Long term evolution (LTE) is firmly on course to be the dominant wireless communications standard in the next decade. There is overwhelming operator commitment to adopting the LTE standard and capitalising on the unquestionable benefits it will bring to their subscribers, even more so than was demonstrated for its 3G predecessors. In this article Ebrahim Bushehri evaluates the prospects for the adoption of LTE by operators worldwide, and describes the challenges this presents for RF transceiver technology in order to make this a reality, along with an explanation of how these challenges can be met.
More than 20 of the leading operators worldwide are already committed to deploying LTE in around 120 networks, a figure that currently represents a total of around 1.8 billion users — approximately half of the world’s total subscriber base. The key operators who have already put their names to the standard include Verizon Wireless, Vodafone, T-Mobile, NTT DoCoMo, AT&T and China Mobile.
According to ABI Research, the number of W-CDMA and cdma2000 (3G) mobile handsets that will be sold during 2009 is expected to exceed 50% of total handset sales. In particular, despite the recession, W-CDMA handset shipments this year are expected to have almost trebled, to 725 million from just 258 million in 2008. By 2013, more than 67% of all handsets shipped will be capable of 3G, 3G+ or 4G communications, and of these more than 32 million will be to LTE subscribers.
Although the operators are pledging that LTE will not be the disappointment that 3G proved to be when it first went into service, there will be a delay before a full range of LTE services will be available to the subscriber. The key driver for LTE is undoubtedly data rate — speeds are expected to be typically 10 times faster than those enjoyed by today’s HSPA subscribers. But the aggressiveness with which operators seek to deploy LTE will be largely dependent on the history and status of their existing networks. Early 3G adopters such as NTT-DoCoMo who are keen to increase their network capacity, plan to roll out LTE at the earliest opportunity. Similarly Verizon Wireless’s cdma2000 networks are not as scaleable as European W-CDMA systems which explains Verizon’s eagerness to embrace 3G. In contrast, both W-CDMA and China’s TD-SCDMA systems still have the potential for further improvements in speed and capacity, and their progression towards LTE is expected to follow a more conservative course. Figure 1 shows the likely evolution routes for the different world standards, while Table 1 shows a comparison of data rate and channel bandwidth for the most common cellular standards.
GSM was originally focused on voice service, with limited data capability later being made available with the addition of GPRS and EDGE. UMTS mainly focuses on providing voice combined with a moderate speed data service, while LTE will provide a high speed data service, with mainly hotpot coverage initially. The US EV-DO (Evolution Data Optimized) technology is unable to match the speeds that are possible in W-CDMA/HSPA networks.
The challenge for silicon vendors
Whatever the rate at which the market will evolve, both handset and basestation manufacturers will be faced with meeting the need for flexible systems that can address multiple standards and multiple bands for many years to come. Historically, any new generation of devices provides back ward compatibility with the previous and this is set to continue for the foreseeable future. The provision of data services will require not only high data rates, but also in-building coverage and efficient usage of the available spectrum.
Analysts largely agree that LTE will be heavily dependent on small cells — picocells and femtocells — in order to meet both commercial and technical requirements, and this will require a new breed of chip sets. Femtocells offer many advantages: they can solve the indoor coverage issues with 3G, using indoor licensed spectrum; they can provide more capacity, coverage and services in the home or small office environment; and they can reduce OPEX by utilising residential IP network backhaul. They also open up the possibility of new home networking solutions, content and applications, while they have the advantage of using mature, standardised user platforms that are already available.
Alongside a femtocell-based deployment, LTE could also be used to ‘fill in’ in rural areas where there is no ADSL or FTTH. It is important to note that 3GPP has designated 40 bands each at uplink and downlink, at least fifteen of which could be used for LTE, and these range between 700MHz and 2.7GHz. Further bands down to 450MHz are likely to be re-farmed in the future from analogue applications, on a regional basis.
This presents a considerable challenge to silicon vendors, especially so of RF transceivers covering multiple frequencies within this range. Added to this will be multiple standards and multiple radio technologies for backward compatibility and covering other complementary air interfaces, in total spanning a range 450MHz to 4GHz. This is compounded by the fact that equipment will be required to support paired and unpaired spectrum on the same hardware, and to support numerous channel bandwidths for backward compatibility or other standards. Coverage and data rate will be a further challenge, and addressing this will necessitate a tiered approach using a high volume of small cells and also utilising MIMO. Overall cost constraints will be a major consideration regardless of the type of solution employed.
Meeting the challenge
The Lime Microsystems LMS6002FN RF transceiver, shown in the block diagram in Figure 2, is a multi-standard platform that covers all the frequencies, air interfaces and bandwidths that are currently allocated for 3G and LTE as specified by both 3GPP and 3GPP2. The transceiver meets the stringent radio requirements for such application, including: gain control resolution; dynamic range; frequency agility; Rx blocking performance, noise; and sensitivity; and Tx noise and linearity. The wideband design removes the need for individual transceiver chips for each of the different bands, and allows a small cell base station to be reconfigured rapidly and simply. As well as offering footprint and cost advantages over discrete solutions, the single-chip device provides a reduction in bill of materials and therefore minimises costs and inventory for OEMs. The transceiver also features auto-calibration of Tx and Rx channels.
Also available is an RF multi-standard enablement platform, shown in figure 3, that can be used to evaluate performance of the wideband single chip transceiver over the 375MHz-4GHz frequency range. The general purpose board allows programming of the modulation bandwidth to values of 1.5, 1.75, 2.5, 2.75, 3, 3.84, 5, 5.5, 6, 7, 8.75, 10, 12, 14, 20 and 28MHz.
A recently-announced collaboration with Blue Wonder has resulted in a complete LTE reference platform, shown in Figure 4, that is based on the LMS6002 integrated single chip RF transceiver and Blue Wonder's baseband subsystem and has been designed to allow OEMs developing LTE equipment to achieve a highly integrated solution in a minimum amount of time.
A high proportion of operators are committed to LTE with the likely deployment in 2011 and sometime before it achieves mass adoption. LTE is expected to be deployed as a small cell solution, but the proliferation of bands and standards will make LTE radios a complex problem to solve. Transceiver flexibility has been identified as a key enabler to making these systems a reality at the earliest opportunity.
Ebrahim Bushehri is CEO of Lime Microsystems, www.limemicro.com, +44 (0) 148 3685063