ML Encoder for Localization Data With Lower Communication Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wireless communication systems face inefficiencies in transmitting and receiving data for localization services, particularly in generating and processing redundant information, which affects the accuracy of location estimation and increases computational costs.
Innovation Solution
An end-to-end communication approach using a machine learning model to encode image-based and sensor-based data into modulation symbols, adding a reference signal, and allocating transmission resources, allowing for joint optimization of the communication and control systems to improve localization accuracy and reduce latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional coding and modulation protocols are used for transmitting localization data, then communication reliability is maintained, but end-to-end service latency increases and localization accuracy deteriorates due to redundant information processing
Solution Approach 1:
The patent extracts and transmits only the essential localization information (encoded modulation symbols) without including redundant data that would require extensive coding and modulation processing. By taking out only the necessary information components, the system reduces processing latency while maintaining localization accuracy, directly resolving the contradiction between measurement precision and time loss.
Solution Approach 2:
The encoder performs preliminary mapping of input data to encoded modulation symbols before transmission, preparing the data in advance in an optimized format. This preliminary action eliminates the need for extensive decoding and modulation processing at the receiver end, reducing service latency while preserving localization accuracy.
2Measurement precision
If comprehensive input data (image-based and sensor-based) is transmitted to improve localization accuracy, then measurement precision improves, but device complexity and processing requirements increase
Solution Approach 1:
The patent merges image-based input data and sensor-based input data into a unified encoded modulation symbol representation. By combining multiple data types into a single integrated format, the system maintains comprehensive information for accurate localization while reducing the complexity of handling separate data streams through conventional coding and modulation protocols.
Solution Approach 2:
The encoder transforms the parameters of input data (image and sensor) into encoded modulation symbols with optimized characteristics for direct transmission. This parameter transformation simplifies the data structure and reduces processing complexity while preserving the essential information needed for accurate localization.
3Productivity
If traditional communication protocols with full coding and modulation are used, then communication reliability is ensured, but productivity and service efficiency decrease due to extensive processing requirements
Solution Approach 1:
The encoder performs preliminary encoding to map input data directly to modulation symbols in an optimized format before transmission. This preliminary action ensures that the transmitted data is already in the correct form for reliable reception, eliminating the need for extensive decoding and modulation processing at the receiver end, thereby improving service efficiency while maintaining communication reliability.
Solution Approach 2:
The system incorporates feedback mechanisms where the receiver sends acknowledgment signals to the transmitter about successful reception. This feedback loop ensures communication reliability by allowing the system to verify successful transmission of encoded modulation symbols, while the streamlined encoding process maintains high service efficiency.
Data Source
AI summary
Disclosed is a method comprising obtaining first input data, that is image-based input data, and second input data, that is sensor-based input data, providing the first input data and the second input data to an encoder, wherein the encoder uses a machine learning model for mapping the first and second input data to encoded modulation symbols, adding a reference signal to the encoded modulation symbols, allocating transmission time interval resource units to the encoded modulation symbols with the reference signal, and transmitting the symbols with the reference signal to a transmission channel.