{"title":"Pursuit-Evasion Game of Unmanded Surface Vehicles Based on Deep Reinforcement Learning","authors":"Xin Wang, Yue-ying Wang, Weixiang Zhou, Jiaming Zhang","doi":"10.1109/ICECAI58670.2023.10176487","DOIUrl":"https://doi.org/10.1109/ICECAI58670.2023.10176487","url":null,"abstract":"Making decisions during the pursuit-evasion game of unmanned surface vehicles (USVs) in a restricted environment with obstacles is a challenging problem. Specifically, in the pursuit game, the pursuer needs to consider how to approach the evader quickly and how to surround the evader and safely avoid obstacles in an environment containing obstacles. This paper proposes a distributed algorithm based on deep reinforcement learning to help USV solve the pursuit problem in a restricted environment. The proposed algorithm can deal with the game problem of multiple USVs on the ocean’s surface. In particular, composite reward function with guiding characteristics are designed based on the artificial potential field (APF) method for pursuit, encirclement, and obstacle avoidance, which can help the USV improve pursuit performance. Then, curriculum learning is used to help the USV improve learning efficiency in the early stage. The simulation results show that the algorithm is effective in different initial conditions and performs well.","PeriodicalId":189631,"journal":{"name":"2023 4th International Conference on Electronic Communication and Artificial Intelligence (ICECAI)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125746826","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The Function of Crop pest and disease identification based on the improved ConvNeXt Network model","authors":"Hailang Chen, Yuan-Hau Liao","doi":"10.1109/ICECAI58670.2023.10176546","DOIUrl":"https://doi.org/10.1109/ICECAI58670.2023.10176546","url":null,"abstract":"A ConvNeXt-ECA network model based on the improved ConvNeXt network is designed for the problem that the pest identification method based on traditional machine learning and simple neural network is not ideal for crop pest identification with unbalanced data, unclear features and multiple categories. The model is based on the ConvNeXt network model as the backbone network, using migration learning to achieve the sharing of pre-trained weight parameters and combining with the attention mechanism to enhance the feature fusion capability. Finally, 16 plant species with a total of 27 types of diseases were studied with a total of 34,200 images, and the training set, validation set and test set were divided in the ratio of 6:2:2. The model converged significantly faster during the experimental training compared with VGG19, GoogLeNet, ResNet50 and ConvNeXt networks. The experimental results show that the accuracy of the improved ConvNeXt network-based crop pest and disease intelligent recognition model reaches 96.73%, while the ablation experiments demonstrate the improvement of the model performance by the optimization method proposed in this paper.","PeriodicalId":189631,"journal":{"name":"2023 4th International Conference on Electronic Communication and Artificial Intelligence (ICECAI)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127988427","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xinghu Jin, Shaopei Ji, Yanhong Liu, Weibing Zhu, Liang Jin, Xinyu Ming
{"title":"Multicarrier NOMA Power Allocation Strategy Based on Adaptive Particle Swarm Optimization Algorithm","authors":"Xinghu Jin, Shaopei Ji, Yanhong Liu, Weibing Zhu, Liang Jin, Xinyu Ming","doi":"10.1109/ICECAI58670.2023.10176543","DOIUrl":"https://doi.org/10.1109/ICECAI58670.2023.10176543","url":null,"abstract":"As non-orthogonal multiple access technology can achieve higher system throughput, spectrum efficiency and energy efficiency than traditional orthogonal access technology, it has become a research hotspot of 5G multiple access technology. In this paper, a power allocation strategy based on adaptive particle swarm optimization is proposed to optimize the energy efficiency of NOMA downlink system. Firstly, aiming at the defects of conventional particle swarm optimization algorithm, an adaptive particle swarm optimization algorithm based on diversity driven (PDD-APSO) is proposed; Then an optimization model based on energy efficiency maximization is established, and PDD-APSO algorithm is used to solve the objective function of maximizing system energy efficiency. The research results show that the PDD-APSO algorithm improves the convergence speed and precision, and the global search ability is significantly improved compared with other existing PSO algorithms; At the optimal power distribution point, the PDD-APSO algorithm can significantly improve the system energy efficiency.","PeriodicalId":189631,"journal":{"name":"2023 4th International Conference on Electronic Communication and Artificial Intelligence (ICECAI)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134109249","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zhixun Zhang, Rengang Li, Yiming Wang, Yuandong Shi
