Startup Tackles 4G Backhaul Bottleneck

Israel-based fabless semiconductor startup DesignArt Networks launched today a WiMax silicon platform aimed at combating 4G backhaul costs and capacity constraints and announced plans for a similar product for Long-Term Evolution (LTE). (See 4G Backhaul: A Problem for All?)

DesignArt’s new WiMax System-on-Chip (SoC) platform integrates base station and backhaul capabilities, which means operators will be able to deploy smaller sized base stations that can be used for backhauling traffic as well as serving customers. The technique is called in-band backhaul, because some of an operator’s spectrum is hived off and used for backhaul. It’s also referred to as in-band relaying or multi-hopping.

The backhaul cost and capacity crunch is especially problematic for so-called 4G WiMax and LTE deployments because these technologies require smaller, more densely deployed base stations. In-band relaying is designed to address these issues. (See Clearwire's Backhaul Bet.)

“As you go to next-generation radio access and true broadband speeds, the only way to get those speeds is to have smaller cell sites,” says Gabriel Brown, senior analyst at Heavy Reading. “The problem with smaller cells is that you have to backhaul them all. For all next-gen systems, people are investigating relaying... It has some potential.”

An in-band backhaul network would typically be deployed in a mesh architecture, but DesignArt says its solution could also be used in a point-to-point long-haul deployment.

The startup claims its new product offers big backhaul cost savings.

“We’ve got a complete base station on one piece of silicon, and we’ve added the backhaul,” says Joachim Hallwachs, DesignArt’s vice president of marketing. “It’s extremely low-cost because of a single piece of silicon and low power consumption.”

But it will be a few years before operators are able to deploy an in-band relaying network, according to Brown. DesignArt, for one, has just unveiled its silicon and it will take about a year before an equipment vendor uses it to build WiMax base stations.

The Institute of Electrical and Electronics Engineers Inc. (IEEE) is also working on a standard for in-band backhaul, called Mobile Multihop Relay (MMR) or 802.16j, and it’s expected to be completed by the end of this year or early next year, according to Hallwachs.

DesignArt is also working on an in-band backhaul product for LTE base stations. “We expect in-band relaying to be much more pronounced with LTE than with WiMax.”

DesignArt was founded in 2006 and employs about 30 people. The company’s original venture capital investors are Motorola Ventures, Carmel Ventures , and Magma Venture Partners. In December 2007, Motorola Ventures joined the original investors in round B funding, which was earmarked for launching the company and the first WiMax product, as well as starting development work on an LTE product. But the company would not reveal how much money it has raised in total.

In-band relaying for LTE and other backhaul issues will be discussed at Light Reading's Backhaul Strategies for Mobile Operators: Europe conference in London later this month.

— Michelle Donegan, European Editor, Unstrung

Interested in learning more on this topic? Then come to Backhaul Strategies for Mobile Operators: Europe, which will provide a unique perspective on the progress that Europe's carrier and vendor community is making in relieving the so-called "backhaul bottleneck" in mobile networks. To be staged in London, June 26, admission is free for attendees meeting our prequalification criteria. For more information, or to register, click here.

joachimh 12/5/2012 | 3:39:04 PM
re: Startup Tackles 4G Backhaul Bottleneck Operators expecting activities '08/'09, real deployments in '09.

WiMAX - yes macro cellular network deployments are indeed well underway. Evolution: this will be a (close) step 2 as radio access underlay network - initially metro fill-in and also campus/malls/airports, then tier 3 markets, rural extensions, ... autonomous extensions.
In the end underlay drives most of the service capacity to customers... macro drives WAN mobility...

LTE - not so sure... Might start out straight with femto and other "underlay" deployments, using 3G network for (initial?) WAN coverage and mobility... autonomous cell clusters are very cost-efficient entry into high-capacity service infrastructure, localized where needed. Low cash-flow, on-demand deployment and investment model ... dual-mode MS/ES... wouldn't dismiss it quite yet.

Yes - they "sign-up" to the existing network, e.g. based on open standard WiMAX. Then OA&M / NOC interaction - can be fully automated / pre-configured. Everything is OTA - that's key point, installation crew only takes care of mounting & low-voltage electricity.

In-band relay, based on 802.16j MMR will of course be an open IEEE standard.

However, most interest found to be in our more efficient mode, based on much higher spectral efficiency, proprietary - but high-speed, multi-hop backhaul capabilities - significantly better coverage and much more distributed capacity within clusters...
joachimh 12/5/2012 | 3:39:07 PM
re: Startup Tackles 4G Backhaul Bottleneck Spectral efficiency is actually where in-band service distribution provides the most benefits.

We are talking about the service distribution layer - in other words "how much capacity density can I offer to the market based on my spectrum asset?" Utilizing high-capacity service distribution to deploy more base stations actually improves the utilization of the spectrum asset. Capacity density can be expected to improve by up to an order of magnitude - simply dependent on how many small cells are deployed.

Thus efficient service distribution provides seamless market coverage and much increased capacity in an SLA grade solution. As an example, we would expect clusters of WiMAX cells, using a single 10 MHz channel, to be able to distribute an overall capacity of several 100 Mbps to indoors and outdoors users, at a fraction of traditional deployment costs. In-band service distribution requires no additional backhaul equipment or (narrow-beam) radios, supporting adaptive multi-hop tree and cluster topologies which will grow driven by service demand over time.

On the other hand, the DAN2400 SoC platform is frequency agnostic and supports in-band and out-of-band backhaul also in unlicensed spectrum.
IPobserver 12/5/2012 | 3:39:07 PM
re: Startup Tackles 4G Backhaul Bottleneck Joachim,

What sort of timeline do you see for this?

The first mobile WiMax and LTE networks will follow standard cell-based deployments.

WouldnGÇÖt these devices need to auto configure? How do you standardize that?

I know some vendors are saying auto-config will come first using their own gear. Getting interop seems a long way off.
wap545 12/5/2012 | 3:39:08 PM
re: Startup Tackles 4G Backhaul Bottleneck "..some of an operatorGÇÖs spectrum is hived off and used for backhaul".
What am I missing here?
Licensed Spectrum is finite and very expensive so what value is it to tie up limited and incredibly expensive licensed spectrum for backhaul when one can easily use free spectrum (3.65Ghz, 5Ghz series etc.) in PTP or even PTMP deployments for backhaul and save all your high $$$ spectrum for 4G services.

Sign In