Bookmark & share

By Julien Grivolas

Several Long Term Evolution (LTE) and Ultra Mobile Broadband (UMB) demonstrations have been performed during the last couple of months, putting the emphasis on what is likely to be the next technological step for the 3GPP and 3GPP2 camps. This comes at a time when mobile WiMAX is being increasingly and aggressively marketed as a pre-4G/4G technology. Senior Analyst Julien Grivolas provides an overview of "beyond 3G" and assesses the positioning of the different candidate technologies.


Ovum view

At a time when mobile operators are focusing on deploying HSDPA and EV-DO Revision A in their networks, there is already lots of speculation about what the next evolutionary step will be. This phenomenon is not new, as operators are always looking forward to how their services and networks will evolve, but it is nevertheless somewhat amplified due to the emergence of mobile WiMAX.

Sprint Nextel is seen as the figurehead of mobile WiMAX in the US, since announcing its $3 billion support at CTIA. The operator has now provided more details on its aggressive rollout strategy, on the size of its device ecosystem and on the reasons for its decision. Even if we think that the marketing campaign associating mobile WiMAX with 4G is a bit exaggerated, as mobile WiMAX performances do not match those expected for 4G, it remains true that mobile WiMAX will bring some of the technical advantages highlighted by next-generation LTE/UMB cellular technologies to the market earlier.

In a context under which several CDMA operators have recently migrated towards UMTS/HSPA for scale reasons, Sprint Nextel's choice of mobile WiMAX adds more pressure on the future success of UMB. However, even if 3GPP LTE is described as the natural evolution path for UMTS/HSPA operators and can leverage on the large scale of 3GPP technologies, some operators want to see an acceleration of LTE development in order to avoid losing market share in the mobile broadband Internet market against mobile WiMAX.

Large UMTS operators can also be tempted to use the WiMAX threat to further increase their influences upon the development of "beyond 3G" technologies. Moreover, several mobile operators decided to create a consortium, called the Next Generation Mobile Network (NGMN), in order to specify their requirements and define a fairer IPR environment for "beyond 3G", regardless of who the technology candidates are.


"Beyond 3G": the landscape

Increasingly positioned as a 4G technology by some vendors and operators, mobile WiMAX is considered by its backers as the main potential competitor to LTE. These parties argue that 16e provides similar capabilities while benefiting from better time to market. The mobile WiMAX standard was approved a year ago and certified products are planned to be available in the near future - compared to LTE or UMB, which are expected in 2009/2010.

This time-to-market advantage has been highlighted by Sprint Nextel as a key reason for the selection of the technology and the US operator is confident that it will leverage this first-mover advantage.

At a Congress keynote session during 3GSM in February 2007, Vodafone CEO Arun Sarin highlighted the need to accelerate the development of LTE in order to minimise the threat of WiMAX.


Main benefits of "beyond 3G" technologies

The vendors are highlighting similar benefits for the three competing technologies. These are: Higher capacity resulting in lower cost per bit; Flatter IP-based network architecture resulting in lower CAPEX/OPEX; Higher speeds, lower latency and better QoS enabling the provision of high-demand applications with greater user experience.

A key difference of mobile WiMAX is its lack of legacy infrastructure. Both LTE and UMB are marketed as natural evolutions of current 3GPP and 3GPP2 technologies. When compared to UMB, the LTE camp is, as usual, mentioning the scale advantage of 3GPP against 3GPP2, which will ultimately enable better economies of scale, in particular in the CPE domain. However, the WiMAX camp has done a great job of being able to create a large ecosystem that includes, among other chipset vendors, device manufacturers and consumer electronics vendors that are committed to embedding WiMAX capability in various types of devices and in volume.

Furthermore, there is no doubt that Sprint Nextel's selection of mobile WiMAX was a strong signal for the industry and brought more credibility to the technology. It will contribute to accelerating the development of 16e-enabled end-user devices and, thanks to volume, lower the costs.

However, even if this announcement is very significant for WiMAX, the fact remains that several points have to be addressed by the WiMAX backers. A key point is the certification process of 16e product, which is the only way to ensure product interoperability and consequently contribute to facilitating the economies of scale by encouraging competition.


Update on LTE work progress

Initially due to be closed by the end of June 2007, the LTE work is now expected to be concluded in September 2007.

In fact, since the 3GPP meeting held in June 2006, the basic requirements have already been agreed; however, further decisions on dedicated items have been added. For example, regarding MIMO smart antenna technique, LTE allows the use of up to four antennas at the mobile station side and four antennas at the cell site. The working group also focused on further enhanced multimedia broadcast multicast service (MBMS).


Update on UMB work progress

In December 2006, 3GPP2 published a white paper setting the technical evolution objectives of the standardisation body for the next ten years. It includes not only the ability to carry native IP packets over the radio air interface at data speeds of more than 200Mbit/s, but also deals with the evolution of core networks, the QoS and security issues, among many other subjects.

UMB, the brand name of the CDMA2000 1XEV-DO Revision C standard, is the first iteration of this vision, with data rates of up to 280Mbit/s for downlink and up to 68Mbit/s for the uplink. UMB combines aspects of various technologies (CDMA, TDM, LS-OFDM, OFDM, and OFDMA) into a single air interface and supports advanced antenna techniques such as MIMO and SDMA. As LTE, UMB also supports flexible bandwidth regime.

