paper-with-me

홈 › Papers

Disaggregated Deep Learning via In-Physics Computing at Radio Frequency

2025-04-24 · Zhihui Gao, Sri Krishna Vadlamani, Kfir Sulimany, Dirk Englund, Tingjun Chen

Modern edge devices, such as cameras, drones, and Internet-of-Things nodes, rely on deep learning to enable a wide range of intelligent applications, including object recognition, environment perception, and autonomous navigation. However, deploying deep learning models directly on the often resource-constrained edge devices demands significant memory footprints and computational power for real-time inference using traditional digital computing architectures. In this paper, we present WISE, a novel computing architecture for wireless edge networks designed to overcome energy constraints in deep learning inference. WISE achieves this goal through two key innovations: disaggregated model access via wireless broadcasting and in-physics computation of general complex-valued matrix-vector multiplications directly at radio frequency. Using a software-defined radio platform with wirelessly broadcast model weights over the air, we demonstrate that WISE achieves 95.7% image classification accuracy with ultra-low operation power of 6.0 fJ/MAC per client, corresponding to a computation efficiency of 165.8 TOPS/W. This approach enables energy-efficient deep learning inference on wirelessly connected edge devices, achieving more than two orders of magnitude improvement in efficiency compared to traditional digital computing.

📄 PDF Abstract BibTeX arXiv:2504.17752

Code (0)

등록된 구현이 없습니다.

Tasks

Autonomous NavigationDeep Learningimage-classificationImage ClassificationObject Recognition

Similar Papers 제목 키워드 기반

VoltanaLLM: Energy-Efficient and SLO-Aware Disaggregated LLM Serving via Adaptive Frequency Control and State-Space Routing

2025-09-05 · Jiahuan Yu, Aryan Taneja, Junfeng Lin, Minjia Zhang arxiv

The energy cost of Large Language Model (LLM) inference is rapidly becoming a barrier to sustainable and scalable deployment. Although modern serving architectures expose distinct prefill and decode behaviors, existing s…

Revisiting Disaggregated Large Language Model Serving for Performance and Energy Implications

2025-11-14 · Jiaxi Li, Yue Zhu, Eun Kyung Lee, Klara Nahrstedt arxiv

Different from traditional Large Language Model (LLM) serving that colocates the prefill and decode stages on the same GPU, disaggregated serving dedicates distinct GPUs to prefill and decode workload. Once the prefill G…

A circuit for precise random frequency synthesis via a frequency locked loop

2019-02-25

Frequency synthesis (FS) is a technique vital for all kinds of radio frequency (RF) communications, such as: mobile phones, Bluetooth, Wi-Fi, radio, TV and satellite, and in other equipment requiring periodic signals of …

Next-generation Co-Packaged Optics for Future Disaggregated AI Systems

2023-03-03 · Sajjad Moazeni

Co-packaged optics is poised to solve the interconnect bandwidth bottleneck for GPUs and AI accelerators in near future. This technology can immediately boost today's AI/ML compute power to train larger neural networks t…

Cloud Computing

Type III solar radio burst detection and classification: A deep learning approach

2021-05-27 · Jeremiah Scully, Ronan Flynn, Eoin Carley, Peter Gallagher 외

Solar Radio Bursts (SRBs) are generally observed in dynamic spectra and have five major spectral classes, labelled Type I to Type V depending on their shape and extent in frequency and time. Due to their complex characte…

Vocal Bursts Type Prediction