The impact of the exponential Kernel’s bandwidth parameter on learning algorithms [PDF]
Exponential kernels have been used considerably in statistics, machine learning, and artificial intelligence for tasks such as kernel principal component analysis (Kernel PCA), support vector machines(SVM), visualization, clustering, and pattern ...
Mahdi A. Almahdawi
doaj +5 more sources
The Optimal Bandwidth Parameter Selection in GPH Estimation [PDF]
In this paper, the optimal bandwidth parameter is investigated in the GPH algorithm. Firstly, combining with the stylized facts of financial time series, we generate long memory sequences by using the ARFIMA (1, d, 1) process.
Weijie Zhou +3 more
doaj +3 more sources
Direct Synthesis of Dual-Parameter Concentric Ring RA with Enhanced Bandwidth [PDF]
Reflectarray antennas (RAs) are nowadays a quite popular technology, used in several applications, due to a significant number of attractive properties, such as low cost, low weight, conformal deployment, and the possibility of introducing suitable ...
Bui Van Ha +3 more
doaj +5 more sources
Multi-Parameter AI-Based Bandwidth Compensation for Energy-Efficient 800G Transmission
We propose a novel energy-efficient AI-based bandwidth compensation technique that jointly optimizes Tx and Rx static filters. Experimental demonstration in a 800G system reveals gains of more than 1 dB when compared with typical digital pre-emphasis.
Fernandes, Marco A. +6 more
+4 more sources
Lower bounds on parameter modulation-estimation under bandwidth constraints [PDF]
Submitted to IEEE Transactions on Information ...
Nir Weinberger, Neri Merhav
+6 more sources
Estimating a Cosmological Mass Bias Parameter with Bootstrap Bandwidth Selection [PDF]
SummaryWe focus on selecting optimal bandwidths for non-parametric estimation of the two-point correlation function of a point pattern. We obtain these optimal bandwidths by using a bootstrap approach to select a bandwidth that minimizes the integrated squared error.
Ji Meng Loh, Woncheol Jang
openaire +2 more sources
A Distributed Source Coding Framework for Bandwidth-Efficient Neural Network Parameter Updates [PDF]
Neural networks have helped solve numerous problems and create applications in various fields in the past few years. In order for the neural networks to perform well, the number of parameters has had to grow considerably in size.
Annabella Chou
doaj +4 more sources
RESEARCH OF LOW-BANDWIDTH RADIONETWORKS QOS PARAMETERS
Background. This article describes the research of QoS parameters in low-bandwidth communication networks based on VHF (Very High Frequency) radio stations produced by Aselsan (Turkey) and Harris (USA). Objective. The aim of the paper is the research of data transfer delay, jitter of data transfer delay, packet losses and analysis of the possibility
Strelkovskaya, Irina, Zolotukhin, Roman
openaire +5 more sources
Model parameters for wide time bandwidth sonar signals [PDF]
Wide time-bandwidth (TW) product signals have been used historically in radar and communications for practical applications. Matched filter processing gain for these signals is predictable and offers highly attractive performance in the presence of gaussian noise.
George F. Rodgers
openaire +2 more sources
Measuring Mean Frequency, Power, and Bandwidth: Key Parameters in Signal Analysis [PDF]
Abstract Signal analysis plays a crucial role in various scientific, engineering, and technological domains, providing valuable insights into the characteristics of different signals. Among the fundamental parameters to be measured, mean frequency, power, and bandwidth stand out as key metrics. In this article, we delve into the significance of
Walid Mohamedi
openaire +2 more sources

