Latency-Reliability-Accuracy Optimization for Wireless Transmission of Sensory Data

Real-time wireless applications, such as Tactile Internet or robotic manipulation, pose stringent latency and reliability requirements. This is critical when a large volume of sensory data is transmitted, requiring joint consideration of compression and communication in terms of latency, reliability, and accuracy. This letter proposes a scheme for uplink latency reduction upon transmission of sensory data with lossy compression. The aggregated latency consists of compression and transmission latency, respectively, each of them represented by a random variable. We investigate the interaction between latency reduction, reliability, and lossy compression through a suitable optimization problem. The results reveal that a latency reduction of up to >20% is achieved with 0.9999 reliability and 1% accuracy loss for 25Mb of data transmission.