This paper studies near-field (NF) channel estimation (CE) for extremely-large hybrid reconfigurable intelligent surface (RIS)-aided systems. To be specific, we consider uplink training, where the mobile station sends...
详细信息
ISBN:
(数字)9798350393187
ISBN:
(纸本)9798350393194
This paper studies near-field (NF) channel estimation (CE) for extremely-large hybrid reconfigurable intelligent surface (RIS)-aided systems. To be specific, we consider uplink training, where the mobile station sends pilot signals to the base station and the hybrid RIS, which is equipped with a single radio frequency chain for sensing purposes. We use the linear total variation regularization (TVR) algorithm to leverage the inherent structured sparsity of NF channels within the angular domain. We further evaluate the impact of carrier frequency, power-splitting factor, and hybrid RIS array configurations on the normalized mean square error (NMSE) performance of the TVR algorithm. Numerical results indicate that the TVR algorithm exhibits superior NMSE performance particularly at higher frequencies, such as 28 GHz. In addition, we observe that the power-splitting factor plays a pivotal role in the NMSE performance.
Tensor decomposition has emerged as an essential means for applications with multidimensional signals such as data compression and feature extraction. However, due to the enormous amount of signalprocessing and stora...
详细信息
ISBN:
(数字)9798350330991
ISBN:
(纸本)9798350331004
Tensor decomposition has emerged as an essential means for applications with multidimensional signals such as data compression and feature extraction. However, due to the enormous amount of signalprocessing and storage, it is challenging to design a low-latency tensor decomposition processor with high hardware efficiency. This paper presents a low-latency algorithm and VLSI architecture for a tensor decomposition processor. The designed circuit is implemented based on the FPGA platform. The estimation results demonstrate that the proposed architecture achieves a low-latency performance and high hardware efficiency.
The recursive least squares (RLS) algorithm is widely used in various adaptive filtering applications, mostly due to its rapid convergence rate. The forgetting factor is a crucial parameter in this algorithm. For a fi...
详细信息
ISBN:
(数字)9798350309249
ISBN:
(纸本)9798350309256
The recursive least squares (RLS) algorithm is widely used in various adaptive filtering applications, mostly due to its rapid convergence rate. The forgetting factor is a crucial parameter in this algorithm. For a fixed value for this parameter, there is a tradeoff between misalignment and tracking. Nevertheless, the use of a variable forgetting factor (VFF) approach gives a better compromise between the performance criteria of the RLS algorithm. Nonetheless, the algorithm's computational complexity presents a notable challenge, particularly in scenarios that involve lengthy adaptive filters, such as acoustic echo cancellation (AEC). This paper incorporates the concept of a variable forgetting factor into the fast normalized least mean squares (FNLMS) algorithm. The FNLMS algorithm demonstrates performances comparable to the RLS algorithm while maintaining a computational complexity similar to that of the normalized least mean squares (NLMS) algorithm. Based on simulation results in the context of AEC, the proposed VFF-FNLMS algorithm exhibits better performances in terms of convergence speed and tracking ability when compared to both FNLMS and NLMS algorithms.
Constraining the amount of CO 2 in the early Martian atmosphere is critical to understanding Mars' geological history and aqueous evolution. We propose an improved and robust algorithm for mapping carbonates from ...
详细信息
ISBN:
(数字)9798331513139
ISBN:
(纸本)9798331513146
Constraining the amount of CO 2 in the early Martian atmosphere is critical to understanding Mars' geological history and aqueous evolution. We propose an improved and robust algorithm for mapping carbonates from CRISM data which involves comprehensive preprocessing to suppress noise, computing multiple spectral parameters, and intelligently combining them using their harmonic mean. Demonstrated over the CRISM image HRL40FF, our innovative approach yields fewer false positives as compared to traditional methods and can also be effectively applied to other hyperspectral datasets. This enhanced method advances carbonate mapping on Mars, thereby contributing to a deeper understanding of its geological and aqueous history.
This work studies the problem of secrecy energy efficiency maximization in multi-user wireless networks aided by reconfigurable intelligent surfaces, in which an eavesdropper overhears the uplink communication. A prov...
详细信息
ISBN:
(数字)9798350393187
ISBN:
(纸本)9798350393194
This work studies the problem of secrecy energy efficiency maximization in multi-user wireless networks aided by reconfigurable intelligent surfaces, in which an eavesdropper overhears the uplink communication. A provably convergent optimization algorithm is proposed which optimizes the user’s transmit power, metasurface reflection coefficients, and base station receive filters. The complexity of the proposed method is analyzed and numerical results are provided to show the performance of the proposed optimization method.
Neural ordinary differential equations (Neural ODEs) propose the idea that a sequence of layers in a neural network is just a discretisation of an ODE, and thus can instead be directly modelled by a parameterised ODE....
详细信息
ISBN:
(数字)9798350372250
ISBN:
(纸本)9798350372267
Neural ordinary differential equations (Neural ODEs) propose the idea that a sequence of layers in a neural network is just a discretisation of an ODE, and thus can instead be directly modelled by a parameterised ODE. This idea has had resounding success in the deep learning literature, with direct or indirect influence in many state of t he art ideas, such as diffusion models or time dependant models. Recently, a continuous version of the U-net architecture has been proposed, showing increased performance over its discrete counterpart in many imaging applications and wrapped with theoretical guarantees around its performance and robustness. In this work, we explore the use of Neural ODEs for learned inverse problems, in particular with the well-known Learned Primal Dual algorithm, and apply it to computed tomography (CT) reconstruction.