The UMB standard is currently being developed by the 3GPP2, with the publication scheduled for Q2 2007 and commercial availability expected during H1 2009. At CTIA 2007, Qualcomm unveiled its MDM8900 modem solution supporting DL speeds of up to 40Mbit/s and UL speeds of up to 10Mbit/s; its CSM8900 solution for base stations supporting 4x4 MIMO and serial input/output. These solutions are expected to be sampled by Q1 2008 and Q2 2008 respectively.


Update on mobile WiMAX work progress

Initially positioned as a complementary or competitive technology of HSPA, mobile WiMAX has increasingly been compared to LTE by some vendors. Why? Some will say because HSPA has already won the battle as there are more than 100 commercial HSDPA networks up and running with HSUPA currently being deployed. Others will say that the comparison with HSPA is not relevant as the technologies are very different and the performance of mobile WiMAX is much higher. Furthermore, comparing LTE and mobile WiMAX makes more sense as they are both OFDM/MIMO based and IP native technologies.

Mobile WiMAX is the brand name of the IEEE 802.16e-2005 standard that has been ratified in early 2006. The WiMAX Forum has defined profiles for certification. The WiMAX Forum Certified stamp on equipment means that the product is interoperable with other certified products for a selected profile. certification is crucial to the success of WiMAX as it promotes economies of scale for the selected profiles defined by the WiMAX Forum while encouraging competition.

Initially, certified products compliant with the Wave 1 mobile WiMAX certification process were due to be available in late 2006. However, the first certified products based on the 2.3GHz profile are now expected for mid-2007, with market availability of WiMAX Forum certified mobile products beginning in late 2007. In order to accelerate and facilitate the certification process, the WiMAX Forum has organised two mobile WiMAX Forum PlugFest events. The latest was hosted by AT4 Wireless Labs in Spain in February 2007, and attracted 35 companies.

Regarding Wave 2, the certification process has not begun yet and there is lots of speculation about when Wave 2 certified products will be available. It's interesting that some of the key benefits, for example, the support for MIMO and beam-forming smart antenna techniques - highlighted by the mobile WiMAX vendors are only part of the Wave 2 profiles. The most aggressive vendors hope to achieve certification in late 2007, while others are claiming that it could not be available before mid-2008.

However, products compliant with Wave 2 will be available before they achieve certification. At 3GSM and CTIA, there were many mobile WiMAX demonstrations based on Wave 2 features. For example, at 3GSM Samsung demonstrated "Mobile WiMAX Wave 2" systems and terminals enabling data speeds of up to 34Mbit/s for downlink and 8Mbit/s for uplink, with theoretical speeds of up to 40Mbit/s and 12Mbit/s respectively.

Several chipset vendors, such as Sequans, Beceem and Runcom, have unveiled Wave 2 chipsets. Huawei claims to be technology agnostic and plans to support all major next-generation technologies: UMB, LTE and WiMAX. For example, Huawei expects to have commercial UMB products available by 2009.


Operators and "beyond 3G" technologies

Naturally, vendors are positioning themselves around technologies, but what are the views of operators regarding "beyond 3G"? First of all, some of the most powerful operators want to play an active role and a handful of them have decided to create a consortium, called Next Generation Mobile Networks (NGMN), in order to influence the specification work and create a favourable environment for "beyond 3G" technologies.


Next Generation Mobile Networks

NGMN was founded in 2006 by China Mobile, KPN Mobile, NTT DoCoMo, Orange, Sprint Nextel, T-Mobile International and Vodafone. During 3GSM in February 2007, Alltel, AT&T, SK Telecom, Telecom Italia and Telefonica joined the consortium as new members.

The creation of NGMN followed a letter sent in November 2005 and co-signed by the CTOs of Vodafone Group, Orange Group, T-Mobile Group and KPN, requesting that the IPR question must be solved before specification begins within 3GPP with regard to LTE. Service providers, which are going to have to invest significantly in hardware if they want to roll out LTE, want to limit the influence of IPR licensing on equipment pricing and avoid what happened with WCDMA. Operators are pushing for the acceptance by the industry of a pre-agreed cumulative cap of 3-5% for all LTE equipment. The ETSI Working Group is currently examining the need for an updated IPR regime for standards.

However, the scope of NGMN is wider than just IPR and LTE; the consortium aims at promoting open standards for "beyond 3G" as a whole. Technologies ultimately selected by NGMN members should be chosen on the basis of considerations such as performance, price and quality. NGMN is complementary to the standard bodies by providing a view of what operators' requirements beyond 2010 will be.

NGMN has defined a network architecture based on a new IP core network and a new radio access network supporting full mobility features. NGMN architecture will initially co-exist with existing circuit-switched voice infrastructure and existing radio access networks, and ultimately will replace the circuit switched network part. To achieve NGMN requirements in accordance with the NGMN system, NGMN has defined detailed recommendations for both the access and core parts of the network.

In terms of timeframe, NGMN plans standard specifications to be completed by late 2008, with systems available for trials for operators in 2009 and commercial rollouts expected by 2010.