International Journal of Intelligent Systems最新文献

Millimeter-wave (mmWave) and massive multiple-input multiple-output (MIMO) technologies are critical in current and future communication research. They play an essential role in meeting the demands for high-capacity, high-speed, and low-latency communication brought about by technological advancements. However, existing mmWave channel estimation schemes rely on idealized common sparse channel support assumptions, and their performance significantly degrades when encountering beam squint scenarios. To address this issue, this paper introduces a dynamic support detection window (DSDW) algorithm. This algorithm dynamically adjusts the position and size of the window based on the received signal strength, thereby better capturing signal strength variations and obtaining a more complete set of signal supports. The DSDW algorithm can better capture and utilize the sparsity of the channel, improving the efficiency and accuracy of the channel state information acquisition. By combining the beam-split pattern (BSP) algorithm with the DSDW algorithm, this paper designs an effective method to address the inherent beam-spreading problem in mmWave scenarios. Simulation results are proposed to demonstrate the effectiveness of the BSP-DSDW algorithm.

Metro system serves as the backbone for urban public transportation. Accurate passenger load prediction for the metro system plays a crucial role in metro service quality improvement, such as helping operators schedule train timetables and passengers plan their trips. However, existing works can only predict low-grained passenger flows of origin-destination (O-D) paths or inflows/outflows of each station but cannot predict passenger load distribution over the whole metro network. To this end, this paper proposes an end-to-end inference framework, PIPE, for passenger load prediction of every metro segment between two adjacent stations, by only utilizing smart card data. In particular, PIPE includes two modules. The first is the core. It formulates the travel time distribution of each metro segment as a truncated Gaussian distribution. Since there might be several possible routes for certain O-D paths, the population-level travel time distribution of these O-D paths would be a mixture of travel times of different routes. Considering the route preference may change over time, a dynamic truncated Gaussian mixture model is proposed for parameter inference of each truncated Gaussian distribution of each metro segment. The second module serves as the supplement, which compiles a bunch of methods for predicting passenger flows of O-D paths. Built upon them, PIPE is able to predict the travel time that future passengers of each O-D path will take for passing each metro segment and consequently can predict the passenger load of each metro segment in the short future. Numerical studies from Singapore’s metro system demonstrate the efficacy of our method.

Arrhythmia is a prevalent cardiovascular disease, which has garnered widespread attention due to its age-related increases in mortality rates. In the analysis of arrhythmia, the electrocardiogram (ECG) plays an important role. Arrhythmia classification often suffers from a significant data imbalance issue due to the limited availability of data for certain arrhythmia categories. This imbalance problem significantly affects the classification performance of the model. To address this challenge, data augmentation emerges as a viable solution, aiming to neutralize the adverse effects of imbalanced datasets on the model. To this end, this paper proposes a novel Multimodality Data Augmentation Network (MM-DANet) for arrhythmia classification. The MM-DANet consists of two modules: the multimodality data matching-based data augmentation module and the multimodality feature encoding module. In the multimodality data matching-based data augmentation module, we expand the underrepresented arrhythmia categories to match the size of the largest category. Subsequently, the multimodality feature encoding module employs convolutional neural networks (CNN) to extract the modality-specific features from both signals and images and concatenate them for efficient and accurate classification. The MM-DANet was evaluated on the MIT-BIH Arrhythmia Database and achieving an accuracy of 98.83%, along with an average specificity of 98.87%, average sensitivity of 92.92%, average precision of 91.05%, and average F1_score of 91.96%. Furthermore, its performance was also assessed on the St. Petersburg INCART arrhythmia database and the MIT-BIH supraventricular arrhythmia database, yielding AUC values of 81.98% and 90.93%, respectively. These outstanding results not only underscore the effectiveness of MM-DANet but also indicate its potential for facilitating reliable automated analysis of arrhythmias.

