Aufgaben:Exercise 2.12Z: Reed-Solomon Syndrome Calculation: Difference between revisions

From LNTwww
Noah (talk | contribs)
No edit summary
Noah (talk | contribs)
No edit summary
Line 1: Line 1:
{{quiz-Header|Buchseite=Kanalcodierung/Fehlerkorrektur nach Reed–Solomon–Codierung}}
{{quiz-Header|Buchseite=Channel_Coding/Error_Correction_According_to_Reed-Solomon_Coding}}


[[File:P_ID2559__KC_T_2_5_Darstellung.png|right|frame|Conversion table for the Galois field  $\rm GF(2^3)$]]
[[File:P_ID2559__KC_T_2_5_Darstellung.png|right|frame|Conversion table for the Galois field  $\rm GF(2^3)$]]
Wie in der  [[Aufgaben:Exercise_2.12:_Decoding_at_RSC_(7,_4,_4)_to_Base_8|"Aufgabe 2.12"]]  betrachten wir den Reed–Solomon–Code  $(7, \, 4, \, 4)_8$, der auf dem Galoisfeld  ${\rm GF}(q)$  mit  $q = 8 = 2^3$  basiert. Die Grafik zeigt die zugehörige Umrechnungstabelle.  
As in the  [[Aufgaben:Exercise_2.12:_Decoding_at_RSC_(7,_4,_4)_to_Base_8|"Exercise 2. 12"]]  we consider the Reed–Solomon code  $(7, \, 4, \, 4)_8$ based on the Galois field  ${\rm GF}(q)$  with  $q = 8 = 2^3$ . The graph shows the corresponding conversion table.  


Gegeben sind die möglichen Codesymbole in Exponentendarstellung $($Potenzen von  $\alpha)$  sowie in Polynom– und Koeffizientenvektordarstellung.
Given are the possible code symbols in exponent representation $($potencies of  $\alpha)$  and in polynomial– and coefficient vector representation.


Vorgegeben ist das Empfangswort  $\underline{y} = (\alpha, \, 0, \, \alpha^3, \, 0, \, 1, \, \alpha, \, 0)$. Anhand des Syndroms
Given the received word  $\underline{y} = (\alpha, \, 0, \, \alpha^3, \, 0, \, 1, \, \alpha, \, 0)$. Based on the syndrome
:$$\underline {s} = (s_0, s_1, s_2) = \underline {y}  \cdot { \boldsymbol{\rm H }}^{\rm T}$$
:$$\underline {s} = (s_0, s_1, s_2) = \underline {y}  \cdot { \boldsymbol{\rm H }}^{\rm T}$$


soll überprüft werden, ob einzelne Symbole des Empfangsvektors  $\underline{y}$  bei der Übertragung verfälscht wurden. Gegeben ist hierzu die Prüfmatrix  $\mathbf{H}$  des betrachteten Codes und deren Transponierte:
is to check whether individual symbols of the received vector  $\underline{y}$  were corrupted during transmission. Given is the parity-check matrix  $\mathbf{H}$  of the considered code and its transpose:
:$${ \boldsymbol{\rm H}} =  
:$${ \boldsymbol{\rm H}} =  
\begin{pmatrix}
\begin{pmatrix}
Line 34: Line 34:




''Hinweis:''
Hints:
* Die Aufgabe bezieht sich auf die Seite  [[Channel_Coding/Fehlerkorrektur_nach_Reed%E2%80%93Solomon%E2%80%93Codierung#Schritt_.28A.29:_Auswertung_des_Syndroms_beim_BDD| Schritt  $\rm (A)$: Auswertung des Syndroms beim BDD]]  des Kapitels "Fehlercodierung nach Reed–Solomon–Codierung".
* This exercise refers to the page  [[Channel_Coding/Error_Correction_According_to_Reed-Solomon_Coding#Step_.28A.29:_Evaluation_of_the_syndrome_in_BDD| "Step (A): Evaluation of the syndrome in BDD"]]  of the chapter "Error coding according to Reed–Solomon coding".
   
