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

Cascaded structure based deep learning network OFDM signal detection method

A deep learning network and signal detection technology, applied in transmission monitoring, digital transmission systems, electrical components, etc., can solve problems such as poor performance, and achieve the effect of improving network detection performance

Active Publication Date: 2019-04-05
SOUTHEAST UNIV
View PDF3 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] To sum up, when using the deep learning network to solve the OFDM symbol detection problem under the Doppler frequency offset, it often encounters the problem that the network iterates to the saddle point or the local optimum point with poor performance

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
  • Cascaded structure based deep learning network OFDM signal detection method
  • Cascaded structure based deep learning network OFDM signal detection method
  • Cascaded structure based deep learning network OFDM signal detection method

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0101] Simulation example 1: The simulation conditions are shown in Table 1 below:

[0102] Number of subcarriers N

32

The number of deep network layers N L

20

Modulation

QPSK

Batch size

500

Symbol sampling interval (s)

1e-6

Normalized Doppler frequency offset f N

0.16

[0103] Table 1

[0104] image 3 Based on the above conditions, the comparison between the cascaded structure and the pure network and the classical structure is shown. It can be seen that compared with the pure network, the cascaded structure achieves a solution with better performance, so in the high SNR section The performance is still excellent, but the pure network has a flat layer, and there is a gain of nearly 10dB compared with the ordinary connection solution correlation detection.

[0105] Simulation example 2: The simulation conditions are shown in Table 2 below:

[0106]

[0107]

[0108] Table 2

[0109] Figure 4 Based on t...

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 relates to a cascaded structure based deep learning network OFDM signal detection method. Two parts of detection devices are cascaded. Detection data is processed by a zero-forcing decorrelator, and then sent to the deep learning network for processing, and detected data is obtained. The cascaded zero-forcing decorrelator preprocesses the data so that network parameters of the deep learning network can be converged to a solution with highest performance in the training and updating processing; for a multi-subcarrier scene, the cascaded structure with a sliding structure is used to reduce the complexity of realization; and an Adam algorithm is used for training of the deep learning network, a small-batch gradient decrease manner is used, an improved least square function serves as a loss function, hyper-parameters and the loss function are optimized correspondingly aimed at the structure, and the detection network parameters are converged to the solution with higher detection performance.

Description

technical field [0001] The invention relates to a deep learning network OFDM signal detection method based on a cascade structure, belonging to the technical field of wireless mobile communication. Background technique [0002] As the key technology of 4GLTE, OFDM technology can put the original high-speed transmission data into parallel sub-channels and use sub-carriers for transmission through FFT transformation, thereby changing the original transmission rate. Due to the use of sub-channels, OFDM signals are inherently resistant to frequency-selective fading caused by multipath transmission. From the perspective of the frequency domain, OFDM technology uses the smallest frequency interval to achieve orthogonality and maximizes spectral efficiency. OFDM technology is suitable for multiple services, such as highly flexible communication systems such as wireless local area networks, WiMAX, and is also listed as a modulation standard technology and industry specification bas...

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): H04L27/26H04L27/00H04B17/391
CPCH04L27/0014H04L27/2666H04L27/2691H04L27/2695H04L2027/0065H04B17/3912
Inventor 赵春明黄启圣姜明李骁敏
Owner SOUTHEAST UNIV