Since data services are penetrating into our daily life rapidly, the mobile network becomes more complicated, and the amount of data transmission is more and more increasing. In this case, the traditional statistical ...Since data services are penetrating into our daily life rapidly, the mobile network becomes more complicated, and the amount of data transmission is more and more increasing. In this case, the traditional statistical methods for anomalous cell detection cannot adapt to the evolution of networks, and data mining becomes the mainstream. In this paper, we propose a novel kernel density-based local outlier factor(KLOF) to assign a degree of being an outlier to each object. Firstly, the notion of KLOF is introduced, which captures exactly the relative degree of isolation. Then, by analyzing its properties, including the tightness of upper and lower bounds, sensitivity of density perturbation, we find that KLOF is much greater than 1 for outliers. Lastly, KLOFis applied on a real-world dataset to detect anomalous cells with abnormal key performance indicators(KPIs) to verify its reliability. The experiment shows that KLOF can find outliers efficiently. It can be a guideline for the operators to perform faster and more efficient trouble shooting.展开更多
1 Introduction Existing university research evaluation mechanisms face issues like unclear value orientation,poor classification,suboptimal expert selection,and unscientific organization[1].A comprehensive,scientific ...1 Introduction Existing university research evaluation mechanisms face issues like unclear value orientation,poor classification,suboptimal expert selection,and unscientific organization[1].A comprehensive,scientific evaluation system is essential for advancing higher education.Motivation.Traditional evaluation methods struggle to keep up with the increasing complexity and variety of evaluation indicators[2].Data mining techniques,such as cooccurrence analysis and clustering,have been effective in uncovering hidden patterns in large datasets,addressing the limitations of traditional qualitative methods[3−5].However,research applying these techniques in this field remains limited.展开更多
The rising demand for diverse service requirements under limited resources necessitates designing codeword indicators with variable distance metrics to exploit prior information,such as data rate,coding schemes,and Si...The rising demand for diverse service requirements under limited resources necessitates designing codeword indicators with variable distance metrics to exploit prior information,such as data rate,coding schemes,and Signal-to-Noise Ratio(SNR).Traditional codebook designs that maximize the minimum Hamming distance provide a global lower bound but often overlook the specific distance requirements of individual codewords.To address this limitation,the weighted Hamming distance metric is introduced,prioritizing codewords with varying distance attributes.Maximizing the weighted Hamming distance leads to the development of hierarchical codebooks,enabling the selection of codewords tailored to specific distance needs.Building on the classical Plotkin bound,an upper bound is derived,revealing trade-offs among the number of codewords,distance layers,and lengths.Optimal hierarchical codebooks are designed using a cyclotomy structure to maintain the desired Hamming distance properties and incorporate a heuristic algorithm to allow flexibility in codebook length and multi-layer attributes.In ultra-wideband multi-service scenarios,allocating higher-distance codewords to lower-SNR services enhances overall performance by matching codeword distances to service demands.Simulations demonstrate that,informed by deep learning-based SNR predictions,this approach reduces packet error rates and improves efficiency compared to existing IEEE 802.15.4z convolutional schemes and classical Hadamard codebooks.展开更多
基金supported by the National Basic Research Program of China (973 Program: 2013CB329004)
文摘Since data services are penetrating into our daily life rapidly, the mobile network becomes more complicated, and the amount of data transmission is more and more increasing. In this case, the traditional statistical methods for anomalous cell detection cannot adapt to the evolution of networks, and data mining becomes the mainstream. In this paper, we propose a novel kernel density-based local outlier factor(KLOF) to assign a degree of being an outlier to each object. Firstly, the notion of KLOF is introduced, which captures exactly the relative degree of isolation. Then, by analyzing its properties, including the tightness of upper and lower bounds, sensitivity of density perturbation, we find that KLOF is much greater than 1 for outliers. Lastly, KLOFis applied on a real-world dataset to detect anomalous cells with abnormal key performance indicators(KPIs) to verify its reliability. The experiment shows that KLOF can find outliers efficiently. It can be a guideline for the operators to perform faster and more efficient trouble shooting.
基金supported by the 2022 Theoretical Research Project on Talent Work of the China Association for Science and Technology(CAST)titled“Research on Talent Discovery and Utilization Mechanisms for Key Core Technical Personnel:A Case Study in Aerospace,Integrated Circuits,and Related Fields”(2022070607CG0905012022).
文摘1 Introduction Existing university research evaluation mechanisms face issues like unclear value orientation,poor classification,suboptimal expert selection,and unscientific organization[1].A comprehensive,scientific evaluation system is essential for advancing higher education.Motivation.Traditional evaluation methods struggle to keep up with the increasing complexity and variety of evaluation indicators[2].Data mining techniques,such as cooccurrence analysis and clustering,have been effective in uncovering hidden patterns in large datasets,addressing the limitations of traditional qualitative methods[3−5].However,research applying these techniques in this field remains limited.
基金supported by the National Natural Science Foundation of China(No.62025110)the Beijing Natural Science Foundation(No.4254068)the China Postdoctoral Science Foundation(Nos.GZC20231378 and 2024M751682).
文摘The rising demand for diverse service requirements under limited resources necessitates designing codeword indicators with variable distance metrics to exploit prior information,such as data rate,coding schemes,and Signal-to-Noise Ratio(SNR).Traditional codebook designs that maximize the minimum Hamming distance provide a global lower bound but often overlook the specific distance requirements of individual codewords.To address this limitation,the weighted Hamming distance metric is introduced,prioritizing codewords with varying distance attributes.Maximizing the weighted Hamming distance leads to the development of hierarchical codebooks,enabling the selection of codewords tailored to specific distance needs.Building on the classical Plotkin bound,an upper bound is derived,revealing trade-offs among the number of codewords,distance layers,and lengths.Optimal hierarchical codebooks are designed using a cyclotomy structure to maintain the desired Hamming distance properties and incorporate a heuristic algorithm to allow flexibility in codebook length and multi-layer attributes.In ultra-wideband multi-service scenarios,allocating higher-distance codewords to lower-SNR services enhances overall performance by matching codeword distances to service demands.Simulations demonstrate that,informed by deep learning-based SNR predictions,this approach reduces packet error rates and improves efficiency compared to existing IEEE 802.15.4z convolutional schemes and classical Hadamard codebooks.