   


Line 41: Line 41:




===Fragebogen===
===Questions===
<quiz display=simple>
<quiz display=simple>
{Empfangen wurde&nbsp; $\underline{y} = (\alpha, \, 0, \, \alpha^3, \, 0, \, 1, \, \alpha, \, 0)$. Geben Sie das erste Element des Syndroms&nbsp; $\underline{s} = (s_0, \, s_1, \, s_2)$&nbsp; an.
{Received was&nbsp; $\underline{y} = (\alpha, \, 0, \, \alpha^3, \, 0, \, 1, \, \alpha, \, 0)$. Specify the first element of the syndrome&nbsp; $\underline{s} = (s_0, \, s_1, \, s_2)$&nbsp;.
|type="()"}
|type="()"}
+ $s_0 = \alpha^4$,
+ $s_0 = \alpha^4$,
- $s_0 = \alpha^5$,
- $s_0 = \alpha^5$,
- $s_0 = \alpha^6$,
- $s_0 = \alpha^6$,
- $s_0 = 0, \, 1, \, \alpha, \, \alpha^2$&nbsp; oder&nbsp; $\alpha^3$.
- $s_0 = 0, \, 1, \, \alpha, \, \alpha^2$&nbsp; or&nbsp; $\alpha^3$.


{Wie lautet bei gleichem Empfangswort das zweite Syndromelement?
{What is the second syndrome element for the same received word?
|type="()"}
|type="()"}
- $s_1 = \alpha^4$,
- $s_1 = \alpha^4$,
+ $s_1 = \alpha^5$,
+ $s_1 = \alpha^5$,
- $s_1 = \alpha^6$,
- $s_1 = \alpha^6$,
- $s_1 = 0, \, 1, \, \alpha, \, \alpha^2$&nbsp; oder&nbsp; $\alpha^3$.
- $s_1 = 0, \, 1, \, \alpha, \, \alpha^2$&nbsp; or&nbsp; $\alpha^3$.


{Wie lautet bei gleichem Empfangswort das dritte Syndromelement?
{What is the third syndrome element for the same received word?
|type="()"}
|type="()"}
- $s_2 = \alpha^4$,
- $s_2 = \alpha^4$,
Line 64: Line 64:
- $s_2 = 0, \, 1, \, \alpha, \, \alpha^2$ oder $\alpha^3$.
- $s_2 = 0, \, 1, \, \alpha, \, \alpha^2$ oder $\alpha^3$.


{Bekannt ist, dass das vorliegende Empfangswort&nbsp; $\underline{y}$&nbsp; richtig decodiert werden kann. Wieviele Symbolfehler beinhaltet das Empfangswort?
{Know that the received word at hand&nbsp; $\underline{y}$&nbsp; can be decoded correctly. How many symbol errors does the received word contain?
|type="{}"}
|type="{}"}
$r \ = \ ${ 1 3% }
$r \ = \ ${ 1 3% }
</quiz>
</quiz>


===Musterlösung===
===Solution===
{{ML-Kopf}}
{{ML-Kopf}}
[[File:P_ID2560__KC_T_2_5_Darstellung.png|right|frame|Umrechnungstabellen für das Galoisfeld $\rm GF(2^3)$]]  
[[File:P_ID2560__KC_T_2_5_Darstellung.png|right|frame|Conversion tables for the Galois field $\rm GF(2^3)$]]  
'''(1)'''&nbsp; Die entsprechende Gleichung zur Syndromberechnung lautet:
'''(1)'''&nbsp; The corresponding equation for syndrome calculation is:
:$$\underline {s} \hspace{-0.15cm} \ = \ \hspace{-0.15cm} (s_0, s_1, s_2) =
:$$\underline {s} \hspace{-0.15cm} \ = \ \hspace{-0.15cm} (s_0, s_1, s_2) =
\begin{pmatrix}
\begin{pmatrix}
Line 88: Line 88:
\hspace{0.05cm}.$$
\hspace{0.05cm}.$$


