56016 research outputs found
Sort by
Optimizing Communication Beamforming for New Multiple Access under Low-Resolution Quantization: A Spectral and Energy Efficiency Perspective
Get PDFDepartment of Electrical EngineeringCurrently, there is growing interest in 6G wireless communication beyond the era of 5G. In addition, the hardware devices require high-speed wireless communication and low-power communications. For example, there are applications such as the internet-of-things (IoT), where devices are limited by battery capacity and have low computing capabilities but require high spectral efficiency. In order to address the issue of power consumption in wireless communication, low-power hardware such as low-resolution analog-to-digital converter (ADC) and digital-to-analog converter (DAC) systems are having attention as a promising transceiver architecture. This is because the power consumption of quantizers decreases exponentially as the number of quantization bits decreases. In this dissertation, low-resolution quantizer system is considered to achieve the trade-off between high spectral efficiency and energy efficiency. Another challenge that needs to be addressed in the development of 6G wireless communications is the severe inter-user interference resulting from the exponential increase in the number of smart devices. For example, in IoT communications, the large number of IoT devices and high channel correlation among them can lead to a significant amount of inter-user interference, which in turn can cause considerable degradation in spectral performance. In this regard, new multiple access approaches are introduced such as rate-splitting multiple access (RSMA), non-orthogonal multiple access (NOMA), spatial-division multiple access (SDMA), and orthogonal multiple access (OMA) to control the interuser interference. Specifically, I consider rate-splitting multiple access to boost the spectral efficiency because rate-splitting multiple access provides extra achievable antenna degree-of freedom by dividing the messages into common and private messages. It is difficult to optimize rate-splitting multiple access precoders due to the minimum rate constraint involved in determining the common rate. Furthermore, the designing quantized precoders is more highly challenging to solve the optimization problem. In this dissertation, I develop a promising RSMA precoder algorithm coupled with quantization errors to maximize the spectral efficiency. To make the optimization problem in smooth function, I first approximate the spectral efficiency of common stream utilizing the Log-Sum Exp technique. Then, I derive the first-order optimality condition in terms of the nonlinear eigenvalue problem (NEP). I suggest computationally efficient method to find a sub-optimal solution for obtaining the principal eigen-vector of the nonlinear eigenvalue problem. In addition, I propose the weighted minimum mean square error-based RSMA precoding algorithm to the considered quantization system. Simulation results demonstrate the performance of the proposed algorithm in terms of the spectral efficiency, and more importantly, ratesplitting multiple access can achieve key benefit than spatial-division multiple access by balancing between the channel gain and quantization error utilizing the common stream in multiuser MIMO systems.clos
????????? ?????? M2 ????????? ??? ZnSe ???????????? ????????? ???????????? ?????? ????????? ????????? ????????? ??????
No full textDepartment of Chemistryclos
Synthesis of dihydroquinolinones based on visible light photocatalysis
No full textDepartment of Chemistryclos
The complete mitochondrial genome of Rana coreana (Anura: Ranidae)
No full textRana coreana is a brown frog species native to the Korean Peninsula. We characterized the complete mitochondrial genome of the species. The mitochondrial genome sequence of R. coreana is 22,262 bp and comprises 13 protein-coding genes, two ribosomal RNA (rRNA) genes, 22 transfer RNA (tRNA) genes, and two control regions (CRs). The CR duplication and gene organization were identical to those observed in Rana kunyuensis and Rana amurensis. A total of 13 protein-coding genes were used to examine the phylogenetic relationships between this species and the genus Rana. R. coreana living on the Korean Peninsula, formed a cluster with R. kunyuensis and R. amurensis, with R. coreana showing the closest phylogenetic affinity for R. kunyuensis
Thrap3 promotes non-alcoholic fatty liver disease by suppressing AMPK-mediated autophagy
No full textMMP Net: A feedforward neural network model with sequential inputs for representing continuous multistage manufacturing processes without intermediate outputs
No full textMachine learning models that are used for the prediction and control of production can improve quality and yield. However, developing models that are highly accurate and reflective of real-world processes is challenging. We propose a feedforward neural network model specifically designed for continuous Multistage Manufacturing Processes (MMPs) without intermediate outputs. This model, which is termed "MMP Net," can accurately represent the control mechanism of continuous MMPs. Whereas existing studies on learning MMPs assume an intermediate output data, the MMP Net does not require such an unrealistic assumption. We use the MMP Net to develop prediction models for the lubricant base oil production process of a world-leading lubricant manufacturer. Evaluation results show that the MMP Net is superior to other deep neural network and machine learning models. Consequently, the MMP Net was actually implemented in a real factory in 2022 and is expected to save 900,000 dollars per year for each production line. We believe that our work can serve as a basis to develop customized machine learning solutions for improving continuous MMPs
Defining the zerogap: cracking along the photolithographically defined Au-Cu-Au lines with sub-nanometer precision
No full textCracks are formed along the photolithographically pre-determined lines with extremely high yield and repeatability, when Cu clusters are introduced between planarized Au thin films sequentially deposited on a PET substrate. These clusters act as nanometer-sized spacers preventing the formation of contiguous metallic bond between the adjacent Au layers which will render prepatterned-cracking impossible. While the effective gap width is initially zero in the optical sense from microwaves all the way to the visible, outer-bending the PET substrate allows the gap width tuning into the 100 nm range, with the stability and controllability in the ranges of 100 s and Angstrom-scale, respectively. It is anticipated that our wafer-scale prepatterned crack technology with an unprecedented mixture of macroscopic length and Angstrom-scale controllability will open-up many applications in optoelectronics, quantum photonics and photocatalysis
Enhanced terahertz nonlinear response of GaAs by the tight field confinement in a nanogap
No full textWe demonstrated that an incident terahertz peak field amplitude below 0.01 MV/cm can trigger Zener tunneling in a semi-insulating GaAs. Moreover, a transmission decrease with an extinction ratio of 60% was observed in the semi-insulating GaAs with an electric field strength of up to 46 MV/cm (maximum incident peak field of similar to 0.29 MV/cm). These experimental results were realized by taking advantage of the nonlinear effects, such as Zener tunneling, impact ionization, and metal-insulator-metal tunneling in 5 nm metallic nanogaps on the GaAs; a strong field was locally confined in the vicinity of these gaps. The 5 nm gap enabled us to lower the voltage across the gap to suppress impact ionization while allowing Zener tunneling. Simulation results indicated that the effective thickness of the semiconductor increased as a function of the gap size. The approach used in this study decreases the threshold incident electric field for nonlinear responses as well as paves the way toward ultrathin high-speed electronic devices and ultrafast light pumps
