X hits on this document

22 views

0 shares

0 downloads

0 comments

4 / 8

WSEAS TRANSACTIONS on ELECTRONICS

ρ (vi ,V fi ) = vi

a0i

+ b0i 2

b0 i

a0i 2

ρ (vi ,Vofi ) = vi

a pi

+ b pi 2

bpi

a pi 2

According to the distance definition of the classical and extensional domains, determines a relational function computation.

Calculate the relational function:

K (vi ) =

( ) ( ) ( ) ( ) , , , , , i f i i f i f i i f i i o f i i f i v V v V V v V v V v V ρ ρ ρ ρ

, i f i v V

Judge the relational function to determine the control output.

Ru (t) = Ru (t 1), Ru (t) = f (Ry , K(vi )),

K(vi ) 0 1 K(vi ) < 0

Ru (t) = Ru max

,

K(vi ) ≤ −1

1. K (vi )

  • 0

    , the output does not change and

maintains the last output.

2. 1 K (vi ) < 0 , according to the element model algorithm, resolve the outputs element.

3. K (vi ) < −1, is the output element of the biggest control amount.

4 Extension Neural Network Algorithm

The extension neural network is a solution method that combines the extension theory with a neural network. The matter element model of the extension theory is combined with the neural network as the learning mechanism. By adjusting the receiver oscillation frequency, it is synchronized to the oscillation frequency of the master atomic clock.

ISSN: 1109-9445

153

Guo-Shing Huang

Figure 2 shows the structure of the extension neural network. It is comprised of an input layer and output layer. The neuron of the input layer is made of the control plant characteristics. The output layer neuron is the control result. After adjusting the weighting factor, learning rate, and the extension distance, an ideal control result will be obtained. [12][13]

Om1

x

p m1

w

L 11

1 U 1 w

p m n x

r L n w

r U n w

Omr

m p t x

s L t w

s U t w

Oms

Figure 2. The structure of the extension neural network

The extension neural network algorithm procedure is as follows:

Step 1: Utilize extension matter element model to determine the weighting factor.

( ) r r r R N c V = =

1 2 r t M N c c c

1 2 r r r t V V V M

r N i s t h e G P S r e c e i v e r ; c i s t h e G P S r e c e i v e r

control characteristics of such as oscillation frequency, integer ambiguity, and so on. And

r n V

=

L , r n r n U w w

expresses

a

region

of

the

characteristics (weighting interval). The method for determining this region is obtained in the training set:

{ } m i n L r r n m n w x =

Issue 7, Volume 4, July 2007

Document info
Document views22
Page views22
Page last viewedMon Dec 05 08:50:04 UTC 2016
Pages8
Paragraphs408
Words4006

Comments