Das erste Element ergibt sich zu
The first element results in
:$$s_0 \hspace{-0.15cm} \ = \ \hspace{-0.15cm}  \alpha \cdot 1 + \alpha^3 \cdot \alpha^2 + 1 \cdot \alpha^4 + \alpha \cdot \alpha^5=
:$$s_0 \hspace{-0.15cm} \ = \ \hspace{-0.15cm}  \alpha \cdot 1 + \alpha^3 \cdot \alpha^2 + 1 \cdot \alpha^4 + \alpha \cdot \alpha^5=
\alpha  + \alpha^5 + \alpha^4+ \alpha^6$$
\alpha  + \alpha^5 + \alpha^4+ \alpha^6$$
:$$\Rightarrow\hspace{0.3cm} s_0 \hspace{-0.15cm} \ = \ \hspace{-0.15cm}  (\alpha) + (\alpha^2 + \alpha+ 1)+ (\alpha^2 + \alpha) + + (\alpha^2 +  1) = \alpha^2 + \alpha = \alpha^4\hspace{0.05cm}.$$
:$$\Rightarrow\hspace{0.3cm} s_0 \hspace{-0.15cm} \ = \ \hspace{-0.15cm}  (\alpha) + (\alpha^2 + \alpha+ 1)+ (\alpha^2 + \alpha) + + (\alpha^2 +  1) = \alpha^2 + \alpha = \alpha^4\hspace{0.05cm}.$$


Richtig ist der <u>Lösungsvorschlag 1</u>.
Correct is the <u>proposed solution 1</u>.




'''(2)'''&nbsp; Entsprechend gilt für das zweite Syndromelement entsprechend dem der <u>Lösungsvorschlag 2</u>:
'''(2)'''&nbsp; Correspondingly, for the second syndrome element, the <u>proposed solution 2</u> applies accordingly:
:$$s_1 \hspace{-0.15cm} \ = \ \hspace{-0.15cm}  \alpha \cdot 1 + \alpha^3 \cdot \alpha^4 + 1 \cdot \alpha^1 + \alpha \cdot \alpha^3=
:$$s_1 \hspace{-0.15cm} \ = \ \hspace{-0.15cm}  \alpha \cdot 1 + \alpha^3 \cdot \alpha^4 + 1 \cdot \alpha^1 + \alpha \cdot \alpha^3=
\alpha  + \alpha^7 + \alpha+ \alpha^4= 1 + \alpha^4 = \alpha^2 + \alpha + 1 =  \alpha^5
\alpha  + \alpha^7 + \alpha+ \alpha^4= 1 + \alpha^4 = \alpha^2 + \alpha + 1 =  \alpha^5
Line 102: Line 102:




'''(3)'''&nbsp; Zur Berechnung von $s_2$ muss mit der letzten Matrixspalte multipliziert werden:
'''(3)'''&nbsp; To calculate $s_2$ must be multiplied by the last matrix column:
:$$s_2 \hspace{-0.15cm} \ = \ \hspace{-0.15cm}  \alpha \cdot 1 + \alpha^3 \cdot \alpha^6 + 1 \cdot \alpha^5 + \alpha \cdot \alpha^1=
:$$s_2 \hspace{-0.15cm} \ = \ \hspace{-0.15cm}  \alpha \cdot 1 + \alpha^3 \cdot \alpha^6 + 1 \cdot \alpha^5 + \alpha \cdot \alpha^1=
\alpha  + \alpha^2 + \alpha^5 + \alpha^2=\alpha^5 + \alpha = (\alpha^2 + \alpha + 1) + \alpha = \alpha^2  + 1 = \alpha^5
\alpha  + \alpha^2 + \alpha^5 + \alpha^2=\alpha^5 + \alpha = (\alpha^2 + \alpha + 1) + \alpha = \alpha^2  + 1 = \alpha^5
\hspace{0.05cm}.$$
\hspace{0.05cm}.$$


Richtig ist der  <u>Lösungsvorschlag 3</u>.
Correct is the <u>proposed solution 3</u>.




