HSPA State Change Measurements

A1-state-changes Last week I did some measurements to get an idea of the time required when switching between different HSPA air interface states. While data is transferred, the mobile is usually in Cell-DCH state on a High Speed Shared Channel. When only little or no data is transferred, the connection transferred to the Forward Access Channel, which is slow but has little overhead for both the network and the mobile device in terms of control measurements and power adjustments commands. If no data is transmitted for a longer duration (e.g. 30 seconds) the connection is put into Idle state. While the IP address is retained, the physical connection between the mobile and the network is severed.

As can be seen in the picture on the left, the round trip time to the first hop in the network of a ping packet is around 100 to 120 milliseconds while the mobile is using a high speed shared channel. While on the slower forward access channel, round trip time increases to 240 to 260 milliseconds. Moving from the high speed shared channel to the forward access channel is relatively quick, it takes around 550 to 600 ms (minus the actual round trip time of the packet itself). Going back to the high speed shared channel takes a little bit more time, around 1000 to 1500 milliseconds.

When using a 3G dongle with a notebook, a connection is rarely set into idle state as there is always one program or another such as an instant messenger, VoIP client, etc., that feels it needs to send a keep alive message to a server in the network before the idle time can expire. Therefore I haven't measured it this time. In the past, I've seen values around 2500 to 2800 milliseconds.

Some say that the effect of this state switching is that web browsing feels a bit more sluggish over HSPA than over a DSL line, which always offers Internet connectivity at full speed without the need of state switching. I use 3G connectivity a lot and quite frankly, while I can feel a difference, it's absolutely no problem to work and live with it.

And here's a quick overview of the test setup: Mobilkom Austria 3.5G HSPA network, a notebook connected via Wi-Fi to a D100 Wi-Fi/3G gateway, connected to a Huawei E220 3G USB stick, HSDPA category 6, no HSUPA.

8 thoughts on “HSPA State Change Measurements”

  1. Regarding “sluggish web browsing”. I’ve measured the effect of state transition from FACH to DCH on web page load times and depending on network parameters it can make quite an impact. For example, one Australian carrier showed an average page load time for Google greater than 6 seconds! Modifying the parameters to transition to DCH faster improved this to 1.7 seconds. The implication is the connection sat in FACH for about 5 seconds and struggled to send / receive any data.
    Where it is most noticeable is when a user loads a page, pauses to read then tries to follow a link. If the connection has dropped back to idle / PCH or FACH, the user perceives a slow connection because of the amount of time it appears for something to start happening.
    The time taken to transition between states will depend on the operators parameters. From my observations, it can be immediate (ie as soon as the first packet is sent) or can be delayed until a certain volume of data has been sent. This avoids transitioning to DCH for small volume traffic e.g pings, keep alives etc and saves on radio resources.
    BTW – Typical ping times I observe in FACH state are 1000 to 2000ms. Compared to sub 200ms in DCH.

  2. Hi,

    Yes, my own experience is the same as Matt’s. Going to Cell_FACH state leads to 1000-2000 ms of ping no matter the device used…

    It’s pretty interesting for me how Mobilkom Austria achieve this improvement 🙂 Probably optimized parameters of the FACH? Or some features?

    BR,
    Bozhidar

  3. Hi Martin,

    Interesting to see the HSPA ping Latency measurements.
    I am also having the similar experience with Orange,France.
    While beling connected in DCH state, RTT is around 120ms and in FACH state it is around 200ms.
    Good to see the improvements.

    Best regards,
    -Srini

  4. Hi Matt and Bozihdar,

    I am quite surprised about the 1000ms+ ping times you observed in cell-FACH state, I have never seen that before. Typically, I see values in the 250 – 300 ms range once cell-DCH is left, i.e. as reported above. That was even the case in non-HSPA times. Also, my observations are not limited to the Mobilkom network in Austria. I’ve also tested all 4 networks in Germany, Orange in France, TIM, Wind, Vodafone and 3 in Italy, Vodafone and 3 in the UK, etc. but the ping times are always in the same range. They differ, however, in the times required to switch state and also in the way the networks decide to switch states when traffic is low / increases again.

    Good Srini can confirm my values for Orange in France 🙂

    Cheers,
    Martin

  5. A suggested test: Compare web browsing on HSDPA to HSPA – you will find that the UL control plane latency for HSPA/HSUPA is much much better than for HSPA. This is especially evident for websites which use cookies and thus require more UL data to be sent. Most people think about speed but latency for web browsing can have a bigger impact on performance. BTW: Pingplotter is a great utility to measure RTT vs Time.

  6. Hi Martin,

    Sorry… It seems I was still sleeping when I posted 🙂

    Actually what I’ve seen is 200-300 ms ping in FACH, 75-120 ms in DCH and 1000-2000 only for the first ping when changing between states. This is for Globul and I think also for MTel in Bulgaria.

    Cheers,
    Bozhidar

  7. I guess here there is a little trick in latency measurements.
    My values are for small ping of 32bytes size.
    If u use big ping of 1024bytes, it takes more time like upto 1000ms in FACH state and 300ms in DCH.
    -Thanks all for sharing ur results.

    How to measure ping latency in Idle? Any one can shed some light?

    -BR,
    -Srini

Comments are closed.