This paper introduces an inclusive class of fixed-time stable continuous-time gradient flows (GFs). This class of GFs is then leveraged to learn optimal control solutions for nonlinear systems in fixed time. It is shown that the presented GF guarantees convergence within a fixed time from any initial condition to the exact minimum of functions that satisfy the Polyak–Łojasiewicz (PL) inequality. The presented fixed-time GF is then utilized to design fixed-time optimal adaptive control algorithms. To this end, a fixed-time reinforcement learning (RL) algorithm is developed on the basis of a single network adaptive critic (SNAC) to learn the solution to an infinite-horizon optimal control problem in a fixed-time convergent, online, adaptive, and forward-in-time manner. It is shown that the PL inequality in the presented RL algorithm amounts to a mild inequality condition on a few collected samples. This condition is much weaker than the standard persistence of excitation (PE) and finite duration PE that relies on a rank condition of a dataset. This is crucial for learning-enabled control systems as control systems can commit to learning an optimal controller from the beginning, in sharp contrast to existing results that rely on the PE and rank condition, and can only commit to learning after rich data samples are collected. Simulation results are provided to validate the performance and efficacy of the presented fixed-time RL algorithm.

Lightweight, high-performance networks are important in vision perception systems. Recent research on convolutional neural networks has shown that attention mechanisms can significantly improve the network performance. However, existing approaches either ignore the significance of using both types of attention mechanisms (channel and space) simultaneously or increase the model complexity. In this study, we propose the adaptive attention module (AAM), which is a truly lightweight yet effective module that comprises channel and spatial submodules to balance model performance and complexity. The AAM initially utilizes the channel submodule to generate intermediate channel-refined features. In this module, an adaptive mechanism enables the model to autonomously learn the weights between features extracted by global max pooling and global average pooling to adapt to different stages of the model, thus enhancing performance. The spatial submodule employs a group-interact-aggregate strategy to enhance the expression of important features. It groups the intermediate channel-refined features along the channel dimension into multiple subfeatures for parallel processing and generates spatial attention feature descriptors and channelwise refined subfeatures for each subfeature; subsequently, it aggregates all the refined subfeatures and employs a “channel shuffle” operator to transfer information between different subfeatures, thereby generating the final refined features and adaptively emphasizing important regions. Additionally, AAM is a plug-and-play architectural unit that can be directly used to replace standard convolutions in various convolutional neural networks. Extensive tests on CIFAR-100, ImageNet-1k, BDD100K, and MS COCO demonstrate that AAM improves the baseline network performance under various models and tasks, thereby validating its versatility.

With the rapid development of media, the role of video quality assessment (VQA) is becoming increasingly significant. VQA has applications in many domains. For example, in the field of remote medical diagnosis, it can enhance the quality of video communication between doctors and patients. Besides, in sports broadcasting, it can improve video clarity. Within VQA, the human visual system (HVS) is a crucial component that should be taken into consideration. Considering that attention is guided by goal-driven and top-down factors, such as anticipated locations or some attractive frames within the video, we propose a blind VQA algorithm based on spatial-temporal attention model. Specifically, we first use two pretrained convolutional networks to extract low-level static-dynamic fusion features. Then, a spatial attention-guided model is established to get more representative features of frame-level quality perception. Next, through a temporal attention-guided model, the video-level features are obtained. Finally, the features are fed into a regression model to calculate the final video quality score. The experiments conducted on seven VQA databases reach the state-of-the-art performance, demonstrating the effectiveness of our proposed method.

This article proposes a novel Mahjong game model, LsAc ^∗-MJ, designed to address challenges posed by data scarcity, difficulty in leveraging contextual information, and the computational resource-intensive nature of self-play zero-shot learning. The model is applied to Japanese Mahjong for experiments. LsAc ^∗-MJ employs long short-term memory (LSTM) neural networks, utilizing hidden nodes to store and propagate contextual historical information, thereby enhancing decision accuracy. Additionally, the paper introduces an optimized Advantage Actor-Critic (A2C) algorithm incorporating an experience replay mechanism to enhance the model’s decision-making capabilities and mitigate convergence difficulties arising from strong data correlations. Furthermore, the paper presents a two-stage training approach for self-play deep reinforcement learning models guided by expert knowledge, thereby improving training efficiency. Extensive ablation experiments and performance comparisons demonstrate that, in contrast to other typical deep reinforcement learning models on the RLcard platform, the LsAc ^∗-MJ model consumes lower computational and time resources, has higher training efficiency, faster average decision time, higher win-rate, and stronger decision-making ability.

