms, symptom features, syndromes, and herbs. These relationships are embedded into the hypergraph to train the model for accurate symptom classification and prescription generation. With optimized hyperparameters (e.g., feature embedding dimensions, HGCN layers), the model achieves high accuracy in symptom classification and exceptional prescription recommendation capabilities on specific datasets, demonstrating significant potential for advancing precision medicine in TCM clinical practice.

With the advancement of technology, Artificial Intelligence (AI) technology has advanced the development of Traditional Chinese Medicine (TCM) prescription recommendation approach. Existing models pay less attention to factors such as patients’ underlying diseases or past medical history as experimental indicators and lack of reasonable interpretations of the results of herbal recommendations. This is not consistent with the clinical diagnostic process that real-world TCM practitioners use to diagnose patients. To address this problem, we propose a personalized recommendation algorithm for Traditional Chinese Medicine prescription based on knowledge graph (MKPR-GAT). The algorithm first pre-trains the constructed TCM knowledge graph (KG) to enrich the collaborative signals of the symptom-herbal medicine node interaction model. Secondly, multi-graph co-learning is performed to model the relationships such as patient personalized information (gender, age and medical history) and symptom-herb in the form of graph embedding. In addition, in order to enhance the interpretability and accuracy of the algorithm, we additionally introduce the Proximity z-score and the Network proximity metrics and construct an association relationship graph. Extensive experiments compared to several baselines on a clinical real dataset and two public datasets demonstrate the effectiveness of MKPR-GAT. It not only provides new insights into the clinical practice of TCM, but also promotes the modernization and innovative development of TCM prescription recommendations.

Artificial intelligence (AI) has received much attention in the field of traditional Chinese medicine (TCM) prescription recommendation. Existing models pay less attention to patient-personalized attribute profiles and the problem of prescription data sparsity (imbalanced herb labels). To remedy this shortcoming, we propose a knowledge graph-guided diffusion model (DM) for personalized herbal prescription recommendation (HPR), namely TCM-KGDPR. We first learn patients’ personalized representations by the prompt fine-tuning technique and enhance the knowledge via the pre-trained model. Subsequently, we leveraged an improved DM to model the multivariate symptom-herb relationship and adopted a mixed node and edge-level strategy to alleviate the problem of imbalanced herb labels. In addition, we utilize a manually constructed TCM knowledge graph as a complement to the knowledge to capture the symptom-herb relationship on a global scale. Extensive experiments compared to several baselines on a real-world dataset and two public datasets demonstrate the effectiveness of TCM-KGDPR. Not only does it provide new information for clinical decision-making in TCM, but also it promotes the modernization and innovative development of TCM diagnosis and treatment.

In Mobile Cognitive Radio Sensor Networks (CRSNs), nodes have mobility which leads to the problem of changing the topology of the network as well as uneven energy consumption of the nodes. In this paper, an event-driven clustering routing algorithm is proposed to solve the above problems. The clustering algorithm fully considers the mobile and cognitive characteristics of the nodes, which is divided into three phases, the first phase is to determine the qualified nodes, the second phase is to determine the nodes that can become the cluster head, and the third phase is that the cluster head determines the cluster members based on the affinity and similarity between the nodes ultimately constructing a finite number of clusters. This algorithm can balance the energy consumption of nodes and increase the stability of clusters. In addition, this paper also proposes a routing scheme that guarantees the minimum number of route hops. Comparing the simulation experiments with the existing algorithms, the proposed algorithm in this paper has better connectivity, lower energy consumption and a better delivery rate.

The combination of Internet of Vehicles and Mobile Edge Computing can offload vehicular tasks to edge servers for execution, thereby reducing the vehicular computation burden and accelerating task execution. However, current research on vehicular edge task offloading focuses on centralized, small-scale environments, with incomplete consideration of distributed vehicular edge computing environments. Meanwhile, existing research ignores the stochastic nature of task arrivals and lacks research on collaborative processing among multiple edge servers. In this paper, considering the randomness of task arrivals and services, task offloading to edge servers for execution is modelled as a queueing-theoretic system and a customized fast search algorithm is used to find the optimal edge servers for further offloading co-execution in order to alleviate the load imbalance problem of edge servers. Meanwhile, a Distributed Distributional Deterministic Policy Gradients Deep Reinforcement Learning Algorithm is introduced to satisfy each user’s task offloading needs using offline distributed learning and centralized training architecture. Simulation results show that this scheme is more effective in reducing the delay and energy consumption of task offload execution compared to the baseline approach.

