Research on mine wireless signal detection method based on dual path network
-
摘要: 目前针对矿井无线信号检测的研究大多只考虑了比较理想的加性高斯白噪声信道和瑞利衰落信道，且信号检测误码率高，网络结构复杂。针对上述问题，提出一种基于双路径网络（DPN）的矿井无线信号检测方法，采用双路网络接收机（DPNR）优化正交频分复用（OFDM）接收端的整体性能，解决常规接收机的误差累积问题。首先采用残差（Res）块的shortcut对浅层特征进行一次卷积，将经过一次卷积后的特征图与经过多次卷积后的特征图相加。然后将密集（Dense）块浅层重复利用，并进行Dense块的卷积计算，得到卷积计算后的特征图。最后将两者的特征图融合成新的特征图，在牺牲较少复杂度的情况下，提取更多的特征，从而提高检测性能。 实验结果表明：① 在OFDM系统中，DPNR的误码率比常规接收机低，在信噪比为13时，误码率为零；在信噪比大于7 时，DPNR的误码率较矿井环境下的常规接收机降低1个数量级以上；在信噪比大于11时，DPNR的误码率较加性高斯白噪声下的常规接收机降低 1个数量级以上。② 在通信系统滤波器组多载波/偏置正交幅度调制中，DPNR的误码率较常规接收机的降低2个数量级以上，说明其具有较好的鲁棒性。③ 随着信噪比的增加，DPNR和残差神经网络（ResNet）接收机的误码率较密集连接卷积网络（DenseNet）接收机低，且DPNR的误码率在最后阶段即信噪比大于13时更低。④ 在较高信噪比情况下，DPNR的误码率远远低于深度接收机，在信噪比大于8时，DPNR的误码率较深度接收机降低1个数量级以上。Abstract: At present, most of the research on mine wireless signal detection only considers the ideal additive Gaussian white noise channel and Rayleigh fading channel. The signal detection has high bit error rate and complex network structure. In order to solve the above problems, a mine wireless signal detection method based on dual path network (DPN) is proposed. The method uses dual path network receiver (DPNR) to optimize the overall performance of the orthogonal frequency division multiplexing (OFDM) receiver and solve the problem of error accumulation in conventional receivers. Firstly, the residual (Res) block's shortcut is used to perform a convolution of shallow features, and the feature map after one convolution is added to the feature map after multiple convolutions. Secondly, the shallow layer of the Dense block is reused. The convolution calculation of the Dense block is performed to obtain the feature map after the convolution calculation. Finally, the feature maps of the two are fused into a new feature map, which extracts more features at the expense of less complexity, thereby improving detection performance. The experimental results show the following points. ① In OFDM systems, the bit error rate of DPNR is lower than that of conventional receivers. When the signal-to-noise ratio is 13, the bit error rate is zero. When the signal-to-noise ratio is greater than 7, the error rate of DPNR is reduced by more than one order of magnitude compared to conventional receivers in mine environments. When the signal-to-noise ratio is greater than 11, the bit error rate of DPNR is more than one order of magnitude lower than that of conventional receivers under additive Gaussian white noise. ② In the multi-carrier/offset orthogonal amplitude modulation of communication system filter banks, the error rate of DPNR is reduced by more than two orders of magnitude compared to conventional receivers, indicating its good robustness. ③ As the signal-to-noise ratio increases, the bit error rate of DPNR and residual neural network (ResNet) receivers is lower than that of densely connected convolutional networks (DenseNet) receivers. The bit error rate of DPNR is lower in the final stage when the signal-to-noise ratio is greater than 13. ④ At higher signal-to-noise ratios, the bit error rate of DPNR is much lower than that of deep receivers. When the signal-to-noise ratio is greater than 8, the bit error rate of DPNR is reduced by more than one order of magnitude compared to deep receivers.
