Model-driven channel modeling approach in laminar channel
WANG Yue1, BAO Xu1, LIN Feng2
1. School of Computer Science and Communication Engineering, Jiangsu University, Zhenjiang, Jiangsu 212013, China; 2. Key Laboratory of Healthy Freshwater Aquaculture of Ministry of Agriculture, Zhejiang Institute of Freshwater Fisheries, Huzhou, Zhejiang 313001, China
摘要 针对已有的层流信道模型不能直接应用于存在目标的复杂层流信道的问题，提出一种基于模型驱动的信道建模方法.研究了存在目标的层流扩散信道的系统模型，在无目标平流模型的基础上加入参数，考虑层流和目标对接收分子的影响.结合仿真结果，将有目标的复杂层流信道近似为两个稳定的层流信道，建立有目标的点源-接收机层流扩散信道模型.结合神经网络使用Levenberg-Marquardt 算法对信道模型参数进行学习和预测，同时提出基于数据和模型驱动结合（combination of data and model driven，CDMD）的检测方法对目标进行检测.结果表明：通过公式数据与仿真数据对比验证了其信道模型的准确性，所有数据的相关系数为0.999 15，该神经网络模型具有可行性；使用神经网络二分类算法验证提出的目标检测方法，检测准确率达到98.8％时，提出的CDMD检测方法所需数据量约为基于数据检测方法的1/6.
Abstract：To solve the problem that the existing laminar channel models could not be directly applied to the complex laminar channels with targets, a model-driven channel modeling approach was proposed. The system model for laminar diffusion channels with targets was extended by incorporating additional parameters in the absence of target advection models, and the effects of laminar flow and targets on received molecules were considered. Based on the simulation results, the complex laminar channel with target was approximated as two steady laminar channels, and the point-source-receiver laminar diffusion channel model with considering the presence of target was established. The Levenberg-Marquardt algorithm was employed to learn and predict channel model parameters by the neural network, and the combination of data and model driven (CDMD) detection method was proposed for target identification. The results show that the accuracy of the channel model can be validated through the comparison of formula data and simulated data with correlation coefficient of 0.999 15, which confirms the feasibility of the neural network model. The proposed target detection method can be verified by the binary classification algorithm within the neural network with detection accuracy rate of 98.8%. The CDMD-based detection method requires approximately one-sixth of the data volume needed for data-driven detection methods for maintaining high detection performance.
AGHABABAIYAN K, KEBRIAEI H, SHAH-MANSOURI V, et al. Enhanced modulation for multi-users molecular communication in internet of nano things [J]. IEEE Internet of Things Journal, 2022,9(20):19787-19802.
［2］
LI Y, LIN L, GUO W S, et al. Error performance and mutual information for IoNT interface system [J]. IEEE Internet of Things Journal,2022,9(12):9831-9842.
［3］
CHENG Z, TU Y C, XIA M, et al. Energy efficiency analysis of multi-hop mobile diffusive molecular communi-cation [J]. Nano Communication Networks, DOI:10.1016/j.nancom.2020.100313.
［4］
LIU Y, WANG H Y, CAI L, et al. Fundamentals and advancements of topology discovery in underwater acoustic sensor networks: a review［J］. IEEE Sensors Journal, 2021,21(19):21159-21174.
［5］
GUO W S, DENG Y S, LI B, et al. Eavesdropper loca-lization in random walk channels［J］. IEEE Communications Letters, 2016, 20(9):1776-1779.
［6］
KHAN T, BILGIN B A, AKAN O B. Diffusion-based model for synaptic molecular communication channel［J］. IEEE Transactions on Nanobioscience,2017,16(4):299-308.
［7］
GENC G, KARA Y E, TUGCU T, et al. Reception modeling of sphere-to-sphere molecular communication via diffusion［J］. Nano Communication Networks, 2018, 16:69-80.
［8］
KHALOOPOUR L, MIRMOHSENI M, NASIRI-KENARI M. Theoretical concept study of cooperative abnormality detection and localization in fluidic-medium molecular communication [J]. IEEE Sensors Journal, 2021,21(15):17118-17130.
［9］
JAMALI V, AHMADZADEH A, WICKE W, et al. Channel modeling for diffusive molecular communication-a tutorial review [J]. Proceedings of the IEEE, 2019,107(7):1256-1301.
［10］
JIN Z G, DING M G, LUO Y M, et al. Integrated time synchronization and multiple access protocol for underwater acoustic sensor networks [J]. IEEE Access, 2019,7:101844-101854.
［11］
ZHU D C, BAO X, ZHANG W C. Target recognition in turbulent diffusion channels [J]. IEEE Communications Letters, 2021,25(11):3694-3698.
［12］
ABBASZADEH M, YILMAZ H B, THOMAS P J, et al. Linearity of sequential molecular signals in turbulent diffusion channels [C]∥Proceedings of the 2019 IEEE International Conference on Communications. Piscataway：IEEE, DOI:10.1109/ICC.2019.8761812.
［13］
BAO X, SHEN Q F, ZHU Y F, et al. Relative localization for silent absorbing target in diffusive molecular communication system [J]. IEEE Internet of Things Journal, 2022,9(7):5009-5018.
LI K Y, ZHAO B, WANG Y P, et al. The integral method for turbulent momentum boundary layers on a porous plate with uniform blowing and suction [J]. Journal of Sichuan University（Natural Science Edition）, 2021，58（2）：155-160.(in Chinese)
2024, Vol. 45 
