Neural Network Reference Signal Pools for Higher Data Rates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems face challenges in efficiently increasing data transmission rates due to limited and poorly distinguishable reference signals, especially in scenarios with a large number of users, leading to restricted system capacity and performance.
Innovation Solution
Utilizing a neural network to train a reference signal pool that can adapt to various channel environments and user features, allowing for the selection of optimal reference signals that enhance data transmission rates by improving signal differentiation and system capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional reference signal allocation methods are used, then system implementation is simple, but data transmission rate is limited due to poor signal distinguishability
Solution Approach 1:
The patent changes the parameters of reference signals by using neural network-generated sequences with varying initialization values, lengths, and scrambling parameters. This creates highly distinguishable reference signals that improve data transmission rate while managing system complexity through parameter optimization rather than structural changes
Solution Approach 2:
The patent replaces traditional mechanical/reference signal generation methods with neural network-based sequence generation. The neural network learns optimal reference signal characteristics from channel conditions and user features, substituting conventional signal design approaches with AI-driven generation that improves transmission performance
2Quantity of substance
If the number of reference signals is increased to support more users, then system capacity improves, but signal distinguishability deteriorates due to resource constraints
Solution Approach 1:
The patent extends reference signal differentiation into multiple dimensions including time-domain shifts, frequency-domain shifts, code-domain spreading, and spatial-domain beamforming. This multi-dimensional approach allows many users to have distinguishable reference signals without increasing the basic reference signal quantity, maintaining signal reliability while supporting high system capacity
Solution Approach 2:
The patent applies local quality by generating user-specific reference signal characteristics through neural network processing of individual user features and channel conditions. Each user receives reference signals optimized for their specific local context, ensuring high distinguishability even with limited overall reference signal resources
3Reliability
If reference signals are optimized for specific channel conditions, then transmission performance improves, but adaptability to diverse channel environments deteriorates
Solution Approach 1:
The patent implements dynamic reference signal optimization through neural networks that continuously adapt to changing channel conditions and user features. The system dynamically generates appropriate reference signal parameters based on real-time inputs, maintaining high transmission performance across diverse and time-varying channel environments rather than being fixed for specific conditions
Data Source
AI summary
Embodiments of this application provide a method for configuring a reference signal, to increase a rate of data transmission between an access network device and a terminal device. The method includes: An access network device determines a first reference signal pool, where the first reference signal pool includes one or more reference signals. The access network device sends first information to a first terminal device, where the first information indicates a first reference signal allocated to the first terminal device, the first reference signal is included in the first reference signal pool, and the first reference signal pool is obtained by the access network device by training a first neural network, or the first reference signal pool is obtained by another network-side node by training a first neural network and sent to the access network device.