Multiple-Antenna Interference Cancellation in WIMAX Networks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing multiple-antenna interference cancellation methods are ineffective in unsynchronized backhaul and cellular WIMAX-based networks due to asynchronous interference scenarios, where the number of interferers exceeds the number of antenna elements, leading to performance degradation and inefficiency in canceling co-channel interference (CCI) components.

Innovation Solution

A receiver system that estimates two sets of antenna weights from different training data sets and uses space-frequency filters to generate symbol candidates, with a selector choosing the best estimate based on quality measures, allowing adaptation to changing interference conditions and improving interference cancellation efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If stationary training-based processing is used to estimate antenna weights over all preambles, then the system can process the entire frame with a single weight vector, but the interference cancellation efficiency deteriorates when the set of CCI components changes between training and data intervals

Engineering Contradiction:
Improveprocessing simplicityVSAvoidinterference cancellation efficiency
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent divides the frame into multiple segments (first training interval, second training interval, and data interval), and estimates antenna weights separately for each segment. This segmentation allows the system to adapt to changing interference conditions in different time intervals, resolving the contradiction between processing simplicity and interference cancellation efficiency.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from static weight estimation over the entire frame to dynamic weight estimation that adapts to changing interference conditions. By estimating weights separately for different training intervals and selecting the most appropriate ones for data intervals, the system dynamically adjusts to varying CCI components, improving reliability while maintaining operational feasibility.

Inventive Principle:
Principle #15Dynamics

2Productivity

If the total number of interferers exceeds the number of antenna elements, then the system cannot cancel all CCI components simultaneously, but reducing the number of estimated weights improves processing efficiency

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidinterference cancellation performance
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent applies partial action by estimating antenna weights only for specific training intervals where interference conditions are favorable, rather than attempting to cancel all interferers simultaneously. This selective approach processes fewer weights at a time, improving processing efficiency while maintaining reliable interference cancellation for the actual data intervals.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The patent performs preliminary weight estimation during training intervals before the actual data transmission. By preparing weight vectors in advance during dedicated training periods, the system reduces the computational burden during data processing while ensuring accurate interference cancellation is ready when needed.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If different sets of CCI components are present in different data intervals, then a single weight vector cannot effectively cancel interference for all intervals, but using multiple weight vectors increases system complexity

Engineering Contradiction:
Improveinterference cancellation accuracyVSAvoidsystem structure complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the weight estimation process into multiple independent estimations for different training intervals. Each segment produces a weight vector tailored to the interference conditions of that specific interval, allowing accurate cancellation without requiring a single complex adaptive system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces training intervals as intermediary periods dedicated to weight estimation. These intermediaries provide controlled environments for measuring channel conditions and calculating appropriate weight vectors, which are then applied to subsequent data intervals. This mediation simplifies the overall system by separating the complex adaptation task into manageable discrete steps.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP2040390B1Multiple-antenna interference cancellation
Publication Date: 2010.04.14 LUCENT TECH INC
  • EP2040390B1 patent drawingFigure 1~2
  • EP2040390B1 patent drawingFigure 3~4
  • EP2040390B1 patent drawingFigure 5~6

AI summary

The present invention relates to multiple-antenna interference cancellation, more particularly to a receiver for wirelessly transmitted digital data bursts and a method for multiple-antenna interference cancellation for wirelessly received digital data bursts. The invention is particular useful for multiple-antenna interference cancellation in unsynchronized backhaul and/or cellular WIMAX-based networks. A receiver (10) for wirelessly transmitted digital data bursts including training data for wireless channel estimation is provided. The receiver comprises: a plurality of antenna elements for receiving input data; a first estimator (13, 14, 15) for estimating a first set of antenna weights for the antenna elements from first training data; a second estimator (13, 14, 15) for estimating a second set of antenna weights for the antenna elements from second training data; a first space-frequency filter (17, 18, 19) for providing a first symbol candidate for the input data based on the first set of antenna weights; a second space-frequency filter (17, 18, 19) for providing a second symbol candidate for the input data based on the second set of antenna weights; and a selector (20) for selecting, as data symbol estimate, the first symbol candidate or the second symbol candidate.