Believe it or not, 4G is coming. Before many of us even have 3G cellular service, the outlook for 4Galthough still unclearis starting to come into focus. It now appears that the first 4G services could arrive, not between 2010 and 2015 as originally planned, but quite a bit sooner. Parts of 4G, in fact, are taking shape right now.
What will 4G be like? Eventually, according to plans of the International Telecommunication Union, it will provide 100 Mbps data rates when you're mobile and 1 Gbps when you're stationary. It will support mobile phones, of course, and also mobile broadband Internet and multimedia services. Even mobile HDTV is a possibility.
Before reaching its full potential, however, 4G is likely to evolve in steps that some technologists are calling 3.5G (Figure 1). The bad news in this is that the most spectacular capabilities of 4G still remain quite a few years out. The good news is that we could have performance significantly beyond that of 3G before 2010, with some of it coming in the next couple of years.
Figure 1: Existing wireless services will evolve toward 4G with increases in data rates and mobility. WiBro, a subset of future standard 802.16e, specifies mobility for the WiMAX technology. (Illustration courtesy of Samsung Electronics Co., Ltd.)
This new 4G outlook stems primarily from new technology developments, such as WiMAX, that are changing not just the schedule, but the possible makeup of 4G. Originally promoted as a wireless replacement for DSL or cable Internet access, WiMAX now is being touted as the very basis of 3G and 4G mobile technology. It can outperform current and planned cellular technologies, WiMAX proponents say, and it will be less expensive to deploy. It can handle high-speed data, of course, as it was intended to do, andwith VoIP (voice over Internet protocol) technologyeven voice.
Other wireless technologies, such as Wi-Fi and Bluetooth, are also contending to be part of 4G. By most accounts, 4G will be a collection of technologies (Table 1) that will allow connection by wireless devices of all types. Furthermore, each type of wireless device will use whatever wireless network is most appropriate at any given time. A mobile phone, for example, will use VoIP on Wi-Fi when it's near a Wi-Fi access point and use the cellular voice network when it's not. This capability, in fact, is already starting to be implemented on 3G and 2.5G networks.
Table 1: Wireless technologies (Data courtesy of Intel Corporation)
What's surprising about this new 4G outlook isn't that it incorporates technologies from the computer world, but that it actually has the potential to displace technologies from the traditional cellular world. The mere possibility of WiMAX dominance in 4G, for example, is causing some cellular operators to consider using WiMAX in 3G, which isn't yet fully deployed despite long-laid plans. Although most cellular companies will stick with their plans to implement previously defined 3G technology, they now have to be concerned about ensuring an easy and cost-effective migration path to a 4G that might be based on WiMAX.
Within the 3G timeframe itself, WiMAX's impact is still uncertain. Some market analysts think it can make significant inroads, but others, such as Allen Nogee, principal wireless analyst for In-Stat/MDR, are less encouraging. "The 3G networks are in their early stages," Nogee notes, "but they're coming out, they're physically here." And, Nogee adds, it has taken cellular 20 years to get to where it is, and "WiMAX, just springing up overnight, is going to have a tough time competing against these well established carriers."
It's also becoming clear that WiMAX will not have a primary role in 3G voice service. Although Intelone of WiMAX's most enthusiastic backershas argued that the technology is, or soon will be, superior to 3G technology in many ways, the company's web site stresses that WiMAX's primary 3G role will be for data functions. According to Intel communications group vice president Jim Johnson, in an interview posted on the site, "Our objective for WiMAX ... is to be a high speed data service that can be used to extend and complement 3G service." Johnson continues: "We know of several 3G service providers that are worried that as data use grows they will not have enough spectrum for both their voice customers and their data customers. So they're interested in WiMAX as a complementary data service technology that they can deploy with their 3G voice service. So essentially, WiMAX is very targeted for wireless data, not wireless voice."
But 4G, compared to 3G, presents additional opportunities for WiMAX. Not only is it less encumbered than 3G with legacy technologies, it's also far enough in the future for WiMAX to evolve toward 4G requirements. "4G will probably be a lot like WiMAX," says In-Stat's Nogee. "It's very possible that [WiMAX and 3G] will converge together in some kind of fashion." In that case, Nogee adds, WiMAX would exist in a 4G form that wireless carriers have embraced, rather than a competitor that they're fighting against.
Before WiMAX can be a serious 4G contender, however, it will have to add mobility features that it currently lacks. Otherwise it won't be able "hand off" a user who moves with a mobile phone or data device from one wireless cell to another. Although WiMAX will soon be adding mobility in the form of IEEE standard 802.16e, broad application of the standard is probably still a couple of years off.
In the meantime, WiMAX will move toward mobility in stagesfrom fixed to portable to nomadic to fully mobile (Figure 2). The fixed phase of WiMAX, in which the technology's main use is basically as a wireless replacement for DSL or cable Internet access, will begin this year. The portable and nomadic phases will probably begin deployment in 2006, as will mobility for data applications, but not voice. The timetable for full mobility is still uncertain.
Figure 2: Intel's vision for WiMAX deployment includes phases for fixed access, portability with simple mobility, and full mobility. (Illustration courtesy of Intel Corporation)
Each new stage of capability in mobility will require improvements to WiMAX. Even simple portability, for example, in which a user moves to a new location and manually links to a new access point, will require enhancements to security, such as strong mutual authentication. In the next phasenomadicity or portability with simple mobilitya user moves at pedestrian speed to a new location with session persistence or automatic reconnection to a new AP. This capability, when implemented, will enable continuing data connections, but it won't meet the low-latency requirements necessary for applications like voice and multimedia.
