SBIR-STTR Award

Communication bandwidth is precious, and transmission of real-time digital video can often be challenging or infeasible. We have developed a machine learning-based imagery codec that can compress videos and still images far beyond the current state-of-the-art, requiring just 20% of the throughput to transmit HD video as the leading industry codecs (HEVC/h.265). It can run in real-time on computationally limited devices (e.g. tablets, smartphones)compression,video,ISR,network,machine learning,sensor,Information

Wireless bandwidth is precious on the battlefield. This presents a particular problem in transmission of HD digital video from edge devices such as UAVs. Entheleon Technologies has developed a state-of-the-art video compression codec that can reduce the bandwidth required to transmit video by up to 90%. The focus of this proposed work is to deploy Entheleon's video codec to hardware environments - computers, sensors, and radios - relevant to USAF. The work is sponsored by the AFRL to test the performance of codec implementations for a drone swarm development program. The work will also develop custom codecs for the MQ-9 digital communications IPT to address challenges with mult-cast transmission of aerial intelligence data and visual telemetry for take-off and landing operations. This work also supports the MQ-9 wing of the California Air National Guard, enabling multi-cast transmission during firefighting and disaster relief efforts.