Modal split prediction in transportation networks has the potential to support network operators in managing traffic congestion and improving transit service reliability. We focus on the problem of hourly prediction o...
详细信息
The gradual decommissioning of fossil fuel-driven power plants, that traditionally provide most operational flexibility in power systems, has led to more frequent grid stability issues. To compensate for the lack of f...
详细信息
In time division duplexing (TDD) millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) systems, downlink channel state information (CSI) can be obtained from uplink channel estimation thanks to channe...
详细信息
The ever increasing need for performance results in increasingly rigorous demands on throughput and positioning accuracy of high-precision motion systems, which often suffer from position dependent effects that origin...
详细信息
This paper aims to forecast high-resolution (hourly) aggregated load for a certain region in the medium term (a few days to over a year). One region is defined as some places with similar climate characteristics becau...
This paper aims to forecast high-resolution (hourly) aggregated load for a certain region in the medium term (a few days to over a year). One region is defined as some places with similar climate characteristics because the climate influences people’s daily lifestyles and hence the electric usage. We decompose the electric usage records into two parts: base load and seasonal load. Considering both temperature and time factors, different deep learning methods are adopted to characterize them. The first goal of our approach is to predict the peak load which is critical for power system planning. Furthermore, our proposed forecast method can provide the depiction of the hourly load profile to provide customized load curves for high-level real-time applications. The proposed method is tested on real-world historical data collected by CAISO, BPA, and PACW. The experimental results show that trained by three years of data, our method could reduce the prediction error for one-year lead hourly load below 5% MAPE, and predict the occurrence of the peak load for next year in CAISO with an error within three days. Furthermore, as a byproduct, an interesting observation on the impact of COVID-19 on human life was made and discussed based on these case studies.
Metamaterials have proven their ability to possess extraordinary physical properties distinct from naturally available materials,leading to exciting sensing functionalities and ***,metamaterial-based sensing applicati...
详细信息
Metamaterials have proven their ability to possess extraordinary physical properties distinct from naturally available materials,leading to exciting sensing functionalities and ***,metamaterial-based sensing applications suffer from severe performance limitations due to noise interference and design ***,we propose a dual-phase strategy that leverages loss-induced different Fano-resonant phases to access both destructive and constructive signals of molecular *** the two reverse signals are innovatively combined,the noise in the detection system is effectively suppressed,thereby breaking through the noise-related ***,by utilizing loss optimization of the plasmon-molecule coupling system,our dual-phase strategy enhances the efficiency of infrared energy transfer into the molecule without any additional fabrication complex,thereby overcoming the trade-off dilemma between performance and fabrication *** to the pioneering breakthroughs in the limitations,our dual-phase strategy possesses an overwhelming competitive advantage in ultrasensitive vibrational spectroscopy over traditional metamaterial technology,including strong signal strength(×4),high sensitivity(×4.2),effective noise suppression(30%),low detection limit(13 ppm),and excellent selectivity among CO_(2),NH_(3),and CH_(4) *** work not only opens the door to various emerging ultrasensitive detection applications,including ultrasensitive in-breath diagnostics and high-information analysis of molecular information in dynamic reactions,but also gains new insights into the plasmon-molecule interactions in advanced metamaterials.
In this paper, we propose an information geometry approach (IGA) for signal detection in ultra-massive multiple-input multiple-output (MIMO) systems. We formulate the signal detection as obtaining the marginals of the...
In this paper, we propose an information geometry approach (IGA) for signal detection in ultra-massive multiple-input multiple-output (MIMO) systems. We formulate the signal detection as obtaining the marginals of the a posteriori probability distribution of the transmitted symbol vector. Then, a maximization of the a posteriori marginals (MPM) for signal detection can be performed. With the information geometry theory, we calculate the approximations of the a posteriori marginals. It is formulated as an iterative m-projection process. We then apply the central-limit-theorem (CLT) to simplify the calculation of the m-projection since the direct calculation of the m-projection is of exponential-complexity. With the CLT, we obtain an approximate solution of the m-projection, which is asymptotically accurate. Simulation results demonstrate that the proposed IGA is a promising and efficient method to implement the signal detector in ultra-massive MIMO systems.
Artificial intelligence generated content (AIGC) technologies, with a predominance of large language models (LLMs), have demonstrated remarkable performance improvements in various applications, which have attracted g...
详细信息
This paper focuses on the application of Conformers in speaker verification. Conformers, initially designed for Automatic Speech Recognition (ASR), excel at modeling both local and global contexts within speech signal...
详细信息
This paper focuses on the application of Conformers in speaker verification. Conformers, initially designed for Automatic Speech Recognition (ASR), excel at modeling both local and global contexts within speech signals effectively. Previous research has established that ASR and speaker verification tasks can naturally complement each other. Building on this synergistic relationship, this study introduces three strategies for leveraging ASR-pretrained Conformers in speaker verification: (1) Transfer learning: We use a pretrained ASR Conformer encoder to initialize the speaker embedding network, thereby enhancing model generalization and mitigating the risk of overfitting. (2) Knowledge distillation: We distill the complex capabilities of an ASR Conformer into a speaker verification model. This not only allows for flexibility in the student mode's network architecture but also incorporates frame-level ASR distillation loss as an auxiliary task to reinforce speaker verification. (3) Parameter-efficient transfer learning with speaker adaptation: A lightweight speaker adaptation module is proposed to convert ASR-derived features into speaker-specific embeddings, without altering the core architecture of the original ASR Conformer. This strategy facilitates the concurrent execution of ASR and speaker verification tasks within a singular model. Experiments were conducted on VoxCeleb datasets. The results are compelling: models employing ASR pretraining and knowledge distillation significantly outperform standard Conformers. Specifically, the best model using the ASR pretraining method achieved a 0.43% equal error rate (EER) on the VoxCeleb1-O test trial, while the knowledge distillation approach yielded a 0.38% EER. Furthermore, by adding a mere 4.92 million parameters to a 130.94 million-parameter ASR Conformer encoder, the speaker adaptation approach achieved a 0.45% EER, enabling parallel speech recognition and speaker verification within a single ASR Conformer encoder.
The purpose of this letter is to study the design and explore vertically stacked complementary tunneling field-effect transistors (CTFETs) using CFET technology for emerging technology nodes. As a prior work, the CTFE...
详细信息
暂无评论