The pros and cons of LTE, according to ZTE

The pros and cons of LTE

by Martyn Warwick

Despite the fact that Long Term Evolution (LTE) has basically won the propaganda war against WiMAX and its proponents say it is now set to become the next generation mobile platform, deploying it in a real world setting and then ensuring that its theoretical performance benefits are translated into practical reality will be a tough task and take a considerable time. At least, so says a senior ZTE executive.

Zhen Donglin, the vice president of wireless technology at ZTE USA, says that while increasing numbers of operators have announced their intention to migrate to LTE, what they are migrating to isn’t necessarily the same platform. Operators in different markets have differing spectrum allocations, forcing them to deploy LTE networks with different carrier sizes and this, naturally enough, results in different performance capabilities.

“In Europe, operators are looking to go to market with 20MHz, because they have 2.6GHz spectrum. After talking to some of the European operators, I think that is what is in their mind. At the world LTE summit in Berlin last month, a lot of operators were talking about initial 20MHz deployments, but that's Europe,” Mr. Zhen told our sister publication Commsday Asia.

He continued, “In the US, things will be different. For the mainstream operators such as Verizon and AT&T, they will have 5MHz-10MHz deployments in the beginning – that is basically because of spectrum limitations: they don’t have any more spectrum. For some second tier and third tier CDMA operators, they are talking about starting with 1.4MHz deployments. Different operators have different things in mind.”

Mr. Zhen believes that this situation seems likely to result in different handsets being provided for different deployment scenarios, with the industry association, NGMN (Next Generation Mobile Network) now proposing separate LTE terminal classes.

He says, “For the handsets, the NGMN has defined four classes. Each class will have different data speeds for bandwidth due to the initial spectrum they can support. I think it will be mandatory for the handset to have the capability to work with 20MHz carriers although data throughput might not be that high. They have to be capable of working with 20MHz, otherwise if you only have a 10MHz handset and you go to a 20MHz base station, they can’t talk to each other – and that’s not good.”

He added: “Eventually, the handset has to support 20MHz, but the throughput doesn’t have to have such high data rate.

The same handset has to work with 1.4MHz spectrum as well as 20MHz spectrum, I think that must be mandatory.”

Zhen Dunglin also says that whilst LTE is expected to be ready for commercial launch before 2009 or 2010, it is highly unlikely that GSM operators will go directly to LTE without first transiting through WCDMA.

“There is some discussion about GSM going directly to LTE and bypassing WCDMA. People can think about it, but in reality it doesn't provide any benefit to the operator because you have GSM, EDGE with data capacity and pretty good coverage, but it’s still narrowband, and these standards cannot provide broadband wireless capabilities. WCDMA though is broadband and it’s apretty good technology. What's more, operators deploying WCDMA have better capex and opex and WCDMA migrates to HSPA, and that’s a major high speed data platform for the mobile industry. LTE, it’s broadband, but initially, it will be deployed in hotspots in urban areas, but even then how can you provide continuous data communications capability? You
still need WCDMA to fill out the gaps for LTE for a while.”

Zhen believes GSM will eventually be replaced by WCDMA for reasons of cost if not because is a superior technology.

He says,“GSM is good and cheap and it’s everywhere, but will GSM last forever? It's hard to say. I think it will n be here for another 10 to 15 years, but for an operator running operating a GSM network from an opex point of view, it is more expensive than operating a WCDMA network. "If you take 5MHz of spectrum and divide that by 200khz (for GSM), you get 25 channels, so you need 25 transceivers. With WCDMA, its one channel. That means capex and opex is lower. Wideband will always give you better economy of scale than narrowband.”