Radar Data Autoencoder Compression for Lower Memory and Latency

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Radar systems require significant memory and processing resources due to the large volume of data generated by multiple sensors, leading to increased costs and latency.

Innovation Solution

Implementing a neural network-based data compression and decompression system using multi-layer perceptron architectures to compress and decompress radar data, reducing the memory and latency requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If traditional radar data processing is used, then data processing capability is maintained, but memory requirements and latency increase significantly

Engineering Contradiction:

Improvememory requirementsVSAvoiddata processing capability

Core Design Contradiction:

Quantity of substanceVSProductivity

Solution Approach 1:

The patent extracts only the essential information from the radar data cube by feeding it into a neural network, which processes and compresses the data to retain only the most important features for target detection, thereby reducing memory requirements while maintaining processing capability

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent creates a compressed representation (copy) of the radar data through neural network processing, which preserves the critical information needed for target detection while significantly reducing the data volume stored in memory

Inventive Principle:

Principle #26Copying

2Quantity of substance

If neural network compression is applied, then memory needs are reduced, but processing time for compression and decompression is added

Engineering Contradiction:

Improvememory needsVSAvoidlatency

Core Design Contradiction:

Quantity of substanceVSLoss of time

Solution Approach 1:

The patent performs compression and decompression operations in advance of the actual target detection process, pre-processing the radar data to reduce its size before analysis, thereby minimizing the impact on overall system latency

Inventive Principle:

Principle #10Preliminary action

3Productivity

If multi-layer perceptron architecture is used, then data compression efficiency improves, but system complexity increases

Engineering Contradiction:

Improvedata compression efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent replaces traditional mechanical data compression methods with a neural network-based approach that uses learned patterns and representations, achieving more efficient compression while the complexity is managed through the intelligent processing capabilities of the network

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20250272881A1Autoencoder for radar data compression

Publication Date: 2025.08.28 INFINEON TECHNOLOGIES AG

AI summary

A radar system includes a radar transmitter to transmit a transmit radar signal into a field. A radar receiver receives a receive radar signal in response to the transmit radar signal and generates received radar data based on the receive radar signal. A neural network compression logic is coupled to the radar receiver and has a multi-layer perceptron architecture. The neural network compression logic is trained to compress the received radar data to generate a compressed radar cube. A memory is coupled to the neural network compression logic and is configured to store the compressed radar cube. A neural network de-compression logic is coupled to the memory. The neural network de-compression logic is trained to de-compress the compressed radar cube to generate de-compressed radar data. A target detection logic is configured to detect whether a target is present in the field based on the de-compressed radar data.