Network Slice Random Access Resource Allocation to Reduce Collisions

The patent segments random access resources by associating them with different network slices. Each network slice is assigned dedicated random access resources (time-frequency resources and/or preambles), dividing the shared resource pool into slice-specific subsets. This segmentation allows multiple terminal devices to access different slices simultaneously without collision, while maintaining high resource utilization through slice-specific allocation.

The patent applies local quality by assigning different random access resource characteristics to different network slices based on their specific service requirements. Each slice receives resources with quality tailored to its needs (e.g., URLLC slices get resources with lower collision probability, while eMBB slices can tolerate higher collision rates). This localized optimization resolves the contradiction by allowing high utilization in less critical slices while ensuring low collision probability in critical slices.

The patent implements universality by creating a unified random access resource allocation framework that serves multiple network slices with different service requirements through a single mechanism. The network device transmits configuration information that simultaneously defines resource associations for all slices, allowing one allocation system to handle diverse services (eMBB, URLLC, mMTC) without requiring separate allocation procedures for each slice type.

The patent enables service differentiation by assigning locally optimized random access resources to each network slice based on its specific QoS requirements. Different slices receive resources with characteristics matched to their needs (e.g., more preambles or dedicated time-frequency resources for slices with higher reliability requirements). This local customization within a unified framework resolves the contradiction between allocation simplicity and service differentiation.