Signal Processing Device for Multi-Sensor Data Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing signal processing systems for camera-mounted equipment face limitations due to CPU performance and increased circuit scale and power consumption when sharing signal processing across multiple sensors, particularly when using a DFP dedicated to data flow.
Innovation Solution
A signal processing device and method that includes multiple input units adding additional information to data from sensors, allowing common signal processing across multiple stages of processing units, thereby bypassing the need for CPU or DFP control and reducing circuit scale and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a CPU or DFP is used to share signal processing across multiple sensors, then signal processing capability is improved, but circuit scale and power consumption increase
Solution Approach 1:
The signal processing function is segmented into two parts: (1) packetization and routing control performed by the CPU at lower frequency, and (2) actual signal processing performed by dedicated hardware blocks (demosaicing unit, noise reduction unit, etc.) that operate independently. This segmentation allows the CPU to be scaled down while maintaining overall processing capability through the distributed hardware architecture.
Solution Approach 2:
A packet-based data structure is introduced as an intermediary between the CPU and signal processing hardware. The packet contains metadata (routing information, processing parameters) that enables hardware blocks to autonomously process data without continuous CPU intervention. This intermediary mechanism decouples the CPU from real-time processing demands, reducing its performance requirements and power consumption.
2Productivity
If a DFP dedicated to data flow is used for signal processing, then processing efficiency is improved, but circuit scale increases
Solution Approach 1:
Multiple signal processing hardware blocks (demosaicing unit, noise reduction unit, color interpolation unit, etc.) are designed with universal interfaces that accept packetized input data with varying metadata. These blocks can process data from different sensor types (RGB, monochrome, depth) and perform multiple functions depending on the packet contents, eliminating the need for separate dedicated hardware for each function and reducing overall circuit scale.
Solution Approach 2:
The signal processing architecture is made dynamic through packet-based routing and configurable processing parameters. The CPU dynamically configures processing blocks by inserting appropriate metadata in packets, and data flows dynamically between blocks based on processing requirements. This dynamic configuration allows a fixed hardware architecture to adapt to different processing needs without requiring additional circuitry for each configuration.
3Productivity
If CPU performance is increased to handle multiple sensors, then signal processing capability is improved, but power consumption increases
Solution Approach 1:
Data from multiple sensors is pre-packaged with all necessary metadata (routing information, processing parameters, sensor identifiers) before reaching the signal processing hardware. This preliminary action of packetization enables the hardware blocks to process data autonomously without requiring the CPU to be high-performance for real-time control decisions, thereby reducing CPU power consumption while maintaining multi-sensor processing capability.
Data Source
AI summary
A signal processing device according to the present disclosure includes: multiple input units that add additional information necessary for signal processing to each of multiple pieces of data inputted from respective multiple external devices, and output the multiple pieces of data; and multiple stages of processing units each configured to perform common signal processing on each of the multiple pieces of data, on the basis of the additional information.