Heartbeat rate estimation using Convolutional Neural Network presents an innovative and non-invasive solution for cardiovascular monitoring. This approach harnesses computer vision and signalprocessing techniques to ...
详细信息
ISBN:
(数字)9798350359299
ISBN:
(纸本)9798350359305
Heartbeat rate estimation using Convolutional Neural Network presents an innovative and non-invasive solution for cardiovascular monitoring. This approach harnesses computer vision and signalprocessing techniques to extract valuable physiological information from facial video data captured by a standard web camera. By analyzing subtle color variations associated with blood flow, the system provides continuous and instantaneous estimations of the user’s heartbeat rate. The proposed method involves the application of advanced signal processing algorithms, specifically utilizing photoplethysmography (PPG) principles, to the facial video stream. These algorithms track changes in pixel intensity over time, correlating with variations in blood volume during each cardiac cycle. To enhance accuracy and adaptability, machine learning models are employed, allowing the system to account for individual differences in skin tones, lighting conditions, and facial features. The result is a real-time analysis that enables continuous and unobtrusive monitoring of the user’s heart rate. A notable advantage of this approach lies in its accessibility and cost-effectiveness, as it leverages widely available web cameras without requiring specialized hardware. The proposed system’s robustness is evaluated across diverse scenarios, including varying lighting conditions and demographic groups, ensuring its reliability in real-world applications. This real-time heartbeat rate estimation using a web camera holds promise for a range of applications, from personalized health tracking to telehealth, marking a significant advancement in accessible cardiovascular monitoring
The subway has become an essential mode of urban transportation due to its high efficiency and environmental benefits. However, as the subway's operation time increases, the train's wheel sets may experience a...
详细信息
ISBN:
(数字)9798350376548
ISBN:
(纸本)9798350376555
The subway has become an essential mode of urban transportation due to its high efficiency and environmental benefits. However, as the subway's operation time increases, the train's wheel sets may experience abnormalities, such as tread wear and coaxial wheel diameter differences exceeding the limit. These abnormalities can worsen the stability and safety of subway train operations. However, the current scheduled maintenance plan for subways is inadequate for timely maintenance of faulty trains. This paper proposes an algorithm that uses optical fiber sensing to identify abnormal subway train wheelset status. The algorithm analyses the train signals collected by the optical fiber sensing system, which can reflect the status of subway train wheelsets. Additionally, the principal component analysis method is used to analyze the train signal matrix. Dimensionality reduction was used to identify abnormal train signals through T2 statistics. Subsequently, subway trains with wheel set abnormalities were identified. Upon checking the original train signal of the fault, abnormal high-frequency fluctuations were observed, and the detection results were accurate.
To minimize the impact of noise on speech signals during transmission or recording. In this paper, a denoising method for speech signals is introduced, which integrates GWO-VMD with MODWT. Firstly, the combination of ...
详细信息
ISBN:
(数字)9798331516550
ISBN:
(纸本)9798331516567
To minimize the impact of noise on speech signals during transmission or recording. In this paper, a denoising method for speech signals is introduced, which integrates GWO-VMD with MODWT. Firstly, the combination of VMD decomposition parameters (k, α) is optimized using the GWO algorithm. The optimized VMD is used to decompose the signal and the IMF component with lower correlation coefficient is eliminated according to the correlation coefficient screening principle. Secondly, the remaining IMFs are recombined and multi-scale analysis is performed by MODWT. Finally, the wavelet coefficients with relative energy greater than a certain threshold are reconstructed to derive the denoised signal. Experimental outcomes demonstrate that by using GWO-VMD-MODWT for noise reduction, the signal-to-noise ratio (SNR) of the original noisy speech signal has increased from -1.0416dB to 11.6057dB. The root mean square error (RMSE) has decreased from 0.0998 to 0.0233. The noise reduction capability of the integrated denoising method presented in this paper surpasses that of the single VMD and MODWT denoising method, as well as the classical EMD and EWT denoising method.
Video analytics is considered the killer application of edge computing and has been successfully deployed across diverse domains. Yet, executing video analytics on mobile devices presents notable challenges owing to t...
详细信息
ISBN:
(数字)9798350317152
ISBN:
(纸本)9798350317169
Video analytics is considered the killer application of edge computing and has been successfully deployed across diverse domains. Yet, executing video analytics on mobile devices presents notable challenges owing to the considerable computational demands and frame rate requirements of DNN models. Current mobile inference frameworks often concentrate on enhancing model inference performance on the CPU or GPU, overlooking the potential of the Digital signal Processor (DSP) – an emerging heterogeneous processor increasingly integrated into modern mobile processors. In this paper, we introduce COUPLE, an orchestration framework for video analytics on heterogeneous mobile processors, with the goal of optimizing real-time video analysis through the collaboration of CPU, GPU and DSP. To tackle the accuracy loss of DSP inference, we introduce the Anchor Frame Calibration mechanism, utilizing high-precision GPU inference results and frame similarities to mitigate accuracy loss on the DSP. Additionally, we design a lightweight progressive scheduler to distribute video frames to GPU and DSP, maximizing inference Average Precision (AP) under performance (i.e., frame rate) and power constraints. COUPLE has been implemented on the Qualcomm's Snapdragon 888 mobile SoC, extensive evaluation results demonstrate its efficacy in imnroving the inference performance and accuracy.
暂无评论