Exercise 2.11Z: Erasure Channel for Symbols

From LNTwww

Erasure channel for symbols:  "m–BEC"

The channel model  "Binary Erasure Channel"  $\rm (BEC)$ describes a bit-level erasure channel:

  • A binary symbol  $(0$  or  $1)$  is correctly transmitted with probability  $1 - \lambda$  and marked as an erasure  "$\rm E$"  with probability  $\lambda$.


A Reed–Solomon code is based on a Galois field   ${\rm GF}(2^m)$   with integer  $m$.  Thus,  each code symbol   $c$   can be represented by  $m$  bits.  If one wants to apply the BEC model here,  one has to modify it to the  "$m$–BEC"  model  as shown in the diagram below for  $m = 2$:

  • All code symbols   $($in binary representation  $00, \ 01, \ 10, \ 11)$   are transmitted correctly with probability  $1 - \lambda_2$ .
  • Thus,  the probability of an erased symbol  is  $\lambda_2$.
  • Note that already a single erased bit leads to the erased received symbol  $y = \rm E$ .




Hints:

  • For a code based on  ${\rm GF}(2^m)$  the outlined  $2$–BEC model is to be extended to  $m$–BEC.
  • The erasure probability of this model is then denoted by  $\lambda_m$.
  • For the first subtasks,  use always  $\lambda = 0.2$  for the erasure probability of the basic model according to the upper graph.



Questions

1

$\lambda = 0.2$  is valid.  With what probabilities do the possible received values occur in the  BEC basic model ?

${\rm Pr}(y = 0) \ = \ $

$\ \%$
${\rm Pr}(y = {\rm E}) \ = \ $

$\ \%$
${\rm Pr}(y = {\rm 1}) \ = \ $

$\ \%$

2

$\lambda = 0.2$  is valid.  What is the erasure probability  $\lambda_2$  at symbol level  $(2$–BEC model$)$  when the Reed-Solomon code is based on  $\rm GF(2^2)$?

$\lambda_2 \ = \ $

$\ \%$

3

$\lambda = 0.2$  is valid.  What is the erasure probability  $\lambda_m$,  when the  $m$–BEC model  is fitted to the  $\rm RSC \, (255, \, 223, \, 33)_{256}$?

$\lambda_m \ = \ $

$\ \%$

4

What is the maximum allowed erasure probability  $\lambda$  of the  "BEC basic model"  for  $\lambda_m ≤ 0.2$  to hold?

${\rm Max} \ \big[\lambda\big ] \ = \ $

$\ \%$

5

What is the probability of receiving the  "zero symbol"  in this case?

${\rm Pr}(y_{\rm bin} = 00000000) \ = \ $

$\ \%$


Solution

(1)  Due to the symmetry of the given BEC model  ("bit-level erasure channel"),  the erasure probability is:

$$\ {\rm Pr}(y = {\rm E}) = \lambda \ \underline{= 20\%}.$$
  • Since the code symbols  $0$  and  $1$  are equally likely,  we obtain   ${\rm Pr}(y = 0) \ \underline{= 40\%}$   and   ${\rm Pr}(y = 1) \ \underline{= 40\%}$   for their probabilities.


(2)  Without limiting generality,  we assume the code symbol  $c_{\rm bin} = $ "$00$"  to solve this exercise.

  • According to the  $2$–BEC model,  the received symbol  $y_{\rm bin}$  can then be either  "$00$"  or erased  $(\rm E)$  and it holds:
$${\rm Pr}(y_{\rm bin} = "00"\hspace{0.05cm} |\hspace{0.05cm} c_{\rm bin} = "00") \hspace{-0.15cm} \ = \ \hspace{-0.15cm} ( 1 - \lambda)^2 = 0.8^2 = 0.64 = 1 - \lambda_2\hspace{0.3cm} \Rightarrow \hspace{0.3cm} \lambda_2 \hspace{-0.15cm} \ = \ \hspace{-0.15cm} 1 - ( 1 - \lambda)^2 \hspace{0.15cm}\underline{= 36\%}\hspace{0.05cm}. $$
  • This assumes that an erasure is avoided only if neither of the two bits has been erased.


(3)  The   $\rm RSC \, (255, \, 223, \, 33)_{256}$   is based on the Galois field  $\rm GF(256) = GF(2^8) \ \Rightarrow \ \it m = \rm 8$.

  • The result of subtask  (2)  must now be adapted to this case.  For the  $8$–BEC holds:
$$1 - \lambda_8 = ( 1 - \lambda)^8 = 0.8^8 \approx 0.168 \hspace{0.3cm} \Rightarrow \hspace{0.3cm} \lambda_m = \lambda_8 \hspace{0.15cm}\underline{\approx 83.2\%}\hspace{0.05cm}. $$


(4)  From the condition   $\lambda_m ≤ 0.2$  it follows directly  $1 - \lambda_m ≥ 0.8$.  From this follows further:

$$( 1 - \lambda)^8 \ge 0.8 \hspace{0.3cm} \Rightarrow \hspace{0.3cm} 1 - \lambda \ge 0.8^{0.125} \approx 0.9725 \hspace{0.3cm} \Rightarrow \hspace{0.3cm}\lambda \hspace{0.15cm} \underline{\le 2.75\%}\hspace{0.05cm}.$$


(5)  Here one proceeds as follows:

  • With  $\lambda = 0.0275 \Rightarrow \ \lambda_m = 0.2$   ⇒   $20\%$  of the received symbols are  erasured.
  • The  $2^8 = 256$  received symbols  $(00000000$   ...   $11111111)$  are all equally likely.  It follows:
$${\rm Pr}(y_{\rm bin} = 00000000) = \hspace{0.1cm}\text{...} \hspace{0.1cm}= {\rm Pr}(y_{\rm bin} = 11111111)= \frac{0.8}{256} \hspace{0.15cm}\underline{= 0.3125\%}\hspace{0.05cm}.$$