Non si dispone dei permessi necessari per modificare questa pagina, per il seguente motivo:

La funzione richiesta è riservata agli utenti che appartengono al gruppo: Utenti.


È possibile visualizzare e copiare il codice sorgente di questa pagina.

Templates used on this page:

Torna a TCP over Hsdpa.

GoodputRTTPLR/TimeoutsGoodput/RTT
Back downlink/ uplink downlink/ uplink downlink / uplink downlink / uplink

TCP Congestion Control Algorithms over a live HSDPA Network[edit]

We report results obtained by an extensive measurement campaign over a live HSDPA network involving more than 3000 flows resulting in around 60 hours of active measurements. The considered TCP variants are the following:


The experimental evaluation has been carried out by accessing the public Internet using a commercial HSDPA card (see figure below) and by employing the linux Kernel 2.6.24.

HsdpaTestbed.png

The following instantaneous variable measurements have been collected by using the libnetmeas library:

The main findings of the investigation can be summarized as follows:

  1. TCP variants provides similar goodputs
  2. TCP Bic/Cubic exhibit inflated values for average RTT due to the aggressive nature of their probing phase. TCP NewReno and TCP Westwood+ provide lower values of RTT, TCP Westwood+ exhibiting the lowest.
  3. TCP Bic/Cubic provoke two times packet retransmission values with respect to TCP Westwood+ and TCP NewReno.
  4. TCP Bic/Cubic provoke roughly three times the number of timeouts wrt TCP Westwood+ and TCP NewReno, TCP Cubic exhibiting the highest number.

Results of the experiments[edit]

We report cumulative distribution function plots:

Congestion window and RTT evolutions[edit]

Typical congestion window and RTT evolutions of the considered TCPs in the case of a single flow over the HSDPA downlink

Hsdpa cwnd 1 flow downlink westwood.png Hsdpa cwnd 1 flow downlink newreno.png Hsdpa cwnd 1 flow downlink bic.png Hsdpa cwnd 1 flow downlink cubic.png