'''(4)'''&nbsp; Aufgrund des errechneten Syndroms $\underline{s} = (\alpha^4, \, \alpha^5, \, \alpha^6) &ne; 0$ beinhaltet das Empfangswort mindestens einen Symbolfehler &nbsp; &#8658; &nbsp; $r > 0$.  
'''(4)'''&nbsp; Due to the calculated syndrome $\underline{s} = (\alpha^4, \, \alpha^5, \, \alpha^6) &ne; 0$ the received word contains at least one symbol error &nbsp; &#8658; &nbsp; $r > 0$.  
*Der vorliegende Reed&ndash;Solomon&ndash;Code $(7, \, 4, \, 4)_8 \ \Rightarrow \ d_{\rm min} = 4$ kann nicht mehr als $t = &lfloor;d_{\rm min}/2&rfloor; = 1$ Fehler korrigieren.
*The present Reed&ndash;Solomon&ndash;code $(7, \, 4, \, 4)_8 \ \Rightarrow \ d_{\rm min} = 4$ cannot correct more than $t = &lfloor;d_{\rm min}/2&rfloor; = 1$ errors.
* Da das Empfangswort gemäß der Angabe tatsächlich decodiert werden kann, gilt&nbsp; $\underline{r = 1}$.
* Since the received word can actually be decoded according to the specification,&nbsp; $\underline{r = 1}$ holds.
{{ML-Fuß}}
{{ML-Fuß}}



Revision as of 18:58, 12 September 2022

Conversion table for the Galois field  $\rm GF(2^3)$

As in the  "Exercise 2. 12"  we consider the Reed–Solomon code  $(7, \, 4, \, 4)_8$ based on the Galois field  ${\rm GF}(q)$  with  $q = 8 = 2^3$ . The graph shows the corresponding conversion table.

Given are the possible code symbols in exponent representation $($potencies of  $\alpha)$  and in polynomial– and coefficient vector representation.

Given the received word  $\underline{y} = (\alpha, \, 0, \, \alpha^3, \, 0, \, 1, \, \alpha, \, 0)$. Based on the syndrome

$$\underline {s} = (s_0, s_1, s_2) = \underline {y} \cdot { \boldsymbol{\rm H }}^{\rm T}$$

is to check whether individual symbols of the received vector  $\underline{y}$  were corrupted during transmission. Given is the parity-check matrix  $\mathbf{H}$  of the considered code and its transpose:

$${ \boldsymbol{\rm H}} =

\begin{pmatrix} 1 & \alpha^1 & \alpha^2 & \alpha^3 & \alpha^4 & \alpha^5 & \alpha^6\\ 1 & \alpha^2 & \alpha^4 & \alpha^6 & \alpha^1 & \alpha^{3} & \alpha^{5}\\ 1 & \alpha^3 & \alpha^6 & \alpha^2 & \alpha^{5} & \alpha^{1} & \alpha^{4} \end{pmatrix} \hspace{0.05cm},\hspace{0.4cm} { \boldsymbol{\rm H}}^{\rm T} = \begin{pmatrix} 1 & 1 & 1 \\ \alpha^1 & \alpha^2 & \alpha^3 \\ \alpha^2 & \alpha^4 & \alpha^6 \\ \alpha^3 & \alpha^6 & \alpha^2 \\ \alpha^4 & \alpha^1 & \alpha^{5} \\ \alpha^5 & \alpha^{3} & \alpha^{1} \\ \alpha^6 & \alpha^{5} & \alpha^{4} \end{pmatrix} \hspace{0.05cm}.$$





Hints:



Questions

1 Received was  $\underline{y} = (\alpha, \, 0, \, \alpha^3, \, 0, \, 1, \, \alpha, \, 0)$. Specify the first element of the syndrome  $\underline{s} = (s_0, \, s_1, \, s_2)$ .

$s_0 = \alpha^4$,
$s_0 = \alpha^5$,
$s_0 = \alpha^6$,
$s_0 = 0, \, 1, \, \alpha, \, \alpha^2$  or  $\alpha^3$.

