Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jan 23;23(3):1302.
doi: 10.3390/s23031302.

Hardware-Based Architecture for DNN Wireless Communication Models

Van Duy Tran  1   2 , Duc Khai Lam  1   2 , Thi Hong Tran  3
Affiliations

Hardware-Based Architecture for DNN Wireless Communication Models

Van Duy Tran et al. Sensors (Basel). .

Abstract

Multiple Input Multiple Output Orthogonal Frequency Division Multiplexing (MIMO OFDM) is a key technology for wireless communication systems. However, because of the problem of a high peak-to-average power ratio (PAPR), OFDM symbols can be distorted at the MIMO OFDM transmitter. It degrades the signal detection and channel estimation performance at the MIMO OFDM receiver. In this paper, three deep neural network (DNN) models are proposed to solve the problem of non-linear distortions introduced by the power amplifier (PA) of the transmitters and replace the conventional digital signal processing (DSP) modules at the receivers in 2 × 2 MIMO OFDM and 4 × 4 MIMO OFDM systems. Proposed model type I uses the DNN model to de-map the signals at the receiver. Proposed model type II uses the DNN model to learn and filter out the channel noises at the receiver. Proposed model type III uses the DNN model to de-map and detect the signals at the receiver. All three model types attempt to solve the non-linear problem. The robust bit error rate (BER) performances of the proposed receivers are achieved through the software and hardware implementation results. In addition, we have also implemented appropriate hardware architectures for the proposed DNN models using special techniques, such as quantization and pipeline to check the feasibility in practice, which recent studies have not done. Our hardware architectures are successfully designed and implemented on the Virtex 7 vc709 FPGA board.

Keywords: MIMO; OFDM; artificial intelligence; deep learning; hardware design.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
MIMO OFDM system.
Figure 2
Figure 2
General DNN structure.
Figure 3
Figure 3
Proposed Model Type I.
Figure 4
Figure 4
Proposed Model Type II.
Figure 5
Figure 5
Proposed Model Type III.
Figure 6
Figure 6
General neural network hardware architecture.
Figure 7
Figure 7
Layer data calculation block.
Figure 8
Figure 8
Illustration of matrix computation block.
Figure 9
Figure 9
The simplified approximate Sigmoid algorithm.
Figure 10
Figure 10
The simplified Sigmoid hardware design.
Figure 11
Figure 11
Layer data calculation block with quantization technique.
Figure 12
Figure 12
BER versus SNR for the conventional receivers with M = 2.
Figure 13
Figure 13
BER versus SNR for the proposed receivers with M = 2, ZF is used and the clipping level of the nonlinear PAs is 5 dB.
Figure 14
Figure 14
BER versus SNR for the proposed receivers with M = 2, ZF is used and the clipping level of the nonlinear PAs is 7 dB.
Figure 15
Figure 15
BER versus SNR for the original receivers with M = 4.
Figure 16
Figure 16
BER versus SNR for the proposed receivers with M = 4, ZF is used and the clipping level of the nonlinear PAs is 5 dB.
Figure 17
Figure 17
BER versus SNR for the proposed receivers with M = 4, ZF is used and the clipping level of the nonlinear PAs is 7 dB.
Figure 18
Figure 18
BER versus SNR for the proposed receivers with M = 4, MLD is used and the clipping level of the nonlinear PAs is 5 dB.
Figure 19
Figure 19
BER versus SNR for the proposed receivers with M = 4, MLD is used and the clipping level of the nonlinear PAs is 7 dB.
Figure 20
Figure 20
BER versus SNR for the receiver hardware architectures with M = 2, ZF is used and the clipping level of the nonlinear PAs is 5 dB.
Figure 21
Figure 21
BER versus SNR for the receiver hardware architectures with M = 2, ZF is used and the clipping level of the nonlinear PAs is 7 dB.
Figure 22
Figure 22
BER versus SNR for the receiver hardware architectures with M = 4, ZF is used and the clipping level of the nonlinear PAs is 5 dB.
Figure 23
Figure 23
BER versus SNR for the receiver hardware architectures with M = 4, ZF is used and the clipping level of the nonlinear PAs is 7 dB.
Figure 24
Figure 24
BER versus SNR for the receiver hardware architectures with M = 4, MLD is used and the clipping level of the nonlinear PAs is 5 dB.
Figure 25
Figure 25
BER versus SNR for the receiver hardware architectures with M = 4, MLD is used and the clipping level of the nonlinear PAs is 7 dB.
See this image and copyright information in PMC

References

    1. Hammed Z.S., Ameen S.Y., Zeebaree S.R.M. Massive MIMO-OFDM Performance Enhancement on 5G; Proceedings of the 2021 International Conference on Software, Telecommunications and Computer Networks (SoftCOM); Hvar, Croatia. 23–25 September 2021; pp. 1–6. - DOI
    1. Riadi A., Boulouird M., Hassani M.M. Least Squares Channel Estimation of an OFDM Massive MIMO System for 5G Wireless Communications; Proceedings of the 8th International Conference on Sciences of Electronics, Technologies of Information and Telecommunications (SETIT’18); Genoa, Italy. 18–20 December 2018; pp. 440–450.
    1. He H., Wen C.K., Jin S. Bayesian Optimal Data Detector for Hybrid mmWave MIMO-OFDM Systems With Low-Resolution ADCs. IEEE J. Sel. Top. Signal Process. 2018;12:469–483. doi: 10.1109/JSTSP.2018.2818063. - DOI
    1. Wang B., Jian M., Gao F., Li G.Y., Lin H. Beam Squint and Channel Estimation for Wideband mmWave Massive MIMO-OFDM Systems. IEEE Trans. Signal Process. 2019;67:5893–5908. doi: 10.1109/TSP.2019.2949502. - DOI
    1. Zhang W., Yin Q., Gao F. Computationally Efficient Blind Estimation of Carrier Frequency Offset for MIMO-OFDM Systems. IEEE Trans. Wirel. Commun. 2016;15:7644–7656. doi: 10.1109/TWC.2016.2605678. - DOI

LinkOut - more resources