Computational offloading in mobile edge computing (MEC) systems provides an efficient solution for resource-intensive applications on devices. However, the frequent communication between devices and edge servers increases the traffic within the network, thereby hindering significant improvements in latency. Furthermore, the benefits of MEC cannot be fully realized when the communication link utilized for offloading tasks experiences severe attenuation. Fortunately, reconfigurable intelligent surfaces (RISs) can mitigate propagation-induced impairments by adjusting the phase shifts imposed on the incident signals using their passive reflecting elements. This paper investigates the performance gains achieved by deploying multiple RISs in MEC systems under energy-constrained conditions to minimize the overall system latency. Considering the high coupling among variables such as the selection of multiple RISs, optimization of their phase shifts, transmit power, and MEC offloading volume, the problem is formulated as a nonconvex problem. We propose two approaches to address this problem. First, we employ an alternating optimization approach based on semidefinite relaxation (AO-SDR) to decompose the original problem into two subproblems, enabling the alternating optimization of multi-RIS communication and MEC offloading volume. Second, due to its capability to model and learn the optimal phase adjustment strategies adaptively in dynamic and uncertain environments, deep reinforcement learning (DRL) offers a promising approach to enhance the performance of phase optimization strategies. We leverage DRL to address the joint design of MEC-offloading volume and multi-RIS communication. Extensive simulations and numerical analysis results demonstrate that compared to conventional MEC systems without RIS assistance, the multi-RIS-assisted schemes based on the AO-SDR and DRL methods achieve a reduction in latency by 23.5% and 29.6%, respectively.

In this paper, a novel integrated long short-term memory (LSTM) network and dynamic update model are proposed for long-term object tracking in video images. The LSTM network tracking method is introduced to improve the effect of tracking failure caused by target occlusion. Stable tracking of the target is achieved using the LSTM method to predict the motion trajectory of the target when it is occluded and dynamically updating the tracking template. First, in target tracking, global average peak-to-correlation energy (GAPCE) is used to determine whether the tracking target is blocked or temporarily disappearing such that the follow-up response tracking strategy can be adjusted accordingly. Second, the data with target motion characteristics are utilized to train the designed LSTM model to obtain an offline model, which effectively predicts the motion trajectory during the period when the target is occluded or has disappeared. Therefore, it can be captured again when the target reappears. Finally, in the dynamic template adjustment stage, the historical information of the target movement is combined, and the corresponding value of the current target is compared with the historical response value to realize the dynamic adjustment of the target tracking template. Compared with the current mainstream efficient convolution operators, namely, the E.T.Track, ToMP, KeepTrack, and RTS algorithms, on the OTB100 and LaSOT datasets, the proposed algorithm increases the distance precision by 9.9% when the distance threshold is 5 pixels, increases the overlap success rate by 0.94% when the overlap threshold is 0.75, and decreases the center location error by 18.9%. The proposed method has higher tracking accuracy and robustness and is more suitable for long-term tracking of targets in actual scenarios than are the main approaches.

Brain tumors significantly impact human health due to their complexity and the challenges in early detection and treatment. Accurate diagnosis is crucial for effective intervention, but existing methods often suffer from limitations in accuracy and efficiency. To address these challenges, this study presents a novel deep learning (DL) approach utilizing the EfficientNet family for enhanced brain tumor classification and detection. Leveraging a comprehensive dataset of 3064 T1-weighted CE MRI images, our methodology incorporates advanced preprocessing and augmentation techniques to optimize model performance. The experiments demonstrate that EfficientNetB(07) achieved 99.14%, 98.76%, 99.07%, 99.69%, 99.07%, 98.76%, 98.76%, and 99.07% accuracy, respectively. The pinnacle of our research is the EfficientNetB3 model, which demonstrated exceptional performance with an accuracy rate of 99.69%. This performance surpasses many existing state-of-the-art (SOTA) techniques, underscoring the efficacy of our approach. The precision of our high-accuracy DL model promises to improve diagnostic reliability and speed in clinical settings, facilitating earlier and more effective treatment strategies. Our findings suggest significant potential for improving patient outcomes in brain tumor diagnosis.