2 What is the second syndrome element for the same received word?

$s_1 = \alpha^4$,
$s_1 = \alpha^5$,
$s_1 = \alpha^6$,
$s_1 = 0, \, 1, \, \alpha, \, \alpha^2$  or  $\alpha^3$.

3 What is the third syndrome element for the same received word?

$s_2 = \alpha^4$,
$s_2 = \alpha^5$,
$s_2 = \alpha^6$,
$s_2 = 0, \, 1, \, \alpha, \, \alpha^2$ oder $\alpha^3$.

4 Know that the received word at hand  $\underline{y}$  can be decoded correctly. How many symbol errors does the received word contain?

$r \ = \ $


Solution

Conversion tables for the Galois field $\rm GF(2^3)$

(1)  The corresponding equation for syndrome calculation is:

$$\underline {s} \hspace{-0.15cm} \ = \ \hspace{-0.15cm} (s_0, s_1, s_2) =

\begin{pmatrix} \alpha,0, \alpha^3,0, 1, \alpha,0 \end{pmatrix}\cdot \begin{pmatrix} 1 & 1 & 1 \\ \alpha^1 & \alpha^2 & \alpha^3 \\ \alpha^2 & \alpha^4 & \alpha^6 \\ \alpha^3 & \alpha^6 & \alpha^2 \\ \alpha^4 & \alpha^1 & \alpha^{5} \\ \alpha^5 & \alpha^{3} & \alpha^{1} \\ \alpha^6 & \alpha^{5} & \alpha^{4} \end{pmatrix} \hspace{0.05cm}.$$

The first element results in

$$s_0 \hspace{-0.15cm} \ = \ \hspace{-0.15cm} \alpha \cdot 1 + \alpha^3 \cdot \alpha^2 + 1 \cdot \alpha^4 + \alpha \cdot \alpha^5=

\alpha + \alpha^5 + \alpha^4+ \alpha^6$$

$$\Rightarrow\hspace{0.3cm} s_0 \hspace{-0.15cm} \ = \ \hspace{-0.15cm} (\alpha) + (\alpha^2 + \alpha+ 1)+ (\alpha^2 + \alpha) + + (\alpha^2 + 1) = \alpha^2 + \alpha = \alpha^4\hspace{0.05cm}.$$

Correct is the proposed solution 1.


(2)  Correspondingly, for the second syndrome element, the proposed solution 2 applies accordingly:

$$s_1 \hspace{-0.15cm} \ = \ \hspace{-0.15cm} \alpha \cdot 1 + \alpha^3 \cdot \alpha^4 + 1 \cdot \alpha^1 + \alpha \cdot \alpha^3=

\alpha + \alpha^7 + \alpha+ \alpha^4= 1 + \alpha^4 = \alpha^2 + \alpha + 1 = \alpha^5 \hspace{0.05cm}.$$


(3)  To calculate $s_2$ must be multiplied by the last matrix column:

$$s_2 \hspace{-0.15cm} \ = \ \hspace{-0.15cm} \alpha \cdot 1 + \alpha^3 \cdot \alpha^6 + 1 \cdot \alpha^5 + \alpha \cdot \alpha^1=

\alpha + \alpha^2 + \alpha^5 + \alpha^2=\alpha^5 + \alpha = (\alpha^2 + \alpha + 1) + \alpha = \alpha^2 + 1 = \alpha^5 \hspace{0.05cm}.$$

Correct is the proposed solution 3.


(4)  Due to the calculated syndrome $\underline{s} = (\alpha^4, \, \alpha^5, \, \alpha^6) ≠ 0$ the received word contains at least one symbol error   ⇒   $r > 0$.

  • The present Reed–Solomon–code $(7, \, 4, \, 4)_8 \ \Rightarrow \ d_{\rm min} = 4$ cannot correct more than $t = ⌊d_{\rm min}/2⌋ = 1$ errors.
  • Since the received word can actually be decoded according to the specification,  $\underline{r = 1}$ holds.