Wyniki wyszukiwania - Biblioteka Nauki

Nowa wersja platformy, zawierająca wyłącznie zasoby pełnotekstowe, jest już dostępna.
Przejdź na https://bibliotekanauki.pl

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: pointing error

Sortuj według:

Ogranicz wyniki do:

Variable antennas positions solution to reduce pointing errors due to wind speed and temperature coupled effects during free space optical link using matrix Rician pointing error model

100%

Teuma Mbezi M. , Eke S. , Som I. H. J. , Mouangue R. M.

tom Vol. 53, nr 3

393--406

Pointing errors (PE) during free space optical (FSO) transmission can be caused by laser beam wander due to thermal and wind dynamic instability. The aim of this work is to study the coupled effects of temperature and wind speed on PE using matrix Rician pointing error (MRPE) model; then show how variable antennas height can reduce PE due to wind speed and temperature coupled effects. To achieve this purposes, average PE expression was established using MRPE model. Then considering a Gaussian beam wave and Monin–Obukhov similarity functions for the structure parameters of temperature, explicit relationship was established between average PE, temperature and wind speed. It comes out of this study that under dynamic turbulence, one can appropriately modify temperature to reduce PE due to dynamic instability and reciprocally. Depending on turbulence large cells or frozen turbulence eddies distribution, PE can be reduced by appropriately modified antennas height or the distance between transmitter and receiver. That is why this work suggests to install variable or dynamic antennas (rather than fixed ones) which could intelligently modify its positions according to laser beam wander created by atmospheric turbulence.

Synthesis and Optimization of Neural Network Parameters for Control of Non-linear Objects

75%

Aleksander M. , Pohrebennyk V. , Strembitskyy M.

tom R. 90, nr 5

207--210

The possibility of using artificial neural network (ANN) for the construction of precise control of dynamic objects with variable parameters is considered. Approach to the synthesis of topology and learning algorithm based on recurrent Elman’s ANN with auxiliary feedback is suggested. The results of modeling of ANN operation in the system of control guided by large-sized antennas are given.

Opisano możliwość wykorzystania sztucznej sieci neuronowej (SSN) do precyzyjnego sterowania dynamicznych obiektów o zmiennych parametrach. Przedstawiono syntezę topologii i uczenia algorytmu opartego na SSN Elmana z dodatkowych informacji. Zaprezentowano wyniki modelowania SSN do zarządzania systemem pozycjonowania dużych anten.