In traditional research on vehicular edge computing, Deep Reinforcement Learning (DRL) is usually used to solve the vehicular task offloading problem. However, DRL requires a large amount of real-time data and DRL is prone to the problem of possible local optimal solutions. For this reason, a Digital Twin(DT)-assisted task offloading scheme for vehicular edge computing is proposed in this paper, and a DT-based data acquisition method is proposed for the problem that DRL training requires a large amount of data. An improved algorithm for asynchronous dominant actor-critic (A3C) is proposed for the local optimal solution problem of A3C. The algorithm combines an ε — greedy strategy with the introduction of a dynamic baseline to dynamically update ε, while using asynchronous gradient descent to optimize the deep neural network controller. Simulation results show that the algorithm can effectively reduce the latency of on-board edge task offloading. Compared with the traditional A3C algorithm and the A3C combined with ε — greedy algorithm, the improved algorithm can effectively reduce the latency of task offloading for vehicular edge computing.

This paper presents the design, implementation, and evaluation of MagWear, a novel biomagnetism-based system that can accurately and inclusively monitor the heart rate and respiration rate of mobile users with diverse skin tones. MagWear's contributions are twofold. Firstly, we build a mathematical model that characterizes the magnetic coupling effect of blood flow under the influence of an external magnetic field. This model uncovers the variations in accuracy when monitoring vital signs among individuals. Secondly, leveraging insights derived from this mathematical model, we present a softwarehardware co-design that effectively handles the impact of human diversity on the performance of vital sign monitoring, pushing this generic solution one big step closer to real adoptions. We have implemented a prototype of MagWear on a two-layer PCB board and followed IRB protocols to conduct system evaluations. Our extensive experiments involving 30 volunteers demonstrate that MagWear achieves high monitoring accuracy with a mean percentage error (MPE) of 1.55% for heart rate and 1.79% for respiration rate. The head-to-head comparison with Apple Watch 8 further demonstrates MagWear's consistently high performance in different user conditions.

Abstract Traditional vehicular edge computing research usually ignores the mobility of vehicles, the dynamic variability of the vehicular edge environment, the large amount of real‐time data required for vehicular edge computing, the limited resources of edge servers, and collaboration issues. In response to these challenges, this article proposes a vehicular edge computing optimization scheme based on the Lyapunov function and Deep Reinforcement Learning. In this solution, this article uses Digital Twin technology (DT) to simulate the vehicular edge environment. The edge server DT is used to simulate the vehicular edge environment under the edge server, and the base station DT is used to simulate the entire vehicular edge system environment. Based on the real‐time data obtained from DT simulation, this paper defines the Lyapunov function to simplify the migration cost of vehicle tasks between servers into a multi‐objective dynamic optimization problem. It solves the problem by applying the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm. Experimental results show that compared with other algorithms, this scheme can effectively optimize the allocation and collaboration of vehicular edge computing resources and reduce the delay and energy consumption caused by vehicle task processing.

This paper presents the design, implementation, and evaluation of MagWear, a novel biomagnetism-based system that can accurately and inclusively monitor the heart rate, respiration rate, and blood pressure of users. MagWear's contributions are twofold. First, we build a mathematical model that characterizes the magnetic coupling effect of blood flow under the influence of an external magnetic field. This model uncovers the variations in accuracy when monitoring vital signs among individuals. Second, leveraging insights derived from this mathematical model, we present a software-hardware co-design that effectively handles the impact of human diversity on the performance of vital sign monitoring, pushing this generic solution one big step closer to real adoptions. Following IRB protocols, our extensive experiments involving 30 volunteers demonstrate that MagWear achieves high monitoring accuracy with a mean percentage error (MPE) of 1.55% for heart rate (HR), 1.79% for respiration rate (RR), 3.35% for systolic blood pressure (SBP), and 3.89% for diastolic blood pressure (DBP). MagWear can also be extended to detect anemia and blood oxygen saturation, which is also our ongoing work.

In this paper, we propose a delay-tolerant network (DTN) based message transmission algorithm, Message Prioritized DTN Routing Based on Node Mobility and Social Attributes (MP-NMSA), for disaster scenarios in which traditional communication devices and wireless devices are damaged, resulting in limited conventional communication technologies and wireless bandwidth congestion. First, machine learning techniques are utilized to process messages in disaster scenarios to achieve transmission prioritization of messages. Then, a mobile self-organizing DTN network is established to achieve message delivery in disaster environments based on the mobility and other social attributes of nodes. The experimental results show that MP-NMSA significantly improves delivery probability, average delay, and network overhead compared with traditional algorithms.