{"title":"Research on single target jamming parameter optimization based on particle swarm optimization","authors":"Zhixun Zhang, Rengang Li, Yiming Wang, Yuandong Shi","doi":"10.1109/ICECAI58670.2023.10176864","DOIUrl":"https://doi.org/10.1109/ICECAI58670.2023.10176864","url":null,"abstract":"In order to achieve effective jamming, it is often necessary to select jamming patterns according to expert experience in radar electronic countermeasures. However, different jamming patterns correspond to a large number of variable parameters a large range of parameters, a large space of inverted parameters, and it is difficult to set the optimal jamming parameters. The application of particle swarm optimization in single target jamming parameter optimization is studied in this work. Construct jamming income function according to jamming incentives of different jamming patterns. The particle swarm optimization algorithm is used to optimize the jamming parameters of different jamming styles and output the optimal parameters and corresponding jamming signals. The research shows that the algorithm model can be quickly optimized in multi-dimensional apace by designing a reasonable jamming revenue function, which can realize the rapid the threat jamming target, and can significantly improve the adaptive jamming waveform generation ability of the jamming to various types of radars.","PeriodicalId":189631,"journal":{"name":"2023 4th International Conference on Electronic Communication and Artificial Intelligence (ICECAI)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132401563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Intermediate pressure analysis of cascade heat pump system based on characteristic variable selection","authors":"Dongfang Yang, Zhixin Li, Weiqing Zhou, Jilong Zhang","doi":"10.1109/ICECAI58670.2023.10176726","DOIUrl":"https://doi.org/10.1109/ICECAI58670.2023.10176726","url":null,"abstract":"It is crucial to the effective and steady functioning of a heat pump system that an appropriate intermediate pressure be chosen for a compound heat pump. In this research, we use a method of data mining called self-organizing data mining to conduct an analysis of the primary characteristic factors that influence the choice of the system’s intermediate pressure so that it can maintain stable operation. The variables that affect the intermediate pressure are derived through an analysis of the stable operation data of the three devices. The results of the analysis, along with the results based on the mechanism calculation, can provide a theoretical basis for the subsequent system data modeling through the determination of the characteristic variables.","PeriodicalId":189631,"journal":{"name":"2023 4th International Conference on Electronic Communication and Artificial Intelligence (ICECAI)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130260812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Density Peak Clustering Based on Global Density","authors":"Min Li, William Zhu","doi":"10.1109/ICECAI58670.2023.10176471","DOIUrl":"https://doi.org/10.1109/ICECAI58670.2023.10176471","url":null,"abstract":"Density peaks clustering is an efficient clustering method. It regards density peaks determined by local density which depends on cutoff distance as cluster centers to recognize clusters. However, it is difficult to find the appropriate cutoff distance to accurately measure local density. In this paper, we define global density to solve above problem. In order to define the global density of a point, we define the density between any two points. The global density of a point is defined as the sum of the densities from a point to other points. We apply our global density to the density peaks clustering and then propose a new clustering method. Experiments illustrate the effectiveness of our new method.","PeriodicalId":189631,"journal":{"name":"2023 4th International Conference on Electronic Communication and Artificial Intelligence (ICECAI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128830009","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"CIPTA: Contrastive-based Iterative Prompt-tuning Using Text Annotation from Large Language Models","authors":"Yuan Yan, Wenzhuo Du, Di Yang, Dechun Yin","doi":"10.1109/ICECAI58670.2023.10176586","DOIUrl":"https://doi.org/10.1109/ICECAI58670.2023.10176586","url":null,"abstract":"In recent years, public opinion analysis has become increasingly important due to the widespread use of social media platforms and the growing influence of online information on public security. Prompt tuning, a typical few-shot learning method, ensures that the model quickly adapts to opinion analysis with different classification rules. However, existing prompt tuning for opinion analysis cannot guarantee the effectiveness of the model in zero-shot or one-shot cases. In this study, we propose the Contrastive-based Iterative Prompt-tuning method using Text-Annotation from Large Language Models (LLMs), CIPTA, for low-resource public opinion analysis. Specifically, with a small amount of manually labeled data, CIPTA leverages the knowledge from LLMs to text annotation and utilizes unsupervised contrastive embedding training to optimize text representation. Based on the prompt tuning method and the iterative training over unlabeled data, the model further utilizes the knowledge from the pre-training stage. Experiment results on tweet data show that our CIPTA achieves