A KWS System for Edge-Computing Applications with Analog-Based Feature Extraction and Learned Step Size Quantized Classifier by Yukai Shen, Binyi Wu, Dietmar Straeussnigg, Eric Gutierrez

“…The filter bank is behaviorally modeled, making use of second-order band-pass transfer functions, simulating the analog front-end (AFE) processing. To enable efficient deployment, the GRU classifier is trained using a Learned Step Size (LSQ) and Look-Up Table (LUT)-aware quantization method. …”
Get full text

Secret Key Generation Driven by Attention-Based Convolutional Autoencoder and Quantile Quantization for IoT Security in 5G and Beyond by Anas Alashqar, Ehsan Olyaei Torshizi, Raed Mesleh, Werner Henkel

“…Additionally, a quantile-based quantization scheme is proposed to enhance key randomness and entropy, thereby strengthening security and resilience against potential threats. …”
Get full text

Deployable Deep Learning for Cross-Domain Plant Leaf Disease Detection via Ensemble Learning, Knowledge Distillation, and Quantization by Mohammad Junayed Hasan, Suvodeep Mazumdar, Sifat Momen

“…Our four-model ensemble (DenseNet-121, ResNet-101, DenseNet-201, EfficientNet-B4) achieves 99.15% accuracy via soft-voting. …”
Get full text

Lightweight faster R-CNN for object detection in optical remote sensing images by Andrew Magdy, Marwa S. Moustafa, Hala M. Ebied, Mohamed F. Tolba

“…First, aware training employs mixed-precision FP16 computation, which enhances training speed by a factor of 1.5 to 5.5 while preserving model accuracy and optimizing memory efficiency. Second, post-training compression applies unstructured weight pruning to eliminate redundant parameters, followed by dynamic quantization to reduce precision, thereby minimizing the model size at runtime and computational load. …”
Get full text

Low-Resolution Quantized Precoding for Multiple-Input Multiple-Output Dual-Functional Radar–Communication Systems Used for Target Sensing by Xiang Feng, Zhongqing Zhao, Jiongshi Wang, Jian Wang, Zhanfeng Zhao, Zhiquan Zhou

“…Furthermore, we propose a dynamic projection refinement algorithm within the alternating direction method of multiplier framework to efficiently solve these sub-problems. Numerical experiments demonstrate that our proposed method outperforms existing state-of-the-art techniques, particularly in terms of bit error rate in low signal-to-noise ratio scenarios.…”
Get full text

Optimizing Machine Learning Models with Data-level Approximate Computing: The Role of Diverse Sampling, Precision Scaling, Quantization and Feature Selection Strategies by Ayad M. Dalloo, Amjad J. Humaidi

“…In this paper, we propose a framework that uses data-level approximate computing techniques, including by diverse sampling strategies, precision scaling, quantization, and feature selection methods, to evaluate the impact of these techniques on the computational efficiency and accuracy of KNN and SVM models. …”
Get full text

An effective approach for real-time drowsy driving prediction using quantized fisher-Gabor features and latent-dynamic conditional random fields by Samy Bakheet, Ayoub Al-Hamadi, Abed Alanazi

Get full text

RiceLeafClassifier‐v1.0: A Quantized Deep Learning Model for Automated Rice Leaf Disease Detection and Edge Deployment by Oluwaseun O. Martins, Christiaan C. Oosthuizen, Dawood A. Desai

“…Additionally, inference time per image was significantly lower at 2.28 ± 0.35 s for the quantized model compared to 0.01 ± 0.01 s for the standard model (p < 0.0001), demonstrating substantial gains in efficiency. …”
Get full text

CADTrans: A code tree-guided CAD generative transformer model with regularized discrete codebooks by Xufei Guo, Xiao Dong, Juan Cao, Zhonggui Chen

“…Firstly, three regularized discrete codebooks are extracted through vector quantized adversarial learning, with each codebook respectively representing the features of Loop, Profile, and Solid. …”
Get full text

Enhancing microgrid forecasting accuracy with SAQ-MTCLSTM: A self-adjusting quantized multi-task ConvLSTM for optimized solar power and load demand predictions by Ehtisham Lodhi, Nadia Dahmani, Syed Muhammad Salman Bukhari, Sujan Gyawali, Sanjog Thapa, Lin Qiu, Muhammad Hamza Zafar, Naureen Akhtar

