Dynamic DSP Cluster Allocation for Mixed Data Rate Channels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for processing high and low data rate channels in DSP clusters face limitations, including individual DSP processing limitations, fixed resource utilization, high costs, and difficulty in parallel processing due to dependency between functions.
Innovation Solution
A dynamic allocation method in DSP clusters that assigns channels based on bit rates, using a scheduler to distribute channels among slave DSPs, with the option of assigning one or multiple DSPs depending on channel bit rates, and downloading corresponding signal processing function chains from an SDR library to maintain throughput and reduce latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If each DSP processes low data rate channels individually, then latency is reduced for low data rate channels, but the system cannot process channels above a certain data rate due to individual DSP limitations
Solution Approach 1:
Multiple slave DSPs are merged to process high data rate channels collectively. The master DSP assigns multiple slave DSPs to handle channels whose bit rates exceed the capability of a single slave DSP, enabling the system to process high data rate channels that would otherwise be impossible for individual DSPs to handle.
Solution Approach 2:
The system dynamically adjusts the allocation of slave DSPs based on channel bit rates. The master DSP scheduler assigns one or more slave DSPs to each channel depending on the specific data rate requirements, allowing the system to adapt to varying channel conditions and maintain low latency for low data rate channels while handling high data rate channels effectively.
2Productivity
If DSPs process high data rate channels in a fixed pipeline manner, then high data rate channels are processed, but resource utilization is poor leading to waste
Solution Approach 1:
The system implements dynamic resource allocation where the master DSP scheduler assigns slave DSPs to channels based on real-time bit rate requirements. This dynamic assignment allows flexible utilization of slave DSPs across different channels and time periods, preventing resource waste while maintaining high processing throughput for high data rate channels.
Solution Approach 2:
Slave DSPs are designed to be universal resources that can be assigned to different channels based on their data rate requirements. Rather than dedicating specific DSPs to fixed functions or channels, the same slave DSPs can serve multiple channels with varying bit rates, improving overall resource utilization while maintaining processing capability.
3Productivity
If a high performance DSP is used to process high data rate channels, then processing capability is sufficient, but the cost increases
Solution Approach 1:
The processing capability is segmented across multiple standard-performance slave DSPs rather than relying on a single high-performance DSP. By dividing the processing workload among multiple less expensive slave DSPs that are assigned dynamically by the master DSP, the system achieves the necessary processing capability for high data rate channels while avoiding the high cost of a single powerful DSP.
Solution Approach 2:
Instead of using one expensive high-performance DSP, the system uses multiple copies of standard-performance slave DSPs. These replicated slave DSPs can collectively handle high data rate channels when assigned in parallel, providing the necessary processing capability at a lower overall cost compared to a single high-performance unit.
4Speed
If functions are divided for parallel processing on a DSP cluster, then processing speed increases, but it is difficult to divide functions due to dependencies between consecutive functions
Solution Approach 1:
The system dynamically determines the number of slave DSPs needed based on channel bit rates rather than statically dividing functions. The master DSP scheduler calculates the required processing capacity and assigns an appropriate number of slave DSPs to each channel, allowing parallel processing to proceed without complex function division while adapting to different data rate requirements.
Data Source
AI summary
The present invention provides a dynamic allocation method in Digital Signal Processors (DSPs) for processing high and low data rate channels. Said method takes advantage of Software Defined Radio (SDR) library and the different requirements for processing high and low speed channels in radio communications, to combine the processing method of high data rate channels and the processing method of low data rate channels into one DSP cluster. Thus said method can process both high and low data rate channels simultaneously in the same DSP cluster. Said method can maintain the processing throughput of high data rate channels, while reducing latency on low data rate channels, so can improve the processing performance and save the cost.