In this paper, we introduce QCBR, a protocol for VANETs in intelligent transportation systems. It addresses challenges, like large Q-learning state space and unstable links, to achieve lower latency and higher message delivery rates while maintaining stability. QCBR includes vehicle clustering, Qvalue table creation, and relay node selection based on stability. Simulations show improved data delivery rates and reduced endto-end delays compared to existing protocols.

This paper explores an orthogonal time frequency space (OTFS)-assisted integrated sensing and communication (ISAC) system in vehicular networks. We present a deep learning (DL)-based framework for the OTFS-assisted ISAC system, leveraging the advantages offered by the Delay-Doppler representation of the time-variant channel. The communication channel matrix is utilized within the framework to infer motion parameters, thereby enabling the establishment of an effective transmission protocol. Therefore, it is crucial to design a channel estimation method that simultaneously fulfills both sensing and communication performance requirements. To this end, a DL-based channel estimation approach is designed to obtain accurate channel state information (CSI), due to the powerful capability of neural networks [1]. Specifically, we model the channel estimation as a denoising problem from the embedded pilot scheme and employ a self-adaptive threshold submodule to eliminate irrelevant features. Finally, simulation results demonstrate that our proposed method can obtain accurate CSI with the available sensing performance.

In this paper, we explore graph analytics on a heterogeneous platform named HetGraph integrating with CPU and a flexible CGRA accelerator called RFU for matrix-based paradigm in this paper. RFU utilizes the lightweight pipeline without data hazards to support various generalized Sparse Matrix-Vector multiplications (SpMVs) of matrix-based graph analytics effectively. HetGraph utilizes the degree-aware workload distribution with vector-scanning sparsity removing scheme to alleviate the impact of highly sparse graph. Furthermore, we propose a heterogeneous work-stealing strategy to balance the workloads between CPU and RFU for HetGraph. To the best of our knowledge, HetGraph is the first heterogeneous CPU-CGRA architecture for matrix-based graph analytics. Overall, HetGraph achieves 9.42x, 2.45x speedup, and 9.80x, 7.70x energy savings on average compared to state-of-the-art (SOTA) CPU-based and GPGPU-based solutions respectively. Compared to the SOTA graph analytics accelerator, HetGraph also achieves 1.42x speedup and 1.06x less energy.

Aspect Sentiment Triplet Extraction (ASTE) task has been widely studied in Natural Language Processing (NLP). Nevertheless, existing ASTE methods are processed on a small public dataset, leading to frequent appeared sentiment words being focused on and neglecting the processing of niche sentiment words with strong sentiment tendencies. Therefore, the model extracts the wrong triples. To address this problem, we combine two attention weights of the Document-Level Sentiment Classification (DSC) and the ASTE based on the attention mechanism to obtain the ASTE representation with enhanced sentiment knowledge. Meanwhile, an attention masking strategy is introduced in the DSC model to ignore noisy words irrelevant to sentiment classification and ensure the effective extraction of sentiment triples. Experiments on four benchmark datasets from ASTE-Data-V1 show that our proposed method can acquire sentiment knowledge from large-scale document-level corpora, outperforming existing baseline models.

Utilizing the intelligence at the network edge, edge computing paradigm emerges to provide time-sensitive computing services for Internet of Things. In this paper, we investigate sustainable computation offloading in an edge-computing system that consists of energy harvesting-enabled mobile devices (MDs) and a dispatcher. The dispatcher collects computation tasks generated by IoT devices with limited computation power, and offloads them to resourceful MDs in exchange for rewards. We propose an online Rewards-optimal Auction (RoA) to optimize the long-term sum-of-rewards for processing offloaded tasks, meanwhile adapting to the highly dynamic energy harvesting (EH) process and computation task arrivals. RoA is designed based on Lyapunov optimization and Vickrey-Clarke-Groves auction, the operation of which does not require a prior knowledge of the energy harvesting, task arrivals, or wireless channel statistics. Our analytical results confirm the optimality of tasks assignment. Furthermore, simulation results validate the analytical analysis, and verify the efficacy of the proposed RoA.