Radio MAC Scheduling with Transport Network Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional radio schedulers in central radio access networks (C-RAN) are unable to effectively adapt to congestion in statistical multiplexing transport networks, leading to uncontrolled delay and packet loss, which affects the quality of service for latency-critical and high-priority flows.
Innovation Solution
Implementing a Cooperative Transport Interface (CTI) to provide feedback from the transport network scheduler to the radio scheduler, allowing for adaptive MAC scheduling decisions based on transport network conditions, ensuring fair share allocation and prioritization of traffic to maintain end-to-end quality of service.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional radio schedulers operate independently without transport network feedback, then radio resource allocation can be optimized for wireless channel conditions, but transport network congestion cannot be detected or responded to, leading to uncontrolled delay and packet loss
Solution Approach 1:
The patent implements a feedback mechanism where the transport network scheduler provides congestion information to the radio scheduler. This allows the radio scheduler to adapt its scheduling decisions based on actual transport network conditions, preventing packet loss and delay by proactively adjusting radio resource allocation when congestion is detected.
Solution Approach 2:
The patent merges the radio scheduling function with transport network scheduling information. By integrating transport network congestion status into the radio scheduling decision-making process, the system achieves coordinated optimization across both radio and transport layers without requiring completely separate scheduling systems.
2Productivity
If radio scheduler allocates resources based solely on wireless channel conditions, then spectral efficiency is maximized, but transport network capacity constraints are violated, causing buffer overflow and packet loss
Solution Approach 1:
The patent makes the radio scheduling dynamically adaptive to transport network conditions. The scheduler continuously adjusts its resource allocation based on real-time feedback from the transport network, transitioning between aggressive allocation (when transport capacity is available) and conservative allocation (when congestion is detected), thereby balancing throughput and reliability.
Solution Approach 2:
The transport network scheduler provides advance congestion information to the radio scheduler before actual packet loss occurs. This preliminary warning allows the radio scheduler to proactively reduce allocation or prioritize traffic before transport buffers overflow, preventing packet loss rather than reacting after it occurs.
3Productivity
If transport network uses statistical multiplexing to increase capacity utilization, then network efficiency improves, but congestion becomes harder to predict and control, affecting latency-critical flows
Solution Approach 1:
The patent applies different scheduling strategies to different traffic types based on their requirements. Latency-critical and high-priority flows receive guaranteed resources and priority handling, while best-effort flows are subject to dynamic allocation based on available transport capacity. This local differentiation allows statistical multiplexing to benefit overall utilization without compromising time-sensitive traffic.
Solution Approach 2:
The patent introduces an intermediary feedback mechanism between the transport network scheduler and radio scheduler that specifically monitors conditions affecting latency-critical flows. This intermediary layer translates complex statistical multiplexing behavior into simple congestion indicators that the radio scheduler can use to protect time-sensitive traffic.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In certain embodiments, a 5G RAN network has a distributed unit (DU) connected to at least one remote unit (RU) via a transport network (TN) that handles both mobile front-haul (MFH) traffic between the DU and the RUs as well as non-MFH traffic with other end users. A radio scheduler in the DU performs medium access control (MAC) scheduling for the mobile user traffic, while a TN scheduler in the transport network performs TN scheduling for both the MFH and non-MFH traffic. The TN scheduler transmits information (e.g., MFH data-rate limits) to the radio scheduler that the radio scheduler uses to perform its subsequent MAC scheduling. In this way, the MAC scheduling can take into account transient constraints that exist in the transport network in order to avoid problems associated with congestion in the transport network due to the statistical nature of mobile user traffic as well as the non-MFH services.