Stochastic Chemical Reaction Networks for MAP Detection in Cellular Receivers

被引:2
作者
Heinlein, Bastian [1 ]
Brand, Lukas [1 ]
Egan, Malcolm [2 ]
Schaefer, Maximilian [1 ]
Schober, Robert [1 ]
Lotter, Sebastian [1 ]
机构
[1] Friedrich Alexander Univ Erlangen Nurnberg, Erlangen, Germany
[2] Univ Lyon, INSA Lyon, Inria, CITI, Villeurbanne, France
来源
PROCEEDINGS OF THE 10TH ACM INTERNATIONAL CONFERENCE ON NANOSCALE COMPUTING AND COMMUNICATION, NANOCOM 2023 | 2023年
关键词
Molecular communication; Boltzmann machine; chemical reaction network; maximum-a-posteriori detection; machine learning;
D O I
10.1145/3576781.3608709
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
In order to fully exploit the potential of molecular communication (MC) for intra-body communication, practically implementable cellular receivers are an important long-term goal. A variety of receiver architectures based on chemical reaction networks (CRNs) and gene-regulatory networks (GRNs) has been introduced in the literature, because cells use these concepts to perform computations in nature. However, practical feasibility is still limited by stochastic fluctuations of chemical reactions and long computation times in GRNs. Therefore, in this paper, we propose two receiver designs based on stochastic CRNs, i.e., CRNs that perform computations by exploiting the intrinsic fluctuations of chemical reactions with very low molecule counts. The first CRN builds on a recent result from chemistry that showed how Boltzmann machines (BMs), a commonly used machine learning model, can be implemented with CRNs. We show that BMs with optimal parameter values and their CRN implementations can act as maximum-a-posteriori (MAP) detectors. Furthermore, we show that BMs can be efficiently trained from simulation data to achieve close-to-MAP performance. While this approach yields a fixed CRN once deployed, our second approach based on a manually designed CRN can be trained with pilot symbols even within the cell and thus adapt to changing channel conditions. We extend the literature by showing that practical robust detectors can achieve close-to-MAP performance even without explicit channel knowledge.
引用
收藏
页码:65 / 71
页数:7
相关论文
共 15 条
[1]   On reaction network implementations of neural networks [J].
Anderson, David F. ;
Joshi, Badal ;
Deshpande, Abhishek .
JOURNAL OF THE ROYAL SOCIETY INTERFACE, 2021, 18 (177)
[2]  
Balasubramaniam S., 2022, arXiv
[3]   Designing Molecular Circuits for Approximate Maximum a Posteriori Demodulation of Concentration Modulated Signals [J].
Chou, Chun Tung .
IEEE TRANSACTIONS ON COMMUNICATIONS, 2019, 67 (08) :5458-5473
[4]   Biological Circuits for Detection in MoSK-Based Molecular Communication [J].
Egan, Malcolm ;
Duong, Trung Q. ;
Di Renzo, Marco .
IEEE ACCESS, 2019, 7 :21094-21102
[5]   A Comprehensive Survey of Recent Advancements in Molecular Communication [J].
Farsad, Nariman ;
Yilmaz, H. Birkan ;
Eckford, Andrew ;
Chae, Chan-Byoung ;
Guo, Weisi .
IEEE COMMUNICATIONS SURVEYS AND TUTORIALS, 2016, 18 (03) :1887-1919
[6]   Implementation issues of diffusion-based molecular communications receivers based on transcriptional elements [J].
Femminella, Mauro ;
Reali, Gianluca .
DIGITAL SIGNAL PROCESSING, 2022, 124
[7]  
Goodfellow I, 2016, ADAPT COMPUT MACH LE, P1
[8]  
Haselmayr W., 2019, P 6G WIR SUMM LEV FI, P1
[9]   Maximum Likelihood Detection With Ligand Receptors for Diffusion-Based Molecular Communications in Internet of Bio-Nano Things [J].
Kuscu, Murat ;
Akan, Ozgur B. .
IEEE TRANSACTIONS ON NANOBIOSCIENCE, 2018, 17 (01) :44-54
[10]  
Lotter S., 2023, IEEE Nanotechnol. Mag.