Realize Ultra-Reliability and Low-Latency in Haptic Communication through Prediction

Abstract

Haptic communications play an essential role in improving the immersive experience of users. However, it requires a lower latency, as well as the higher continuity and reliability of the received information. Unfortunately, the existing technologies are still not adequate to fully satisfy these requirements, especially when radio resources are limited and the communication environment is fluctuating. To this end, we propose a prediction-based resource allocation scheme for haptic communications to satisfy the low latency and high reliability. Specifically, in order to reduce the risk of latency, we construct a prediction model, which has an optimized prediction window size and uses the LSTM (Long Short-Term Memory) method to predict future haptic information. Then, the predicted information would be transmitted in advance according to the window size. In addition, since the prediction model inevitably brings the error, which may well reduce the reliability of haptic transmission. In order to further satisfy reliability and save resources, taking power resources as an example, according to the relationship between power and reliability, we propose optimization problems and solutions based on the relationship between power and reliability. Further, power determines the predicted window size by optimizing the power. Due to optimization problems intractability, we develop a joint bisection and trisection algorithm to obtain the optimal solution. Simulation results show that the optimal transmit power can be obtained, and the error can be minimized through our scheme.