Compressed Occupancy Grid Aggregation for Low-Overhead OTA Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems face challenges in efficiently aggregating occupancy grids due to resource overheads when multiple user equipments (UEs) contribute their occupancy vectors, especially in dynamic environments with high resolution and frequent updates, limiting the scalability and efficiency of occupancy grid aggregation.
Innovation Solution
The use of compressed sensing techniques to generate a compressed occupancy vector by applying a sensing matrix to the occupancy vector, allowing for reduced resource usage by transmitting fewer elements, and enabling over-the-air aggregation of these vectors across UEs, which are then recovered by a network entity using a common sensing matrix.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple UEs transmit full occupancy vectors for aggregation, then aggregation accuracy is maintained, but resource overhead increases significantly
Solution Approach 1:
The patent extracts and transmits only the essential occupancy information by applying a sensing matrix to the occupancy vector, generating a compressed representation that retains the necessary data for accurate aggregation while removing redundant elements. This extraction process reduces the number of elements that need to be transmitted over the air interface.
Solution Approach 2:
The patent changes the parameter representation of occupancy data by transforming the original occupancy vector into a compressed form through matrix multiplication. This parameter transformation maintains the informational content needed for aggregation while reducing the dimensionality and resource requirements for transmission.
2Measurement precision
If high resolution occupancy grids are used, then detection precision is improved, but transmission and processing complexity increases
Solution Approach 1:
The patent extracts the essential occupancy information from high-resolution grids by applying a sensing matrix that identifies and retains only the critical elements needed for accurate detection. This extraction maintains detection precision while reducing the complexity of transmission and processing by eliminating redundant high-resolution data.
Solution Approach 2:
The patent segments the high-resolution occupancy grid into compressed representations through matrix transformation. This segmentation divides the complex high-resolution data into manageable compressed components that can be transmitted and processed more efficiently while preserving the essential detection information.
3Speed
If occupancy grids are updated frequently, then dynamic environment tracking is improved, but resource consumption increases
Solution Approach 1:
The patent extracts only the necessary occupancy changes and transmits compressed updates rather than full occupancy vectors. This extraction approach enables frequent updates to track dynamic environments effectively while reducing the resource consumption associated with transmitting and processing complete grid data at high frequencies.
Solution Approach 2:
The patent implements periodic compressed sensing and transmission of occupancy updates. This periodic action allows the system to maintain accurate environmental tracking by updating at appropriate intervals while consuming fewer resources compared to continuous transmission of full occupancy vectors.
Data Source
AI summary
Aspects of the present disclosure include methods, apparatuses, and computer-readable medium for generating an occupancy vector based on one or more sensor signals captured by one or more sensors configured to detect a presence of one or more objects in an area, wherein each element of the occupancy vector corresponds to a cell in an occupancy grid that divides the area into a number of cells, wherein each element of the occupancy vector has a value indicating that a respective cell of the occupancy grid is occupied responsive to an object being detected in the respective cell by the one or more sensors; applying a sensing matrix to the occupancy vector to generate a compressed occupancy vector that has fewer elements than the occupancy vector; and transmitting, to a network entity, one or more signals indicative of the compressed occupancy vector.