MIMO Radio Receiver Using Neural Equalization for Stream Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radio receiver algorithms, particularly for MIMO systems, are labor-intensive to develop and struggle to align with actual channel conditions, leading to suboptimal performance and challenges in separating overlapping spatial streams.
Innovation Solution
A radio receiver employing a machine learning model with a neural network and multiplicative layers to process data arrays, enabling efficient channel estimation and bit/symbol detection, even under complex MIMO conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual algorithms are used for MIMO detection, then implementation is straightforward, but performance is suboptimal and development is labor-intensive
Solution Approach 1:
The patent replaces manual algorithm design and traditional signal processing methods with a machine learning model that automatically learns optimal detection strategies from training data, eliminating the need for labor-intensive manual algorithm development while achieving superior receiver performance
Solution Approach 2:
The patent transforms the receiver from using fixed manually-designed algorithms to using adaptive machine learning models whose parameters are automatically optimized through training on channel data, allowing the system to adapt to different channel conditions without manual reconfiguration
2Measurement precision
If traditional algorithms are designed based on theoretical channel conditions, then implementation is simple, but they do not align well with actual physical channel conditions
Solution Approach 1:
The patent performs preliminary training of the machine learning model using extensive channel data that captures actual physical channel conditions before deployment. This pre-training phase allows the model to learn realistic channel characteristics and patterns that theoretical models cannot capture, improving alignment with actual conditions
Solution Approach 2:
The machine learning model automatically adapts to and learns the specific characteristics of the deployment environment's channel conditions through continuous operation and potential fine-tuning, enabling the system to self-optimize without requiring manual adjustment or complex theoretical modeling
3Productivity
If MIMO detection separates multiple overlapping spatial streams, then capacity increases, but detection complexity and challenges increase
Solution Approach 1:
The patent merges the separation of multiple spatial streams into a unified machine learning model that processes all streams simultaneously through learned transformation operations, avoiding the need for complex sequential separation algorithms while maintaining the capacity benefits of MIMO
Data Source
AI summary
According to an example embodiment, a radio receiver is configured to: obtain a data array including a plurality of elements, wherein each element in the plurality of elements in the data array corresponds to a sub-carrier in a plurality of subcarriers, to a timeslot in a time interval, and to an antenna stream; implement a machine learning model including at least one neural network and a transformation, wherein the transformation includes at least one multiplicative layer or equalisation; and input data into the machine learning model, wherein the data includes at least the data array; wherein the machine learning model is configured to, based on the data, output an output array representing values of the plurality of elements in the data array, wherein the values include bits or symbols. A radio receiver, a method and a computer program product are disclosed.