“…The SAQ-MTCLSTM incorporates a sophisticated architecture that combines convolutional and LSTM layers with self-aware quantization to enhance computational efficiency and model adaptability. …”
Get full text

Switched 32-Bit Fixed-Point Format for Laplacian-Distributed Data by Bojan Denić, Zoran Perić, Milan Dinčić, Sofija Perić, Nikola Simić, Marko Anđelković

“…Precision analysis is achieved using the signal-to-quantization noise ratio (SQNR) as a performance metric, introduced based on the analogy between digital formats and quantization. …”
Get full text

Construction of a Deep Learning Model for Unmanned Aerial Vehicle-Assisted Safe Lightweight Industrial Quality Inspection in Complex Environments by Zhongyuan Jing, Ruyan Wang

“…A network-sparsifying pruning training method based on a channel importance mechanism is proposed to transform the pruning training process into a constrained optimization problem. A quantization-aware training method is proposed to automate the learning of quantization bitwidths to improve the adaptability between features and data representation accuracy. …”
Get full text

Adaptive Fixed-Time Practically Tracking Control for n-Link Flexible-Joint Manipulator With Input Quantization and Input Delay via Command Filter-Based Approach by Rongyang Qiu, Meizhen Xia, Jiu Zhang, Chao Ma

“…Furthermore, the effects of communication pressure and input delay are explicitly handled with a hysteresis uniform quantizer. Concurrently, the proposed method efficiently mitigates the adverse impacts of unmodelled dynamics by an auxiliary signal. …”
Get full text

On-Edge Deployment of Vision Transformers for Medical Diagnostics Using the Kvasir-Capsule Dataset by Dara Varam, Lujain Khalil, Tamer Shanableh

“…The seven ViT models selected for this study are EfficientFormerV2S2, EfficientViT_B0, EfficientViT_M4, MobileViT_V2_050, MobileViT_V2_100, MobileViT_V2_175, and RepViT_M11. …”
Get full text

Fast intra prediction algorithm applied in QS enhancement layer by WANG Da-yong1 SHU Jian2

“…A fast intra coding algorithm was proposed which was suitable for enhancement layer in quality salability.Base on the coding structure of quality scalability,our algorithm used the relationships between layers,the relationships between rate distortion and modes,and temporal and spatial relationships to predict the modes and coding order to be ap-plied,with an early termination determined by the relationships between quantization parameters and modes,residual co-efficients,and rate distortion.The experimental results show that the encoding speed of the proposed algorithm can be improved by 70%,with negligible coding loss.…”
Get full text

A Novel Training Strategy for Deep Learning Model Compression Applied to Automatic Modulation Classification by Mateus A. Goldbarg, Micael Balza, Sergio N. Silva, Lucileide M. D. Silva, Marcelo A. C. Fernandes

“…This strategy aims to minimize both pruning and quantization losses during the training of compressed models, thereby reducing the computational complexity of DNNs. …”
Get full text

Neural network quantification for solar radiation prediction: An approach for low power devices by Brenda Alejandra Villamizar-Medina, Angelo Joseph Soto Vergel, Byron Medina-Delgado, Darwin Orlando Cardozo-Sarmiento, Dinael Guevara-Ibarra, Oriana Alexandra Lopez-Bustamante

“…Quantization enables complex predictive models to run on low-cost, energy-efficient devices, thereby democratizing advanced prediction technologies for critical applications like solar energy generation. …”
Get full text

Adaptive key distillation from channel characteristics by Qiao DAI, Liang JIN, Kai-zhi HUANG

“…Approaches generating secret keys based on radio channel characteristics can’t guarantee the key length and system efficiency at the same time because of the low entropy rate or high disagreement ratio of keys.An adaptive key distillation scheme based on the quantization of channel state information was designed.An upper bounding function was used as an approximation of the real one to improve the entropy of keys under the constraint of disagreement rate.Based on this,the key agreement scheme resulting in longer keys was selected.Simulation results show that with this scheme,the length of key and efficiency of the system can be guaranteed at the same time.…”
Get full text

An Improved Low-Bit-Rate Image Compression Framework Based on Semantic-Aware Model and Neighborhood Attention by Chengbin Zeng, Liang Zhang

“…This process can minimize quantization errors and facilitate efficient vector quantization. …”
Get full text

Intelligent energy saving measurement method and realization of power grid based on IoT by Dapin ZHANG, Qinghai OU, Yeshen HE, Yiying ZHANG

Get full text