Two-Phase Device Grouping for MIMO Signaling Overhead
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current user communication device grouping in MIMO systems faces high signaling overhead and computational complexity due to large channel covariance matrices, which hinders efficient communication network management as the number of devices grows.
Innovation Solution
Implementing a two-phase user communication device grouping mechanism where devices self-organize into groups via device-to-device communication, reducing feedback overhead and computational complexity by performing initial grouping locally before the network device clusters these groups.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If centralized processing is used at the BS for user communication device grouping, then grouping can be performed, but signaling overhead becomes huge due to feedback of complete covariance matrices
Solution Approach 1:
The patent segments the grouping process into two phases: a first phase where user communication devices perform local grouping based on covariance matrices of neighbor devices, and a second phase where the BS performs clustering on the groups formed in the first phase. This segmentation reduces the amount of data that needs to be feedback and processed centrally, thereby reducing signaling overhead while maintaining grouping functionality.
Solution Approach 2:
The patent applies preliminary action by having user communication devices perform initial grouping before the BS intervenes. The user devices first group themselves based on local covariance matrix information, and then the BS receives these pre-formed groups for further processing. This preliminary action reduces the computational burden and signaling requirements at the BS.
2Extent of automation
If centralized processing of all covariance matrices is performed, then grouping can be executed, but computational complexity and processing resources requirements increase significantly
Solution Approach 1:
The patent divides the computational task into segments: user communication devices compute and process covariance matrices of their neighbor devices locally during the first phase, while the BS only processes the aggregated group information during the second phase. This segmentation significantly reduces the computational complexity at the BS compared to processing all individual device covariance matrices centrally.
Solution Approach 2:
The patent implements preliminary action by having user devices perform the computationally intensive task of analyzing covariance matrices and forming groups before the BS needs to intervene. This preliminary computation reduces the amount of data and complexity the BS must handle in the second phase.
3Extent of automation
If centralized processing is used, then grouping can be performed, but processing speed decreases and network management becomes difficult as the number of devices grows
Solution Approach 1:
The patent segments the processing workload between user communication devices and the BS. User devices perform local grouping operations independently and in parallel, while the BS performs clustering on the aggregated group data. This segmentation enables concurrent processing across multiple devices, significantly improving overall processing speed and network management capability as the number of devices increases.
Solution Approach 2:
The patent applies preliminary action by having user devices complete their grouping computations before the BS needs to process the results. This time-separated approach allows user devices to work on local grouping independently, improving overall processing throughput and reducing the bottleneck at the BS.
4Ease of manufacture
If existing grouping algorithms are used without exploiting proximity information, then algorithms can be implemented, but local optimum is poor due to bad initial points in iteration
Solution Approach 1:
The patent applies local quality by having each user communication device perform grouping based on local covariance matrix information and proximity to neighbor devices. This local-based approach ensures that devices are grouped according to their physical proximity and local channel characteristics, providing good initial points for the algorithm iteration and improving the quality of local optima.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
The present invention is directed to a user communication device and to a communication network device, both arranged to support grouping of user communication devices in a communication network. The user communication device determines a neighbor user communication device set comprising the user communication device and neighbor user communication devices and determines, by communicating with neighbor user communication devices of the neighbor user communication device set via D2D communication, a user communication device group that the user communication device joins. The communication network device receives information on a plurality of user communication device groups, determined by user communication devices in the communication network, and executes clustering on the plurality of user communication device groups.