Mobile emergency networks improve image delivery with in-network adaptation
Packet-Level In-Network Semantic Adaptation for Unstable Mobile Emergency Networks
Networking and Internet Architecture
Summary
Mobile emergency networks often face unstable wireless links that change faster than devices can adjust, causing delays and loss in image data needed for tasks like disaster response. The authors developed DINA, a method that breaks images into small parts and adapts them inside the network nodes without waiting for endpoint feedback. This approach helps deliver more useful image data on time, improving things like fire detection and rescue operations. Their tests showed notable increases in how much of the image data arrives before deadlines and how accurate related machine tasks become.
What this means in practice
- •For emergency response teams: Enhance timely image-based detection and classification in unstable wireless networks during disasters using packet-level adaptation.
- •For uav network operators: Improve data throughput and task accuracy in drone networks monitoring large areas with rapidly changing link conditions.
Authors
Zhiyuan Ren, Tao Zhang, Wenchi Cheng
Abstract
Mobile emergency networks can experience independently changing intermediate wireless links on timescales shorter than endpoint feedback can track. When an egress changes after packet emission, feedback affects only later source data, while the on-path node observes the current condition with the affected packet still mutable. This paper presents DINA, a packet-level in-network semantic adaptation method. An image is divided into self-describing spatial packets carrying coordinates, a current representation identifier, and payload. At each eligible node, an offline-trained frozen selector scores compatible operators, immediately transforms the packet, and forwards it without image reconstruction or cross-packet adaptation state. Later nodes can retain or further compact the packet through the same typed compatibility contract. The receiver places available packets by coordinate, fills missing regions with black, and runs a fixed machine task. We realize DINA in a 24-node UAV environment using XDP and AF_XDP. In the primary forest-fire trace, DINA raises deadline tile coverage from 40.4% to 72.6% and classification accuracy from 77.5% to 95.0% relative to forwarding. In an independently trained RescueNet segmentation case, it raises coverage from 65.6% to 91.7% and foreground mIoU from 0.486 to 0.541. Sufficient- and extreme-capacity profiles expose a no-gain boundary and a common task-failure boundary, respectively.