Distributed Scheduling for CPRI Over Ethernet
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Ethernet-based CPRI transport networks in 5G wireless communication systems face challenges in ensuring low latency, high bandwidth, and synchronization due to the best-effort nature of Ethernet, which does not support frame delivery guarantees, leading to potential delays and jitter in CPRI traffic.
Innovation Solution
A distributed scheduling algorithm for Ethernet is introduced to manage CPRI traffic by using a distributed timeslot scheduling mechanism (DTSCoE) that reserves network resources for CPRI flows, allowing for periodic timeslot reservations and flexible frame sizing to minimize queuing delays and jitter, and employs techniques like time-shifting and frame preemption to optimize link utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If Ethernet is used as a best-effort communication protocol for CPRI transport, then equipment cost is reduced and shared infrastructure is enabled, but frame delivery guarantee is lost leading to increased latency and jitter
Solution Approach 1:
The Ethernet network is segmented into virtual channels (VCs) that provide dedicated bandwidth guarantees for CPRI traffic while sharing the physical infrastructure. Each VC acts as an independent logical channel with reserved resources, separating CPRI traffic from best-effort traffic while using the same physical Ethernet medium.
Solution Approach 2:
A scheduling mechanism acts as an intermediary between the best-effort Ethernet protocol and the deterministic requirements of CPRI traffic. The scheduler monitors queue depths, applies weighting factors, and dynamically adjusts transmission priorities to ensure frame delivery guarantees without requiring specialized hardware.
2Productivity
If statistical multiplexing is used in Ethernet shared infrastructure, then resource utilization is improved, but latency and jitter increase for time-sensitive CPRI traffic
Solution Approach 1:
The scheduling algorithm dynamically adjusts the behavior of virtual channels based on real-time traffic conditions. When CPRI traffic arrives, the scheduler dynamically prioritizes its transmission by adjusting queue service order and applying weighting factors, allowing the system to adapt between high resource utilization and low latency requirements.
Solution Approach 2:
The scheduling mechanism continuously monitors queue depths and transmission delays, using this feedback to adjust scheduling decisions. When CPRI queues build up or latency thresholds are approached, the scheduler increases priority for those channels, creating a closed-loop control system that maintains performance guarantees while maximizing resource utilization.
3Loss of time
If distributed timeslot scheduling is implemented for CPRI over Ethernet, then queuing delays and jitter are reduced, but scheduling complexity increases
Solution Approach 1:
The scheduler implements partial timeslot reservation specifically for CPRI traffic rather than fully reserving all resources. By applying scheduling algorithms only to the extent necessary for CPRI guarantees while allowing best-effort traffic to use remaining capacity, the system achieves low jitter without the complexity of complete resource reservation.
Solution Approach 2:
The scheduling complexity is managed by changing key parameters such as weighting factors and queue thresholds rather than implementing complex algorithms. These parameter adjustments allow the scheduler to adapt to different traffic patterns and performance requirements without requiring sophisticated control logic.
Data Source
AI summary
Systems and methods are provided for scheduling the transmission of data in a datapath. A source node can send a registration request indicating its preferred sending time for data transmission. Intermediate nodes can determine if there are overlaps in timeslot reservations and adjust, and schedule, the requested preferred sending time accordingly.