视频传输：wireless error channels

Video Transmission over wireless error channels

[Preface]

The examples are included in .

 

[Background]

1.      I don’t use the default error models provided in NS2. I add Prawit Chumchu’s work in the. The model is shown blow. The error model is implemented at the receiver side.

 

 

2.      wireless GE model --- shows the burst error pattern

In the “good” state (G) losses occur with lower probability  while in the “bad” state (B) they happen with higher probability . Also,  is the probability of the state transiting from a good state to a bad state, and is the transition from a bad state to a good state. The steady state probabilities of being in states G and B are

 and ,

respectively. The average packet loss rate produced by the GE error model is

.

 

 

 

3.      wireless random uniform model --- the errors are distributed uniformly.

 

4.      Unicast vs. Multicast over 802.11 wireless network

l          In unicasting, the MAC layer at a sender retransmits an unacknowledged packet at a maximum of N times before it gives up. The perceived correct rate at application-level is thus , where N is the maximum number of retransmission at the MAC layer and p is the packet error rate at the physical-level. As a consequence, the application-level error rate is p_effective = p^N. (The default value for N is 4)

l          Due to the absence of retransmissions in broadcasting and multicasting, the packet error rate at the physical-level is the same as at the application-level.

 

[Run the examples]

1.      Change the path to ns-allinone-2.28/ns-2.28/myexample/video-wireless_error/

 

2.      Example 1: The video packets are delivered via multicast over a random uniform error model of wireless channels.

Description:

a.       ns test_bsc_multicast.tcl 0 0 0.1 0 1 1024 (test_bsc_multicast.tcl: simulation script; 0: don’t care; 0: don’t care; 0.1:average error rate; 0:don’t care; 1: seed; 1024: the max packet size)

b.      the error rate = (163682-147621)/163682 = 0.098 ~ 0.1 (very close to the setting value)

 

3.      Example 2: The video packets are delivered via multicast over a GE error model of wireless channels.

 

Description:

c.       ns test_ge_multicast.tcl 0.96 0.94 0.001 0.05 1 1024 (0.96: P_GG; 0.94: P_BB; 0.001:P_G; 0.05:P_B; 1: seed; 1024: the max packet size)

d.      the error rate = (163682-160391)/163682 = 0.02  [ theory value : 0.01*(1-0.94)+0.05*(1-0.96)/((1-0.96)+(1-0.94))= 0.0206]

 

5.      Example 3: The video packets are delivered via unicast over a random error model of wireless channels.

Description:

e.       ns test_ge_multicast.tcl 0 0 0.5 0 1 1024 (0: don’t care; 0: don’t care; 0.5:average error rate at the physical  level; 0:don’t care; 1: seed; 1024: the max packet size)

f.        the error rate = (163682-153625)/153682=0.061  [ theory value : 0.5*0.5*0.5*0.5=0.0625]

 

Last modified date: 2006/3/3

 

Author: Chih-Heng, Ke (柯志亨)

Email:

MSN: