In this section, we introduce an adaptive offloading strategy into the model and deploy a practical edge computing scenario based on multiple types of embedded terminals. We first verify the performance and advantages of our proposed innovative…
Category: 7. Maths
-

Comparison of conventional and radiomics-based analysis of myocardial infarction using multimodal non-linear optical microscopy
Ovine myocardial infarction model and sample preparation
An ovine infarction model was provided by the Heart Rhythm Disease Institute, Bordeaux University Hospital, France in accordance with the guidelines from Directive 2010/63/EU – 2010 of the…
Continue Reading
-

Quantifying teachers’ readiness for artificial intelligence adoption in education: a mathematical modeling perspective
Context and participants
The study was conducted within the context of Nigeria, focusing on teachers’ readiness to adopt Artificial Intelligence (AI) in Education (AIED). Nigeria presents a unique environment characterized by diverse levels of…
Continue Reading
-

Deep learning for enhancing automatic classification of M-PSK and M-QAM waveform signals dedicated to single-relay cooperative MIMO 5G systems
Relevant feature extraction and processing
To accurately estimate the M-PSK and M-QAM waveform signals of each \({\textbf{y}}_{d ^(j)}\) at the \(j^{\textrm{th}}\) antenna, selecting a relevant set of important features is essential. Higher-order…
Continue Reading
-

A Generative Framework for Probabilistic, Spatiotemporally Coherent Downscaling of Climate Simulation
Data
We analyze four interrelated atmospheric variables, namely the zonal (uas) and meridional (vas) components of near-surface (10 m) wind speeds, surface (2 m) air temperature (tas), and sea level pressure (psl), due to their crucial…
Continue Reading
-

Dual-stream interactive mechanism with multi-modal hierarchical aggregation transformer for gait recognition
This section evaluates the performance of the GaitSMAT model on three mainstream datasets (Grew24, Gait3D25 and SUSTech1k26). We will describe these datasets and their corresponding training configurations in detail. We also compare GaitSMAT with…
Continue Reading
-

Multitasking vision language models for vehicle plate recognition with VehiclePaliGemma
This section presents the results of evaluating and comparing our proposed solution, which leverages the OCR capabilities of VLMs to address the challenging problem of car plate recognition. Several VLMs were evaluated and compared in terms of…
Continue Reading
-

A synergistic approach using digital twins and statistical machine learning for intelligent residential energy modelling
Table 4 provides a comparative analysis of four models: Random Forest, Deep Neural Network (DNN), Long Short-Term Memory (LSTM), and Bidirectional LSTM (BiLSTM) based on precision, recall, F1-score, and support for two classes (Class 0 and Class…
Continue Reading
-

Deep learning-based profiling side-channel attacks in SPECK cipher
Notation
Calligraphic letters \(\mathscr {X}\) represent sets and where the calligraphic letter is finite, \(|\mathscr {X}|\) will represent the cardinality of \(\mathscr {X}\). X denote random variables and random vectors X over \(\mathscr {X}\)….
Continue Reading
-

Application and evaluation of industrial waste residue and bentonite for solidification and stabilization of sewage sludge
UCS test
Figure 3 highlights the variations in the UCS of cement-solidified sludge at various PC contents and curing ages. Notably, the early strength of the cement-solidified sludge increased relatively rapidly, with the strength at 7 days…
Continue Reading
