Research On The Mutiplexing Of EMBB And URLLC In 5G NR

The evolution of the 5G standardization was devoted to solving the diversified and differentiated performance indexes in different application scenarios.The lack of applicable spectrum resources required that the 5G NR spectrum resources should meet the requirements of simultaneously carrying ultra high-speed and ultra-low latency services.This thesis mainly discusses and studies eMBB,URLLC KPI and related technologies from the perspective of system level modeling and physical layer design.In order to achieve coexistence of two service models with different numerology in the same spectrum,the thesis analyzes the multiplexing method of eMBB and URLLC based on Short TTI scheduling,aiming at reducing resource waste and improve system spectrum efficiency.The main contributions and innovations are as follows:The thesis firstly performs a system-level modeling of URLLC based on Short TTI technology and queuing theory model,analyzes how to efficiently transmit URLLC services under limited bandwidth and strict QoS constraints,and designs Compact DCI to increase its transmission effeciency.For eMBB,the standardized CBG retransmission technology is applied.In the thesis,aiming at the multi-service scenario in cellular network,a scheme of resource allocation on Short TTI transmission mode is designed.The design of this scheme is to give priority to scheduling bursty traffic so that the whole system delay is as low as possible.Through system-level simulation,the throughput and latency of single-service and multi-service are analyzed.The Short TTI application scenario that is compatible with the eMBB/URLLC is proposed,which lays a foundation for the service multiplexing technology.Based on the hybrid service with Short TTI scheduling,different scheduling TTIs for different services are studied to evaluate the system performance when coexisting.For resource reservation multiplexing scheme,we analyze the impact of the reserved bandwidth and system load of the URLLC on the eMBB service performance through the queuing theory model.For dynamic resource multiplexing,this thesis uses the convex optimization theory to perform system-level modeling analysis of the eMBB loss,aiming at optimizing the performance of the URLLC while optimizing the eMBB transmission efficiency.Finally,the thesis designs the parameter configuration of pre-emption indication signaling in dynamic resource multiplexing in detail,and combines CBG retransmission technology to evaluate the rationality of resource allocation scheme and throughput through system level simulation,which is important to the eMBB region with high hotspot and high capacity.