-
-
![]()
图 1 不同m值下Nakagami−m模型的概率密度函数
Figure 1. The probability density function of Nakagami-m model under different m values
![]()
图 2 基于深度接收机的OFDM无线通信系统
Figure 2. Deep receiver-based OFDM wireless communication system
![]()
图 9 DPNR与深度接收机的误码性能对比
Figure 9. Comparison of BER performance between DPNR and deep receivers
表 1 DPNR网络节点的卷积设置
Table 1 Convolution settings for the DPNR network nodes
网络块 输出 卷积设置 Cov1 (None,160, 64) [31,64]×1 过渡块 (None,79,128) [5,128]×1 DPN块 (None,79,320) [1,128]×1,[5,128]×1 过渡块 (None,39,64) [5,64]×1 DPN块 (None,39,224) [1,128]×1,[5,128]×2 过渡块 (None,19,64) [5,64]×1 DPN块 (None,19,288) [1,128]×1,[5,128]×3 过渡块 (None,9,64) [5,64]×1 DPN块 (None,9,224) [1,128]×1,[5,128]×2 Cov (None,9,150) [5,150]×1 全连接层 (None,9,150) − 
下载: 导出CSV
表 2 OFDM参数的设置
Table 2 OFDM parameter settings
参数 值 FFT点数 64 CP长度/chip 16 子载波 52 调制 QPSK 信道 Nakagam−m+AWGN
下载: 导出CSV
表 3 网络架构初始参数
Table 3 Initial parameters of the network architecture
参数 值 优化器 Adam 学习率 10−3 批大小 256 训练次数 15 训练数据集数量 320 000 测试数据集数量 320 000
下载: 导出CSV
-
[1] ZHANG Hui,ZHU Mengzhi,LI Xingwang,et al. Very low frequency propagation characteristics analysis in coal mines[J]. IEEE Access,2020(8):95483-95490.
[2] 张帆,李闯,李昊,等. 面向智能矿山与新工科的数字孪生技术研究[J]. 工矿自动化,2020,46(5):15-20. DOI: 10.13272/j.issn.1671-251x.2020040042 ZHANG Fan,LI Chuang,LI Hao,et al. Research on digital twin technology for smart mine and new engineering discipline[J]. Industry and Mine Automation,2020,46(5):15-20. DOI: 10.13272/j.issn.1671-251x.2020040042
[3] ZHENG Shilian,CHEN Shichuan,YANG Xiaoniu. Deepreceiver:a deep learning-based intelligent receiver for wireless communications in the physical layer[J]. IEEE Transactions on Cognitive Communications and Networking,2020,7(1):5-20.
[4] LUONG T V,KO Y,VIEN N A,et al. Deep learning-based detector for OFDM-IM[J]. IEEE Wireless Communications Letters,2019,8(4):1159-1162. DOI: 10.1109/LWC.2019.2909893
[5] ZHAO Zhongyuan,VURAN M C,GUO Fujuan,et al. Deep-waveform:a learned OFDM receiver based on deep complex-valued convolutional networks[J]. IEEE Journal on Selected Areas in Communications,2021,39(8):2407-2420. DOI: 10.1109/JSAC.2021.3087241
[6] 姚善化. 基于镜像法的矿井隧道电磁波多径信道模型[J]. 工矿自动化,2017,43(4):46-49. DOI: 10.13272/j.issn.1671-251x.2017.04.011 YAO Shanhua. Electromagnetic wave multipath channel model based on image method in mine tunnel[J]. Industry and Mine Automation,2017,43(4):46-49. DOI: 10.13272/j.issn.1671-251x.2017.04.011
[7] LIU Boyan, YANG Xiaohui, CHEN Zifeng, et al. The Internet of Things（IoT） system for bolt looseness detection in coal mines[C]. The 3rd International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering （ICBAIE）, Xi'an, 2022: 293-296.
