Statistical Delay and Error-Rate Bounded QoS Control for mURLLC Over 6G Multimedia Mobile Wireless Networks in the Non-Asymptotic Regime

To support the exponentially increasing demand for bandwidth-intensive and delay-sensitive real-time multimedia data services, researchers have made tremendous efforts to provide a better quality-of-service (QoS) when designing next generation wireless network architecture models for massive ultra-reliable and low-latency communications (mURLLC). One of the major design issues raised by mURLLC is how to guarantee stringent delay and error-rate bounded QoS requirements when implementing short-packet data communications, such as finite blocklength coding (FBC), over highly time-varying wireless fading channels. To efficiently accommodate statistical QoS for mURLLC over 6G mobile wireless networks, it is important to remodel wireless fading channel's stochastic-characteristics by defining statistical QoS metrics and their analytical relationships, such as delay-bound-violating probability, effective capacity, error probability, outage capacity, etc., when applying FBC. However, when being integrated with FBC, how to rigorously characterize stochastic dynamics of wireless networks in terms of statistically upper-bounding both delay and error-rate QoS metrics has been neither well understood not thoroughly studied. To overcome these obstacles, in this paper we develop analytical frameworks and control mechanisms for statistical delay and error-rate bounded QoS in non-asymptotic regime. First, we establish FBC-based wireless-fading channels model by characterizing various information-theoretic specifications. Then, we develop a set of new statistical delay and error-rate bounded QoS metrics, tradeoff-functions, and control mechanisms including epsilon-effective capacity, delay-bound-violating probability, Markov-model-based QoS-exponents functions, and FBC-based outage-capacity in finite blocklength regime. Finally, our simulation results validate and evaluate our developed mechanisms for statistical QoS in supporting mURLLC over 6G wireless networks.