Dynamic Wireless Baseband Cluster Scheduling for Power Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional wireless baseband processing architectures are unsustainable due to overdesign for extreme scenarios, leading to wasted capabilities in common working scenarios and high power consumption, as they often feature extraordinarily powerful CPUs, large memories, and high bandwidth, which are underutilized.
Innovation Solution
A method and device that dynamically schedule wireless baseband processing clusters with varying capacities to match task requirements by monitoring current load states and allocating only necessary resources for specific tasks, allowing for efficient distribution of processing capacities and reducing power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If baseband chips are designed with extraordinarily powerful CPUs, large memories, and high bandwidth to satisfy maximum working loads, then the processing capability is improved, but power consumption and device complexity increase significantly
Solution Approach 1:
The baseband chip is divided into multiple processing clusters, each with different processing capabilities (first processing clusters with stronger capability, second processing clusters with weaker capability). This segmentation allows the system to activate only the necessary number and type of clusters based on current workload, avoiding the need to power all high-capability resources continuously.
Solution Approach 2:
The system dynamically adjusts the working state of processing clusters based on real-time load monitoring. The scheduling component can activate or deactivate clusters as needed, making the processing capability flexible and adaptive to changing conditions rather than static and always at maximum capacity.
2Productivity
If baseband chips are designed with ultra-large capacity memories and ultra-large bandwidth internetworking, then the processing capability is improved, but device complexity and investment cost increase
Solution Approach 1:
The memory and bandwidth resources are segmented and distributed across multiple processing clusters rather than providing one ultra-large memory and ultra-large bandwidth interconnection for the entire chip. Each cluster has its own memory and interconnection resources scaled to its specific needs, reducing overall complexity while maintaining adequate capacity.
Solution Approach 2:
Different processing clusters are equipped with memory and bandwidth resources matched to their specific processing capabilities. First processing clusters (stronger capability) have larger memory and higher bandwidth, while second processing clusters (weaker capability) have smaller memory and lower bandwidth, optimizing the balance between capability and complexity.
3Productivity
If base station processing capability is increased to satisfy hotspot areas, then the ability to handle maximum working load is improved, but the processing capability is wasted in common working scenarios
Solution Approach 1:
The system uses dynamic scheduling to activate only the necessary processing clusters based on current workload demands. In common scenarios with lower traffic, only a subset of clusters (or lower-capability clusters) are activated. In hotspot scenarios with maximum load, additional clusters or higher-capability clusters are activated, ensuring processing capability matches actual demand without waste.
Solution Approach 2:
The scheduling component continuously monitors the working loads of processing clusters and adjusts cluster activation decisions based on this feedback. This closed-loop control ensures that processing resources are allocated efficiently according to real-time conditions, preventing both over-provisioning and under-provisioning.
Data Source
AI summary
A method and device for processing a wireless baseband capable of expanding dynamically are provided. The method includes the following steps: a plurality of groups of wireless baseband processing clusters which have different processing capacities and are relatively independent are pre-constructed (S201); a current load working state of each group of wireless baseband processing clusters is acquired by monitoring the processing loads of the plurality of groups of wireless baseband processing clusters (S202); and according to baseband processing task requirements and the current load working state of each group of wireless baseband processing clusters, a certain number of wireless baseband processing clusters of which the current load working states are suitable for performing a baseband processing task are scheduled to enter a working state for performing the wireless baseband processing task (S203).