Hardware Accelerator Buffer Pool Management for WWAN Latency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Wireless wide area networks (WWAN) face challenges in processing speed and memory performance due to underperforming components in mobile devices, leading to network latency issues.

Innovation Solution

The implementation of hardware acceleration techniques such as early data deciphering, adaptive data aggregation, optimized data path processing, buffer pool management, and keystream processing to enhance processing performance by reducing operations, logic, and latency, and minimizing data copies.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If data deciphering is performed before copying to external memory, then processing speed is improved, but hardware complexity increases

Engineering Contradiction:
Improveprocessing speedVSAvoidhardware complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The patent performs data deciphering operations before copying data to external memory, rather than after. This preliminary processing approach ensures that data is decrypted while still in the processor's internal buffers, reducing the need for subsequent processing operations on encrypted data and improving overall processing throughput.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces specialized hardware components including a deciphering engine and buffer pool manager that act as intermediaries between the processor and external memory. These components handle the complex deciphering operations independently, allowing the main processor to continue executing other tasks while maintaining high processing speed.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of time

If adaptive data aggregation is implemented, then network latency is reduced, but device complexity increases

Engineering Contradiction:
Improvenetwork latencyVSAvoiddevice complexity
Core Design Contradiction:
Loss of timeVSDevice complexity

Solution Approach 1:

The patent implements adaptive data aggregation that dynamically adjusts aggregation parameters based on network conditions, data priority, and current system load. The buffer pool manager continuously monitors queue depths and latency metrics, adjusting aggregation thresholds in real-time to optimize performance without requiring complex manual configuration.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system incorporates feedback mechanisms where the buffer pool manager monitors aggregation performance metrics and uses this information to adjust aggregation parameters. This closed-loop control allows the system to automatically optimize latency performance based on actual network conditions while keeping the control logic manageable through standardized feedback processing.

Inventive Principle:
Principle #23Feedback

3Productivity

If multiple hardware paths are used for data processing, then processing performance is enhanced, but device complexity increases

Engineering Contradiction:
Improveprocessing performanceVSAvoiddevice complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent divides data processing into multiple specialized hardware paths including a fast path for already-decrypted data and a slow path for data requiring deciphering. Each path is optimized for its specific function, allowing parallel processing of different data types simultaneously. The segmentation is managed through simple routing logic that directs data based on its encryption state, maintaining low complexity despite multiple paths.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP2922323B1Hardware acceleration for WWAN technologies
Publication Date: 2020.06.17 QUALCOMM INC
  • EP2922323B1 patent drawingFigure 1
  • EP2922323B1 patent drawingFigure 2
  • EP2922323B1 patent drawingFigure 3

AI summary

A method, a processor, a computer program product and an apparatus are described for providing buffers for a hardware accelerator in a telecommunications device for packetized data. The method comprising providing a dynamic pool of buffers (520) for use by the hardware accelerator (540), wherein access to the pool of buffers (520) is controlled by software (560) in the telecommunications device, wherein the software (560) dynamically adjusts the pool of buffers (520) according to hardware needs indicated by at least one of status signals from the hardware (540) and threshold information on the pool of buffers (520), wherein buffers in the pool (520) are replenished via software control.