[ns] Steady state throughput
Dear NS-users, I would like to evaluate the steady-state throughput of a network of few nodes with respect to the packet arrival rate (PAR). For each PAR, I extract from the trace file the corresponding Throughput by using the following formula Throughput = (total number of bits received) / (simulation duration). Let's mentioned that in the simulation duration, the transient period is not considered. My problem is that I don't know how to compute the steady-state throughput. Because the throughput I am computing varies with the simulation duration. I tried to simulate with very long simulation durations (4000...6000 seconds), but there are still significant changes in the throughput values and the simulation takes too long time. How could I simulate the steady-state throughput? Thank you very much in advance for your help. Mozam -- View this message in context: http://www.nabble.com/Steady-state-throughput-tp18807257p18807257.html Sent from the ns-users mailing list archive at Nabble.com.
[ns] steady state throughput
Hello, Excuse me if i'm in the wrong forum, just do not consider this post message. I am a NS 2.29's user and I am evaluating the IEEE 802.15.4 performances. I would like to evaluate the throughput of a network of few nodes with respect to the packet arrival rate (PAR). For each PAR, I extract from the trace file the corresponding Throughput. However, I notice that the throughput varies with the simulation duration. But in my work, I am looking for steady state throughput, a kind of mean throughput. I tried to simulate with very long simulation durations (2000...3 seconds), but there are still changes in the throughput values and the simulation takes too long time. Please, could someone help me? How should I simulate that throughput? Thank you very much in advance. Mozam -- View this message in context: http://www.nabble.com/%22steady-state-throughput%22-tp17755772p17755772.html Sent from the ns-users mailing list archive at Nabble.com.
[ns] Steady state throughput
Hello, Excuse me if i'm in the wrong forum, just do not consider this post message. I am a NS 2.29's user and I am evaluating the IEEE 802.15.4 performances. I would like to evaluate the throughput of a network of few nodes with respect to the packet arrival rate (PAR). For each PAR, I extract from the trace file the corresponding Throughput. However, I notice that the throughput varies with the simulation duration. But in my work, I am looking for steady state throughput, a kind of mean throughput. I tried to simulate with very long simulation durations (2000...3 seconds), but there are still changes in the throughput values and the simulation takes too long time. Please, could someone help me? How should I simulate that throughput? Thank you very much in advance. Mozam