encouraging performance in public opinion analysis.","PeriodicalId":189631,"journal":{"name":"2023 4th International Conference on Electronic Communication and Artificial Intelligence (ICECAI)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124495338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Evaluation criterion of space surveillance sensor scheduling based on relative orbit analysis","authors":"Jiakang Shen, Junling Wang, Mengqi Zhao, Peng Lv","doi":"10.1109/ICECAI58670.2023.10176684","DOIUrl":"https://doi.org/10.1109/ICECAI58670.2023.10176684","url":null,"abstract":"The rapid growth of the number of space targets leads to the shortage of sensor observation resources in the space surveillance network. Improving the efficiency of sensor scheduling helps to alleviate the observation pressure of the space surveillance network. In this paper, a unified observation model of space-based sensors and ground-based sensors for space targets is established based on the relative orbit of space targets. Based on the covariance matrix, a criterion function for calculating the scheduling efficiency of space surveillance network sensors is proposed, and the analytical solution of the scheduling criterion function under the assumption of plane circular orbit is given. The analytical solution can be used to analyze the influence of observation geometry, observation accuracy, observation times and other factors on scheduling efficiency. Finally, the influence of observation geometry on radar sensor scheduling efficiency is simulated and analyzed. The simulation results show that the orbit uncertainty decreases with the increase of the observation geometric angle under the same observation resources, therefore the radar sensor scheduling efficiency increases with the increase of the observation geometric angle. This conclusion is applicable to both space-based and ground-based radar sensors.","PeriodicalId":189631,"journal":{"name":"2023 4th International Conference on Electronic Communication and Artificial Intelligence (ICECAI)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124552702","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zihan Zhou, D. Zhai, Wenxin Tang, Caina Qin, Zhangjie Cai
{"title":"Access Latency Minimization for Aerial Data Collection Networks with NOMA","authors":"Zihan Zhou, D. Zhai, Wenxin Tang, Caina Qin, Zhangjie Cai","doi":"10.1109/ICECAI58670.2023.10176633","DOIUrl":"https://doi.org/10.1109/ICECAI58670.2023.10176633","url":null,"abstract":"optimizing the access and transmission of massive devices in the uplink wireless network is crucial for the sixth-generation mobile communication system (6G). In this paper, we present a novel optimization approach based on non-orthogonal multiple access (NOMA) which consist of a large number of Internet of Things (IoT) devices. To minimize the access latency of devices, we propose a joint optimization scheme including unmanned aerial vehicle (UAV) trajectory planning, device scheduling, and power control. For UAV trajectory planning, we discretize the space and propose the Minimum Set Cover (MSC) based algorithm to minimize the UAV hovering positions. For user scheduling, we recast it as a k-cut problem, build a directional interference graph to model interference between devices, and establish a viable scheduling scheme. Moreover, we optimize the power allocation by fixed-point equations. Finally, simulation results verify that the proposed scheme can accomplish massive device access and quick data collection.","PeriodicalId":189631,"journal":{"name":"2023 4th International Conference on Electronic Communication and Artificial Intelligence (ICECAI)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133936695","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Design and Development of Emotional Analysis System for Chinese Online Comment Text","authors":"J. Yao, Q. Hu, Tianyi Zhou, Yilin Wang","doi":"10.1109/ICECAI58670.2023.10176889","DOIUrl":"https://doi.org/10.1109/ICECAI58670.2023.10176889","url":null,"abstract":"Performing sentiment analysis on a massive volume of online comments data has significant commercial value. Therefore, this paper proposes a sentiment analysis platform focusing on Chinese comment text. In system design, we utilized deep learning techniques and implemented a large-scale pre-trained language model ERNIE for text feature extraction. We fine-tuned the network structure targeting downstream tasks to achieve Chinese sentiment analysis, overcoming the challenges of traditional machine learning methods, such as low accuracy and scalability. Additionally, we compared the experimental results with prominent pre-trained models, analyzed and evaluated the experimental data. The experimental results indicate that the platform can enhance the data analytic capabilities of online Chinese comment text, ameliorate the communication efficiency between industry audiences and creators, and extract commercial values from Internet comments.","PeriodicalId":189631,"journal":{"name":"2023 4th International Conference on Electronic Communication and Artificial Intelligence (ICECAI)","volume":"2 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125717245","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}