Multi-Chip Daisy Chain for Low-Latency LiDAR Output Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lidar systems face challenges in efficiently aggregating data from multiple photodetectors without the need for shared memory buffers, leading to increased latency and storage requirements.
Innovation Solution
A daisy-chained optical receiver architecture with a common processed data pipeline and multiple processors that output data in an out-of-order manner, allowing for efficient data streaming without physical ordering or buffering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data from multiple processors is aggregated using traditional methods with shared memory buffers, then data aggregation is achieved, but latency increases and storage requirements increase
Solution Approach 1:
The patent divides the data aggregation system into multiple independent processing pipelines, each handling data from specific photodetectors. Each pipeline processes data independently and outputs to a shared data stream, eliminating the need for centralized buffering and reducing latency while maintaining data aggregation reliability.
Solution Approach 2:
The patent introduces a shared data stream as an intermediary that receives processed data from multiple independent pipelines. This mediator structure allows data to flow continuously without blocking, eliminating the need for traditional shared memory buffers and reducing both latency and storage requirements.
2Reliability
If data from multiple processors is aggregated using traditional methods with shared memory buffers, then data aggregation is achieved, but storage requirements increase
Solution Approach 1:
The patent segments the data aggregation architecture into independent processing pipelines that each handle specific portions of the data stream. This segmentation eliminates the need for large centralized memory buffers, reducing storage requirements while maintaining complete data aggregation through the distributed pipeline structure.
Solution Approach 2:
The shared data stream acts as an intermediary that receives data from multiple pipelines without requiring substantial buffering capacity. This mediator approach allows continuous data flow with minimal storage requirements, contrasting with traditional methods that require large shared memory buffers.
3Ease of operation
If processors are arranged in a specific physical order, then data ordering is simplified, but device complexity increases
Solution Approach 1:
Instead of arranging processors in a specific physical order to achieve data ordering, the patent inverts the approach by using logical ordering within independent pipelines. Each pipeline maintains its own data order, and the shared data stream combines these ordered streams without requiring the processors themselves to be physically ordered, thereby reducing device complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables reliable and efficient data aggregation with reduced latency and storage needs, facilitating improved performance in lidar applications.
Implementation Method 1
A non-limiting example of such an optical receiver could include a particular arrangement of: a) a plurality of photodetectors
Data Source
AI summary
A system may include a common processed data pipeline and a plurality of processors. Outputs of the plurality of processors are communicatively coupled to the common processed data pipeline. Each processor is configured to accept a plurality of input signals, and each of the plurality of input signals represents a light signal detected by a photodetector. Each processor is also configured to process the input signals to provide processed data. Each processor is also configured to output the processed data into one or more predetermined data locations of a data stream of the common processed data pipeline.