Unlock instant, AI-driven research and patent intelligence for your innovation.

Method and device for realizing matrix QR decomposition with low complexity

A QR decomposition, low-complexity technology, applied in digital transmission systems, electrical components, error prevention, etc., can solve the problems of high division resource overhead, waste of power consumption resources, inconvenient use, etc., to waste power consumption and avoid calculation , The effect of convenient hardware implementation

Inactive Publication Date: 2020-10-09
上海擎昆信息科技有限公司
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in order to ensure performance, the number of iterations is large, and a fixed number of iterations is used for all data, which consumes a lot of power
[0005] 3. Since there are multiple combinations of receiving antennas and layers m*n, it is necessary to implement multiple sets of QR decomposition, which wastes resources
[0006] 4. For the equation y=Hx, first do the QR decomposition of H, and then solve the equation. It is necessary to do the same operation on a unit matrix I while doing the QR decomposition of H to output the Q matrix, and it is necessary to calculate (Q^ T)y, resulting in waste of power consumption and resources
[0007] 5. For the application scenario of solving equations, the output R matrix generally needs to be solved again. When solving, division is required, and the resource overhead of division is large, which makes it inconvenient to use the result of QR

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and device for realizing matrix QR decomposition with low complexity
  • Method and device for realizing matrix QR decomposition with low complexity
  • Method and device for realizing matrix QR decomposition with low complexity

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] Embodiment 1: refer to Figure 1-4 : A low-complexity method for realizing matrix QR decomposition, including the following steps:

[0038] Assuming that the maximum receiving antenna of the system is m_max, the maximum number of layers is n_max, m_max>=n_max, the implementation steps are as follows:

[0039] Step 1: Obtain complex channel estimation matrix H and complex received signal vector y, H contains m rows and n columns of complex data, where m>=n, y contains m rows and 1 column of complex signals, forming an augmented matrix H1=[H y] , H1 contains m rows and n+1 columns. Expand the complex matrix H1 into a complex matrix H2 with m_max rows and m_max+1 columns, and expand by filling 0. Initialize the m_max row m_max+1 column matrix RR to be all zeros, the number of CORDIC iterations is 16, die_para, dieabspara, div_en.

[0040] Step 2: For the matrix H2, go through the first CORDIC iterative unit from the first row, and each CORDIC iterative unit has the same...

Embodiment 2

[0052] Embodiment 2: refer to Figure 1-4 : A low-complexity method for realizing matrix QR decomposition, including the following steps:

[0053] Assuming that the maximum receiving antenna of the system is m_max, the maximum number of layers is n_max, m_max>=n_max, the implementation steps are as follows:

[0054] Step 1: Obtain complex channel estimation matrix H and complex received signal vector y, H contains m rows and n columns of complex data, where m>=n, y contains m rows and 1 column of complex signals, forming an augmented matrix H1=[H y] , H1 contains m rows and n+1 columns. Expand the complex matrix H1 into a complex matrix H2 with m_max rows and m_max+1 columns, and expand by filling 0. Initialize the m_max row m_max+1 column matrix RR to be all zeros, the number of CORDIC iterations is 16, die_para, dieabspara, div_en.

[0055] Step 2: For the matrix H2, go through the first CORDIC iterative unit from the first row, and each CORDIC iterative unit has the same...

Embodiment 3

[0068] Embodiment 3: refer to Figure 1-4 : A low-complexity method for realizing matrix QR decomposition, including the following steps:

[0069] Assuming that the maximum receiving antenna of the system is m_max, the maximum number of layers is n_max, m_max>=n_max, the implementation steps are as follows:

[0070] Step 1: Obtain complex channel estimation matrix H and complex received signal vector y, H contains m rows and n columns of complex data, where m>=n, y contains m rows and 1 column of complex signals, forming an augmented matrix H1=[H y] , H1 contains m rows and n+1 columns. Expand the complex matrix H1 into a complex matrix H2 with m_max rows and m_max+1 columns, and expand by filling 0. Initialize the m_max row m_max+1 column matrix RR to be all zeros, the number of CORDIC iterations is 16, die_para, dieabspara, div_en.

[0071] Step 2: For the matrix H2, go through the first CORDIC iterative unit from the first row, and each CORDIC iterative unit has the same...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a method and a device for realizing matrix QR decomposition with low complexity. The method comprises the following steps of: S1, setting a maximum receiving antenna of a systemas m_max, setting a maximum layer number as n_max, and setting m_max>=n_max; and S2, acquiring a complex channel to obtain a matrix H and a complex received signal vector y, wherein H comprises m rows and n columns of complex data, m>=n, and y comprises m rows and 1 column of complex signals. The method adopts a QR decomposition method of Givens rotation, is implemented by using CORDIC, is low inresource consumption, facilitates hardware implementation, carries out QR decomposition by using a brightness enhancement matrix, avoids output of a Q matrix and calculation of (Q^T)y at the cost ofincreasing the computation amount of one column of data, and expands an arbitrary m*(n+1)(m>=n) matrix into an m_max*(m_max+) matrix, so that a set of QR decomposition scheme is applicable to arbitrary m*(n+1) matrix QR decomposition.

Description

technical field [0001] The invention relates to the technical field of wireless communication, in particular to a method and device for realizing matrix QR decomposition with low complexity. Background technique [0002] Matrix QR decomposition is a commonly used technique in multi-signal detection systems. There are many ways to implement QR decomposition, commonly used are Householder QR decomposition, Gram-Schmidt method QR decomposition and QR decomposition based on Givens rotation. The prior art is based on all three approaches. But there are the following disadvantages: [0003] 1. Householder QR decomposition, Gram-Schmidt method QR decomposition requires the number of multipliers and dividers, and the bit width is relatively large, which consumes resources. [0004] 2. The QR decomposition of Givens rotation can be realized by CORDIC (coordinate rotation digital calculation method), which can be realized only by addition, shift and multiplication, which is conveni...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): H04L1/00
CPCH04L1/0047H04L1/005
Inventor 谭定富
Owner 上海擎昆信息科技有限公司