Embedded DMA Data Transfer for Real-Time DFOS Streaming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing distributed fiber optic sensing systems face challenges in delivering high-bandwidth data to cloud systems in real-time due to limited computing resources and software overhead, leading to throughput bottlenecks.
Innovation Solution
Employing a cyclic Direct Memory Access (DMA) technique that writes data from firmware to host memory in a cyclic mode, using a bit to indicate data availability and overflow status, with driver code monitoring and user space software handling jitter through large buffers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If high-bandwidth data is transmitted from embedded DFOS systems to cloud systems, then data throughput is improved, but software overhead and processing complexity increase
Solution Approach 1:
The patent extracts the data transfer function from the main CPU processing workflow by implementing a dedicated DMA engine. This separate hardware component handles memory-to-memory data movement independently, removing the software overhead burden from the main system while maintaining high throughput capability.
Solution Approach 2:
The patent introduces cyclic buffers as an intermediary data structure between the DMA engine and cloud transmission. These buffers act as a mediator that decouples the data generation rate from the transmission rate, absorbing software overhead variations while ensuring continuous high-bandwidth data flow to the cloud.
2Loss of time
If real-time data delivery is implemented, then data freshness is improved, but system resource consumption increases
Solution Approach 1:
The patent implements periodic data sampling and transmission cycles using the DMA engine. Instead of continuous high-power processing, data is transferred in periodic bursts at predetermined intervals, achieving real-time delivery requirements while allowing the embedded system to enter low-power states between transmission cycles.
Solution Approach 2:
The patent pre-configures cyclic buffers with sufficient capacity before data transmission begins. This preliminary action allows the system to accumulate data locally without immediate transmission, enabling batched real-time delivery that reduces the frequency of high-power transmission operations while maintaining data freshness.
Data Source
AI summary
Distributed fiber optic sensing (DFOS)/distributed acoustic sensing (DAS) that illustrate inventive techniques—applicable to all embedded systems—that deliver high-bandwidth traffic from a DFOS system to a cloud or other processing resources in real-time, employ firmware that operates at a register-transfer level and writes data repeatedly into the embedded host memory directly—without software intervention after initial configuration. This technique advantageously enables the use of the relatively large host memory to compensate for any software jitter. Preferably configured, the firmware employs only a small buffer to compensate for a transaction and host-memory arbitration latency. A second dedicated thread sends out data from the user space buffer to a cloud, or a connected server. Remote procedures that are executed on a remote system (a computer, network of computers, or cloud), receive data transmitted from the embedded system, and perform further processing as necessary.