Luciferase in rDNA technology (biotechnology).pptx
Елена Малютина - Оценка параметров хаотического процесса с помощью Ukf-фильтра для построения прогноза
1. Parameters Estimation for Chaotic Systems Using
the Unscented Kalman Filter (UKF) to Construct a Forecast
Elena I. Malyutina
Federal State State-Financed Educational Institution of High Professional Education
“South Ural State University”
05.13.01 System analysis, management and information processing (Postgraduate)
AIST Conference. April 2014. Yekaterinburg
2. e.mankevich@gmail.com
What is deterministic chaos?
Implementation of irregular oscillation regime similar to a random process but strictly
predictable in the sense of the law of evolution.
Actuality
Complexity of solving the problem of data processing which contain chaotic component is:
- Short time series
- The implementation process is unique
- No information about the probability distribution of errors.
Objective
To improve the forecast accuracy by the identification of the chaotic component of the time
process using deterministic nonlinear dynamic systems with chaotic solutions.
Application
Engineering, cryptography, medicine, economics and other fields of science and technology.
Problem statement
The model of chaotic process
( )
1
, 1,2,...,
n
i
k i k k
i
y a x k N
( )
1 , , 0,1, , 1, 1,2, , ,ii
k i k ix f x k N i n
( )
1 1 , 0,1, , , 1,2, ,i ii
k i k kx x x k N i n
0 0,1 , ,4 , 3.57i
ix
Fig.1. The objective function
3. e.mankevich@gmail.com
Practical results
Logistic map
In order to check the UKF it is considered a model example:
1 1
,
1 , 1,2,..., .
k k k
k k k
y ax
x x x k N
0 0.3, 3.69, 10 , 10%, (8 ) 10%x SNR dB REE RPE steps
Fig. 2. (a) The investigated process kx (–) and approximation (- -), (b) The estimation (- -), true value (–) of the
parameter
Fig. 3. (а) The forecast of the initial process, (b) Relative prediction error
The communication traffic prediction
Fig.4. (а) The forecast, (b) The residues and approximation of residues