A.Yu. Parshin – Ph.D.(Eng.), Associate Professor, 

Department «Radio Engineering Devices», Ryazan State Radio Engineering University named after V.F. Utkin E-mail: parshin.a.y@rsreu.ru

Yu.N. Parshin – Dr.Sc.(Eng.), Professor, Head of Department «Radio Engineering Devices»,  Ryazan State Radio Engineering University named after V.F. Utkin E-mail: parshin.y.n@rsreu.ru

The article deals with the methods for compensating flicker noise in ultra-narrowband signal processing systems as applied to the implementation of the Internet of Things (IoT) device systems. High requirements for energy efficiency in IoT systems on the one hand and a low information transmission rate on the other hand allows the usage of signals with a small spectrum width, including the lowfrequency region of the spectrum, where flicker noise also has a noticeable effect along with thermal noise. Due to the extremely small width of the spectrum of the information signal, the transition to processing on a zero carrier will lead to a sharp increase in the influence of low-frequency noise on the processing results. Flicker noise has a power spectral density that increases with decreasing frequency. The fractional nature of the power spectral density makes it possible to use models of random processes with a fractional dimension to describe low-frequency noise. A measure of self-similarity is the dimension of an object. Thus, the evaluation of the parameters of objects allows to determine the properties of low-frequency noise and perform its subsequent compensation.

An important area in signal processing is the development of methods and algorithms for optimal processing based on the theory of statistical decisions. The general statement of the problem is implemented in the estimation-correlation-compensation approach. The article discusses the non-Gaussian flicker noise model, which allows a more accurate description of low-frequency noise with a fractional character, compared with the common Gaussian model of fractal Brownian motion, the measure of fractality of which is the Hurst exponent. The article presents mathematical expressions for a non-Gaussian flicker noise model based on differential equations. The analysis of the signal-to-noise ratio by a simulation method is carried out, the optimal estimation-correlation-compensation method, non-linear compensation methods are compared with the optimal or matched correlation processing by the parameter of the obtained signal-tonoise ratios. It is shown that increasing the signal duration on the one hand increases the signal energy, and on the other hand reduces the width of its spectrum. For a linear, consistent algorithm, reducing the width of the spectrum of the signal shifts the spectrum to the low-frequency region where flicker noise is more intense, which compensates for the increase in signal energy. As a result, as the signal duration increases, the linear processing efficiency tends to a constant value. With non-Gaussian flicker noise, the usage of non-linear noise compensation significantly increases the signal-to-noise ratio.

1. Mangalvedhe N., Ratasuk R., Ghosh A. NB-IoT Deployment Study for Low Power Wide Area Cellular IoT. IEEE 27th Annual International Symposium on Personal, Indoor and Mobile Radio Communication. 2016.
2. Potapov A.A. Fraktaly v radiofizike i radiolokatsii: topologiya vyborki. M.: Universitetskaya kniga. 2005. 848 s. (in Russian)
3. Crownover R.M. Introduction to Fractals and Chaos. Jones and Bartlett. 1995.
4. Sosulin Yu.G., Kostrov V.V, Parshin Yu.N. Otsenochno-korrelyatsionnaya obrabotka signalov i kompensatsiya pomekh. M.: Radiotekhnika. 2014. 632 s. (in Russian)
5. Sosulin Yu.G., Parshin Yu.N. Nelineinoe otsenivanie stokhasticheskikh signalov s adaptatsiei parametrov algoritmov. Radiotekhnika i elektronika. 1986. T. 31. № 5. S. 904−910. (in Russian)
6. Parshin Yu.N., Parshin A.Yu. Maksimalno pravdopodobnoe otsenivanie korrelyatsionnoi razmernosti s uchetom vliyaniya smeshcheniya otsenki i usecheniya diapazona masshtabov. Vestnik Ryazanskogo gosudarstvennogo radiotekhnicheskogo universiteta. 2013. № 4. S. 18−25. (in Russian)
7. Parshin A.Yu., Parshin Yu.N. Vydelenie granits fraktalnogo ob'ekta metodom maksimalnogo pravdopodobiya po nezavisimoi i zavisimoi vyborkam. Vestnik Ryazanskogo gosudarstvennogo radiotekhnicheskogo universiteta. 2014. № 4. S. 3−9. (in Russian)
8. Bochkov G.N., Kuzovlev Yu.E. Novoe v issledovaniyakh 1/f shuma. Uspekhi fizicheskikh nauk. 1983. T. 141. № 1. S. 151−176. (in Russian)
9. Kaukalys B., Ruseckas J. Stochastic nonlinear differential equation generating 1/f noise. Physical Review E 70, 020101(R). 2004.
10. Kaukalys B., Alaburda M., Ruseckas J. Modeling non-Gaussian 1/f noise by the stochastic differential equations. 19th International Conference on Noise and Fluctuations (ICNF-2007). AIP Conf. Proc. 2007. 922. P. 439−442.
11. Mamontov Y.V., Willander M. Long Asymptotic Correlation Time for Non-Linear Autonomous Ito’s Stochastic Differential Equation. Nonlinear Dynamics. 1997 V. 12. P. 399−411 (Kluwer Academic Publishers. Printed in the Netherlands).
12. Parshin A.Yu., Parshin Yu.N. Optimalnaya obrabotka uzkopolosnogo signala v prisutstvii gaussovskogo i negaussovskogo flikker shumov. Radiotekhnika. 2019. T. 83. № 5(6). S. 161−167. (in Russian)
13. Seidzh E., Mels Dzh. Teoriya otsenivaniya i ee primenenie v svyazi i upravlenii: Per. s angl. M.: Svyaz. 1976. 496 s. (in Russian)