Taqua reinvents itself with new take on radio backhaul

4G is presenting an opportunity for many companies to reinvent themselves, pursuing product lines they previously never touched or only tackled through partnerships.

Harris Stratex changed its name to Aviat Network (NASDAQ:AVNW) to mark its move from microwave backhaul into end-to-end WiMax networks. Juniper Networks (NASDAQ:JNPR) produced its first stand-alone 4G packet core business after years of making 3G cores with partner Ericsson (NASDAQ:ERIC). And Tellabs (NASDAQ:TLAB) branched out from wireless backhaul into the mobile core with the purchase of WiChorus.

The latest company to reinvent itself for 4G is Taqua, a next-generation wireline and wireless switching vendor that had previously never built a radio in its life but this week announced its first wireless backhaul product. Taqua, however, isn’t producing the typical long-haul point-to-point radio used to connect cellsites miles away. Rather Taqua is proposing its platform be used in new cellular topologies that make heavy use of picocells, outdoor femtocells and even Wi-Fi hotspots. Its solution wouldn’t replace the backhaul network itself, but rather function as last mile extension for backhaul linking up remote sites.

Providing adequate backhaul links to dozens of small cells is much more expensive than providing one big link to a large cell site, though, which has made these kind of hyper-cellular architectures few and far between, explained John Hoadley, Taqua’s new wireless chief technology officer and former wireless CTO of Nortel. Taqua’s architecture relies on small, short-distance links using a variation of the IEEE 802.16e technology used for WiMax. Using inexpensive but licensed time division duplexing frequencies such as the 2.5 GHz and 2.6 GHz bands, an operator can overlay a low-cost backhaul network over their macro-cellular footprint, which could then be used to support numerous small cells on the macro cells’ fringes, Hoadley said.

“We’re at a point that devices and applications are miles ahead of the infrastructure—it’s usually the other way around,” Hoadley said. In order to meet those increasing demands for wireless data, operators are having to deploy more infrastructure, creating more wireless carriers to support more traffic on the same sites. But the logical thing is for operators to shrink down their networks, spreading the increased traffic among many more much smaller sites, Hoadley said. “We’re on a path where we have to do very dense deployments using small cells,” Hoadley said. “It’s got to be about small cell cites and to do that we need better backhaul.”

The backhaul platform, which Hoadley’s team began developing after he left Nortel, uses many of the same techniques and technologies as WiMax, from orthogonal frequency division multiplexing (OFDM) to multiple-input multiple output (MIMO).The links are non-line of site, meaning it can punch through foliage and bend around obstacles, much like its mobile broadband cousin. But it’s designed to be a fixed point-to-multipoint system with a single hub delivering 40 Mb/s to 60 Mb/s of capacity shared by a cluster of two or more pico- or micro-cells. Ideally an operator would deploy the remote backhaul modules next to picocells mounted on lampposts sitting at the edge of a macrocell, Hoadley said. The backhaul hubs would sit on the cell tower, using the same fiber backhaul link to that feeds to the macrocell to access the network, Hoadley said.

The catch is that in order to function properly the system has to use licensed spectrum that’s been typically designated for broadband access. But Hoadley pointed out that much of the TDD spectrum its backhaul system would use is already lying dormant at operators that have deployed their 3G and 4G networks over frequency division duplexing bands. There are exceptions, such as Clearwire, but for the most part operators can get those TDD bands fairly inexpensively. “If you look at the TDD bands around the world, they’re anywhere from 30 to 100 times cheaper than FDD bands,” Hoadley said. “Many operators already have TDD spectrum that they don’t know what to do with.”