Pilot Allocation Method for Multi-Antenna Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multi-antenna systems, particularly in Massive MIMO, pilot contamination limits the accuracy of channel estimation due to the limited number of orthogonal pilot sequences, leading to interference from adjacent cells, which restricts the system's capacity and throughput.
Innovation Solution
A pilot allocation method that determines the number of pilots and training cycles based on the number of users in each cell, allocating pilots to different users in other cells during each training cycle to randomize pilot contamination, thereby enabling accurate channel estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If pilot sequences are reused among several cells to improve spectrum utilization, then system capacity increases, but channel estimation accuracy deteriorates due to pilot contamination from adjacent cells
Solution Approach 1:
The patent divides the pilot allocation into multiple training cycles, where each cycle uses a different orthogonal pilot sequence. This segmentation allows the system to reuse pilots across cells while maintaining estimation accuracy by separating the contamination from different cycles.
Solution Approach 2:
The patent implements periodic pilot allocation where pilot sequences are reused in a structured periodic manner across multiple training cycles. This periodicity enables the receiver to distinguish between desired signals and contaminated signals by knowing which pilot sequences should be present in which cells at which times.
2Productivity
If the number of antennas at base station increases to improve spatial resolution and antenna array gain, then data rate increases, but the complexity of pilot management and interference coordination increases
Solution Approach 1:
The patent creates a universal pilot allocation framework that works for any number of antennas and users. The method uses standardized orthogonal pilot sequences and training cycles that can be applied regardless of the specific Massive MIMO system configuration, reducing the complexity of pilot management.
Solution Approach 2:
The patent changes the temporal parameter by introducing multiple training cycles with different pilot allocations. Instead of trying to manage complex spatial parameters with more antennas, the system uses temporal diversification through repeated training cycles to simplify pilot management while maintaining high data rates.
3Productivity
If orthogonal pilot sequences are limited due to coherence interval length, then training efficiency decreases, but if more pilot sequences are used, then pilot contamination increases
Solution Approach 1:
The patent uses periodic repetition of training cycles with different orthogonal pilot sequences. By periodically rotating through available pilot sequences across multiple cells and cycles, the system achieves effective reuse of limited pilot resources while minimizing contamination through the periodic structure that allows interference identification and cancellation.
Solution Approach 2:
The patent performs preliminary pilot allocation planning across multiple training cycles before actual data transmission. By pre-assigning different orthogonal pilot sequences to different cells and training cycles, the system proactively prevents pilot contamination rather than reacting to it, improving training efficiency from the outset.
Data Source
AI summary
A pilot allocation method for a multi-antenna system and a pilot allocation apparatus are provided. The method comprises: determining the number of pilots to be allocated and the number of training cycles at least according to the number of users in each cell, and determining a pilot allocation scheme of the each cell according to the number of pilots to be allocated and the number of training cycles. The determining a pilot allocation scheme of the each cell can comprise determining pilots allocated for users in the l-th cell used during each training cycle, and during each training cycle, allocating pilots of at least the k-th user of the l-th cell to different users in other cells of the system. By allocating pilots allocated to a desired user to different users during each training cycle, pilot contamination can be randomized during several different continuous training cycles, thereby providing a basis for the later accurate channel estimation.