the traffic models are implemented according to RAN R1-070674. For FTP, it is defined there as: “An FTP session is a sequence of file transfers separated by reading times”. That means that a file of random size is created and afterwards there is a random waiting time until the next file is put in the buffer. The file sizes and the waiting time are relatively long (compared to 1000 tti) so to transmit several packets in one simulation demands a very long total time. You can adapt the traffic model such that, e.g., only one file is created (maybe with smaller size) and then set the waiting time sufficiently large, larger than the total duration of your simulation.
1 – files are created consecutively, i.e., the longer your simulation, the more files will be transmitted.
2 – reading time: a file of arbitrary size is generated and the transmit buffer is being emptied by transmitting over the channel. Once a file is completely transmitted, there is a reading time (chosen from a random distribution as specified in the document I referenced above). After this reading time a new file is generated. (there is no term ‘waiting time’ specified in the document, I used in instead of ‘reading time’ – sorry if that caused confusion)
3 – the simulator is transmitting bits, while the file size is generated in bytes. Therefore, it is multiplied by 8.