MIMO Virtual Training Signal Block for Delay Wavelength Extension

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In MIMO systems, the delay wavelength of the virtual communication path often exceeds the assumed value due to environmental differences between the transmission and reception sides, leading to incorrect calculation of the reception equalization weight and increased bit error rates. Extending the actual training signal section to account for this can deteriorate transmission capacity.

Innovation Solution

A communication path presuming method that calculates an adjoint matrix as a transmission weight, transmits a training signal beamformed by this weight, generates a virtual training signal block, and calculates a communication path response using a slide correlation scheme, allowing for the extension of the presumed delay wavelength without extending the actual training signal section.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the actual training signal section is extended to account for the delay wavelength, then the delay wavelength can be presumed accurately, but the transmission capacity deteriorates

Engineering Contradiction:
Improvedelay wavelength presumption accuracyVSAvoidtransmission capacity
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent creates a virtual training signal block by copying and concatenating the actual training signal multiple times. This virtual copy has an extended length that covers the maximum delay wavelength, allowing accurate communication path presumption without actually transmitting the extended signal, thus preserving transmission capacity while achieving accurate delay wavelength measurement

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent transitions from the time domain to the frequency domain by performing FFT (Fast Fourier Transform) on both the virtual training signal block and the received signal. This dimensional change allows the communication path response to be calculated in the frequency domain, enabling accurate delay wavelength presumption across the extended virtual signal length without increasing actual transmission time

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Reliability

If the training signal section is sufficiently extended, then the accuracy of reception equalization weight calculation is improved, but the transmission capacity deteriorates

Engineering Contradiction:
Improvereception equalization weight accuracyVSAvoidtransmission capacity
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent generates a virtual training signal block by concatenating multiple copies of the actual training signal. This virtual extension provides sufficient length for accurate communication path presumption across the maximum delay wavelength, enabling reliable reception equalization weight calculation without actually transmitting the extended signal, thus maintaining transmission capacity

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent performs communication path presumption using the virtual training signal block before actual data transmission. By preliminarily calculating the communication path response and reception equalization weight using the extended virtual signal, the system ensures accurate equalization for subsequent data transmission without requiring the actual training signal to be extended

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS12273178B2Communication path presuming method and wireless communication device
Publication Date: 2025.04.08 NIPPON TELEGRAPH & TELEPHONE CORP
  • US12273178B2 patent drawing
  • US12273178B2 patent drawing
  • US12273178B2 patent drawing

AI summary

An adjoint matrix adjH(z, t) of a transfer function matrix H(z, t) established between a transmission station and a reception station is defined as a transmission weight WT(z). A training signal multiplied by the transmission weight WT(z) is transmitted from transmission antennas toward a reception station. M sequence portions S included in the received training signals #1 and #2 are connected in series to generate a virtual training signal block. A communication path response R(m) implemented by using the transmission weight WT(z) is calculated by calculating a correlation between both an M sequence portion S and a comparison sequence portion at each position while sliding the comparison sequence portion with respect to the virtual training signal block. Based on the communication path response R(m), a reception equalization weight WR(z) corresponding to an inverse response det{H(z, t)}−1 of a determinant det{H(z, t)} of the transfer function matrix H(z, t) is calculated.