Prevent the data tsunami from swamping cell networks


Nokia Siemens Networks(s nok) (s si) published a blog post Friday called “Wake-up call: Industry collaboration needed to make Beyond 4G networks carry 1000 times more traffic by 2020.” Such a headline is designed to strike fear into the hearts of mobile operators (and NSN customers) everywhere as the demand for mobile broadband outstrips the carriers’ ability to supply it. Even AT&T (s t) is prepping for a traffic explosion, not just at the edge of the network, where its wireless base stations sit, but at the core of the network as the world transitions to an all-IP future complete with video.

From an interview with Krish Prabhu, of AT&T Labs, in Fierce Wireless:

Over the next five years, our traffic volumes tell us that when we launch LTE each one of our 11 regional cores will have a throughput of two to three terabits, and the national core will have a throughput at least 10 times that. … We are very involved in the solution to that problem. We’ve identified a layered approach to get us there even as we support the launch of our LTE network and get LTE to 90 to 95 percent of our end-users. That to me is the biggest challenge.

However, instead of fear mongering, both stories actually try to discuss some of the technical challenges the industry needs to meet to support the demand for data. The NSN report discussed the need for more spectrum, but also the need to figure out ways to cram more bits into a single megahertz of spectrum, so every airwave can work a little harder. We’ve covered some of the ways this can happen, from carrier aggregation technology to better use of available spectrum to more base stations to help with capacity. NSN goes further and discusses the need for cognitive radios and self-optimizing networks, a concept that major vendors are pushing as networks become more complicated.

Over at AT&T, Prabhu told Fierce Wireless that when the network was carrying mostly voice traffic, managing the network itself was simpler. However, with the switch to data, and soon to all-IP networks in the form of LTE, the way traffic is handled changes. Data traffic becomes harder to anticipate and predict, and can overflow the network or the handset. He said AT&T Labs is working with developers to understand how apps behave on the network (products such as those from Mu Dynamics can help with this) as well as researching things such as algorithmic flow control on the network and better signaling and control of how data flows through the networks. Companies such as ByteMobile, Starrent (acquired by Cisco), (s csco) Camiant (acquired by Tekelec) (s tklc) and others are providing some of these products. I expect we’ll hear more about this from Cole Brodman, the CMO of T-Mobile USA or Stephen Bye, the CTO of Sprint (s s), when they hit the stage at our Mobilize 2011 conference on Sept. 26 and 27.

However, both of these articles ignore a critical element to help meet mobile broadband demand: Wi-Fi. AT&T is already using Wi-Fi as a means to offload traffic from its cellular network, and Metro PCS (s pcs) may be offloading some 20 percent of its traffic via the technology. But there is still a lot that needs to happen to help integrate Wi-Fi into the cellular experience in a way that’s seamless and encourages the customer to use it and trust it. It’s not surprising that NSN wouldn’t want to focus on the topic, since it’s not an area where it’s selling gear, but I hope that AT&T is keeping its commitment to better Wi-Fi even as it expands capacity and the capability on its core network.

Here’s the brief NSN presentation from SlideShare:

Mobile Broadband Beyond 4G
View more presentations from Nokia Siemens Networks



One thing that doesn’t get discussed much is spectrum utilization as opposed to network utilization. In a single geographic location, one carrier’s spectrum allocation may be at capacity which another carrier’s is under-utilized.

The FCC needs to insure maximum spectrum utilization in any given location. This means two things:

1. That all mobile phones approved for use in the US can operate on all available frequencies allocated for their use irregardless of carrier.

2. That appropriate data roaming algorithms are developed and equable revenue sharing agreements are in place when a phone is forced due to data congestion to roam off its home network onto an underutilized one.

With some engineering and arm twisting, we could greatly lesson the effects of the oncoming “data tsunami.”

Harri Holma

We see WiFi offloading as a near term topic that going forward will be an integral part of wireless access. NSN already has an offloading solution (see and we will have an ever closer cooperation in the future. Looking Beyond 4G by 2020 requires far higher capabilities in terms of data rates, efficiency and latency than existing 3GPP 3G/4G networks or IEEE WiFi radio technologies. Future radio access networks need to be able to utilize both licensed bands (like 3G/4G networks) and unlicensed bands (like WiFi).

Harri Holma, author of NSN Beyond 4G blog

Volker Joksch

Hi Stacey, NokiaSiemens sells solutions around offloadig mobile traffic via WiFi – launched Feb. 2011 in Barcelona. Cheers, Volker Joksch


It is true that network is the part of the problem. We can’t rely on only one sources of network. For example iPhone able to send push with 3G (via SMS protocol) and WiFi (via Internet connection). so, At least they have 2 system backup.


They are over-managing their network. It’s been shown over and over that more capacity beats more management every time for cost-effectiveness. It is particularly ironic that AT&T got caught with their own numbers indicating that it costs 10% as much to build out an LTE networks as it does to buy a competitor.

Andrew Cutler

Stacey, great point about how both articles ignore Wi-Fi. T-Mobile U.S.A. has taken a step in the right direction by supporting Wi-Fi calling (UMA), which recent inclusion of an OTA Wi-Fi calling app for Android based phones. What’s perplexing is that this Wi-Fi app has yet to be made available to the Google Nexus S (Google’s showcase Android development machine) despite the Nexus S originally making its debut with a T-Mobile USA 3G radio modem. So it seems Google is also partly responsible for holding back innovative Wi-Fi telephony (not just NSN and Ma Bell Labs).

Steve Crowley

Here’s some info on NSN’s Wi-Fi offloading solution that the author of the above presentation may find useful:

This is one of several cases of two departments within the same company at odds. Cisco is another one; the Wi-Fi folks roll their eyes at the 4G core network people who prepare the Visual Networking Index, and underestimate Wi-Fi offloading.

Michael Schmidlen

Here’s a suggestion: How about knowing the device capabilities of the smart phones & tablets that are connecting to the web servers and only sending relevant (based on the actual connected device capabilities and the actual physical location, which may or may not impact the devices performance) content to these devices. This would substantially reduce the amount of info required to be transmitted…


Here’s the thing. Mobile and Cloud app developers don’t care (and in some cases don’t know) about the network. What gets transmitted and how it gets to the other side is the responsibility of the operators. Given you an example. We put together this quadrant that compares the user/network friendliness of apps and you can see what’s good for the user is not always friendly on the network:

– @pcapr

Steve Crowley

Not knowing about the network is part of the problem. It costs the operator some amount of money to send each bit of information. The app developer largely doesn’t care today, but apps can be coded more tightly to use less data in many ways. For example, instead of a web app sending separate HTML requests for each button on a screen, all the buttons at once on one picture can be sent using one HTML request, with the buttons cut out in the app.

There are also large differences in data requirements depending on phone operating system with Blackberry using the least of all, due in part to data compression implemented in its network. With the new tiered pricing plans, maybe this distinction becomes more important to the consumer, who has long been insulated from these differences.

Comments are closed.