Full mobility for WiMAX presents a whole new set of problems. In a fully mobile situation, a user will be able to engage in activities such as voice conversations or multimedia streaming while moving from one access point to another. The challenge for WiMAX in these cases will be to automatically hand the user over from one AP to another without dropping data packets and while maintaining low latency. As noted in an Intel technology paper, "Global, Interoperable Broadband Wireless Networks: Extending WiMAX Technology to Mobility," this is "clearly a challenging task."
Fortunately, work is progressing on proposed IEEE standard 802.16e, which will define WiMAX mobility. Late last year, WiMAX heavyweight Intel agreed to let 802.16e be based on WiBro (Wireless Broadband), a technology developed in Korea by Samsung Electronics and LG Electronics. Without the Intel-Korean compromise, the standard might well have languished, causing WiMAX to miss market opportunities. It now appears, however, that 802.16e will receive approval sometime this year.
WiBro is only a beginning for WiMAX portability, but it's a good beginning. It doesn't handle mobile cellular voice calls, but, according to Samsung, it will allow wireless users to travel across large geographical areas at speeds up to 75 miles per hour while receiving broadband data and multimedia content. Samsung, in conjunction with three wireless carriers already selected by the Korean government, intends to commercially launch WiBro high-speed data access systems and handsets in Korea next year with data throughput speeds of up to 30 Mbps. According to London-based Rethink Research, which specializes in wireless market analysis, that puts Samsung and LG in a position to deliver "real-world 4G before 2010."
The cellular industry, of course, isn't sitting on its hands while WiMAX develops. For example, a technology called High Speed Downlink Packet Access (HSDPA), a technology upgrade to 3G's Universal Mobile Telecommunications Service (UMTS) networks, now enables downlink data rates of 10 Mbps and higher. And unlike WiMAX, HSDPA doesn't require new cellular infrastructure.
It's quite possible, though, that that the real beginnings of 4G are taking shape outside the realms of both 3G and WiMAXin Wi-Fi. Although VoIP has been slow to take off commercially, it seems now to be on the verge of a big growth spurt thanks to Wi-Fi's huge success. In doing so, VoIP on Wi-Fi (sometimes called VoFi) could establish Internet protocol (IP) as the unifying lingua franca of the wireless world, demonstrating that one technology can serve multiple purposes.
Wi-Fi with VoIP will be only a tiny step, however, because Wi-Fi has limitations: very limited range, no mobility, anddespite recent improvementsno guaranteed QoS (quality of service). WiMAX, on the other hand, has range out to several miles, built-in QoS features, and upcoming provisions for mobility. In short, it looks a lot like cellular technology.
And Wi-Fi is starting to get some mobility of its own, a development that could create consumer expectations for mobile VoIP that would benefit WiMAX down the road. Several industry groups are developing technology and products for VoIP on Wi-Fi that can automatically switch voice calls or data sessions back and forth between Wi-Fi access points and a cellular network. The dual-mode handsets and the network equipment needed to handle the switches are already starting to appear.
One of the more prominent Wi-Fi VoIP groups-called UMA, for Unlicensed Mobile Access-includes players like Motorola, Nokia, Nortel, British Telecom, Cingular, Siemens, Sony Ericsson, and T-Mobile US. UMA technology provides access to mobile services on GSM (Global System for Mobile) and GPRS (General Packet Radio Service) over unlicensed spectrum technologies, including Bluetooth and 802.11. The support for GSM and GPRS reflects the current commercial dominance of those technologies; the principles of UMA technology also apply to CDMA technology.
With UMA technology (Figure 3), handovers between a cellular network and a Wi-Fi or Bluetooth network are completely transparent to the user. If a user with a UMA-enabled, dual-mode handset moves within range of a Wi-Fi or Bluetooth network that it's allowed to connect to, it will automatically connect. After connecting, the UMA Network Controller (UNC) allows the handsetif appropriateto access GSM voice and GPRS data services via the Wi-Fi or Bluetooth network. If the handset moves outside the range of the Wi-Fi or Bluetooth network, the UNC and handset will facilitate roaming back to the licensed cellular network. Similarly, if a user is on an active GSM voice call or GPRS data session and comes within range of a Wi-Fi or Bluetooth network, the voice call or data session will automatically switch to that local network.
Figure 3: Unlicensed Mobile Access (UMA) technology permits dual-mode handsets to roam between cellular networks and local networks, such as Wi-Fi or Bluetooth, that operate in unlicensed spectrum.
The roaming that UMA's automatic handovers permitalthough limited to the small areas of Wi-Fi and Bluetooth coveragedemonstrate the potential for roaming on a larger scale with WiMAX. If VoIP on Wi-Fi does take off, as many market analysts expect it to this year, it seems entirely likely that VoIP will also move to WiMAX when WiMAX networks start appearing. In fact, although many early WiMAX deployments will simply provide wireless broadband Internet access, WiMAX service providers might well bundle VoIP with their service. According to In-Stat/MDR, they will need to in order to compete successfully with DSL and cable.
Perhaps the most important aspect of Wi-Fi with VoIP, however, is the IP part. VoIP on Wi-Fi sets up a real-world connection between cellular and Internet protocol that could help prepare for an eventual key role by IP in G4. VoFi could be the beginning of cellular technology that's actually based on IP.
About the Author
Gary Legg is a Boston-based freelance writer. He holds a BSEE degree and is a former editor and executive editor of EDN magazine. He can be reached at firstname.lastname@example.org