[8] 王安义,李立. 基于高阶累积量和DNN模型的井下信号识别方法[J]. 工矿自动化,2020,46(2):82-87. DOI: 10.13272/j.issn.1671-251x.2019100064 WANG Anyi,LI Li. Underground signal recognition method based on higher-order cumulants and DNN model[J]. Industry and Mine Automation,2020,46(2):82-87. DOI: 10.13272/j.issn.1671-251x.2019100064
[9] HAO Ye,LI G Y,JUANG B H F. Power of deep learning for channel estimation and signal detection in OFDM systems[J]. IEEE Wireless Communications Letters,2018,7(1):114-117. DOI: 10.1109/LWC.2017.2757490
[10] LIU Hongfu,WEI Ziping,ZHANG Hengsheng,et al. Tiny machine learning (Tiny-ML) for efficient channel estimation and signal detection[J]. IEEE Transactions on Vehicular Technology,2022,71(6):6795-6800. DOI: 10.1109/TVT.2022.3163786
[11] YI Xuemei,ZHONG Gaijun. Deep learning for joint channel estimation and signal detection in OFDM systems[J]. IEEE Communications Letters,2020,24(12):2780-2784. DOI: 10.1109/LCOMM.2020.3014382
[12] FELIX A, CAMMERER S, DORNER S, et al. OFDM-autoencoder for end-to-end learning of communications systems[C]. IEEE 19th International Workshop on Signal Processing Advances in Wireless Communications （SPAWC）, Kalamata, 2018: 1-5.
[13] BALEVI E,ANDREWS J G. One-bit OFDM receivers via deep learning[J]. IEEE Transactions on Communications,2019,67(6):4326-4336. DOI: 10.1109/TCOMM.2019.2903811
[14] CHEN Xiaolong, JIANG Qiaowen, SU Ningyuan, et al. LFM signal detection and estimation based on deep convolutional neural network[C]. Asia-Pacific Signal and Information Processing Association Annual Summit and Conference （APSIPA ASC）, Lanzhou, 2019: 753-758.
[15] DONG Yaning, WU Chuanzhang, ZHU Huizhu, et al. A weak signal detection method based on spatial spectrum-LSTM neural network[C]. The 5th International Conference on Information Communication and Signal Processing （ICICSP）, Shenzhen, 2022: 1-6.
[16] NIE Donghu,XIE Kai,ZHOU Feng,et al. A correlation detection method of low SNR based on multi-channelization[J]. IEEE Signal Processing Letters,2020,27:1375-1379. DOI: 10.1109/LSP.2020.3013769
[17] LI Chun, ZHAO Zhijin, CHEN Ying. Detection algorithm of frequency hopping signals based on S transform and deep learning[C]. 16th IEEE International Conference on Signal Processing （ICSP）, Beijing, 2022: 310-313.
[18] GATERA O, SIBOMANA L, LLHAN H, et al. On analysis of signal detection in relays networks over time-varying rayleigh channels[C]. International Conference on Communications, Signal Processing, and their Applications （ICCSPA）, Sharjah, 2019: 1-5.
[19] ZHANG Zhaoming, CHENG Baixiao, YANG Minglei, et al. Target detection of optimum frequencies selection based on time reversal[C]. The 5th International Conference on Frontiers of Signal Processing （ICFSP）, Marseille, 2019: 40-44.
[20] 崔建华,袁正道,王忠勇,等. 基于隐聚类和狄利特雷过程的大规模MIMO-OFDM接收机设计[J]. 电子学报,2019,47(12):2515-2523. CUI Jianhua,YUAN Zhengdao,WANG Zhongyong,et al. Massive MIMO-OFDM receiver design based on hidden cluster hypothesis and dirichlet process[J]. Acta Electronica Sinica,2019,47(12):2515-2523.
[21] YANG Aiping, WANG Haixin, JI Zhong, et al. Dual-path in dual-path network for single image dehazing[C]. Teenty-Eighth International Joint Conference on Artificial Intelligence, 2017: 4627-4634.
-
期刊类型引用(3)
1. 雍明超，王震，金士魁，冯俊文. 艾维尔沟矿区坚硬顶板回采巷道围岩结构特征分析. 陕西煤炭. 2022(02): 32-36+57 . 
百度学术
2. 周海龙. 单元型巷道超前液压支架的研制. 煤矿机械. 2020(11): 54-55 . 百度学术
3. 苏志毅. 矩形巷道无支护状态下巷道围岩变形数值模拟分析. 煤炭与化工. 2019(05): 42-44+64 . 百度学术
其他类型引用(7)
计量
- 文章访问数: 366
- HTML全文浏览量: 70
- PDF下载量: 14
- 被引次数: 10
