| DC Field | Value | Language |
|---|
| dc.contributor.author | Qikai Wang | - |
| dc.coverage.spatial | Минск | en_US |
| dc.date.accessioned | 2026-08-28T08:05:39Z | - |
| dc.date.available | 2026-08-28T08:05:39Z | - |
| dc.date.issued | 2026 | - |
| dc.identifier.citation | Qikai Wang. Semantic communication system for vehicle detection / Qikai Wang // Технологии передачи и обработки информации : материалы Международного научно-технического семинара, Минск, апрель 2026 г. / Белорусский государственный университет информатики и радиоэлектроники ; редкол.: В. Ю. Цветков [и др.]. – Минск, 2026. – С. 165–167. | en_US |
| dc.identifier.uri | https://libeldoc.bsuir.by/handle/123456789/65614 | - |
| dc.description.abstract | This paper presents a semantic communication system for vehicle detection over noisy wireless channels. A Convolutional Neural Network (CNN) semantic encoder compresses image features with multiple compression ratios to transmit task-related information only, which is then
processed by a pre-trained You Only Look Once version 12 (YOLOv12) object detector. A Supernet with shared encoder weights and Sandwich Rule training is proposed to reduce redundancy. Experiments on Additive White Gaussian Noise (AWGN) channels show that the semantic system
significantly outperforms traditional transmission at low signal-to-noise ratio (SNR), and the Supernet achieves comparable performance with independent models. | en_US |
| dc.language.iso | en | en_US |
| dc.publisher | БГУИР | en_US |
| dc.subject | материалы конференций | en_US |
| dc.subject | semantic communication | en_US |
| dc.subject | vehicle detection | en_US |
| dc.subject | supernet | en_US |
| dc.title | Semantic communication system for vehicle detection | en_US |
| dc.type | Article | en_US |
| Appears in Collections: | Технологии передачи и обработки информации : материалы Международного научно-технического семинара (2026)
|
