Feature Stories | Apr 03 2007
By Investech Technology Consultants
(This is part two of a two parts series. Both parts will be published in text format without accompanying graphs and charts. FNArena will publish a Special Report under the same title in pdf later this month which will contain similar content with the graphs and charts.)
The trade-off between Capacity versus Coverage
Both the WiMAX and 3G {PP} camps will argue the technological superiority of their respective technology in terms of capacity and reach. The world of wireless unfortunately is full of trade-offs. There is a trade-off between Coverage versus Capacity. Equally, there is a trade-off between power output and coverage. There is also a trade-off between frequency and coverage. Consequently, there is no “silver bullet” technology solution that can cost-effectively deliver large bandwidths in all environments. Coverage and Capacity (data speed) are a trade-off (see figure 1), such that if you seek capacity, your range will be reduced; and if you seek range, the data rate will be reduced. In this respect no singular technology can resolve the trade-off completely.
Improvements in technology such as NLOS, OFDM, MIMO, et al and better performance make it possible to move upwards on the Capacity-Coverage trade-off curve. But there is really is no escaping the fact that to resolve the capacity versus coverage trade-off to deliver higher rates to users while mobile will require building (or having access to) many sites and towers.
WiMAX propagates a signal that, at most, is only twice as powerful as existing WiFi 802.11a and 1/10th that of a 3G/UMTS base station. Due to this low output power and relatively high frequency band (>3.5GHz) the coverage footprint is less than 3km. It will be difficult for WiMAX to create coverage within and between buildings. The consequence is that in order to ensure proper operation the CPE antenna must be placed outdoors above roof tops. This results in problems of coexistence (including handoff) of base stations where other wireless systems are in close proximity.
The proponents of WiMAX also go to great lengths to expound the virtues of NLOS operation and the many deployment advantages over LOS. Indeed, one of the clear advantages of WiMAX , over its BWA predecessors, is its ability to operate under blocked or obstructed line-of-sight (LOS) conditions. Wireless Internet Service Providers do not have to engage in detailed cell planning, using high resolution, three-dimensional data that takes into account terrain, rain data and the footprint, location and height of buildings, and foliage. BWA links that use such technology can serve a greater number of sites.
While this is true, we believe that NLOS operation will always yield significantly lower throughput, due to the effect of multi-paths and interference rather than having a clear, unobstructed line of sight path between the two end points of the wireless link.
Several examples may serve to clarify. Let’s say the bandwidth allocated to a WISP is 10MHz and this operator uses a high modulation rate (of say 64 QAM), with a relatively high spectral efficiency of 1.91bits/Hz. It is conceivable that a data rate of 37 Mbps [downlink] per sector can be achieved. If the bandwidth allocation of the license was increased to 20MHz then the frequently cited 70 Mbps may be achievable. These are ideal (almost impossible to achieve in the field) and not real world figures. Unwired, with a network of around 80 base stations, owns a 10MHz license operating at 3.5GHz. It claims to deliver a modest 12 Mbps of downlink capacity per sector. Bear in mind that this is the total capacity of the entire base station sector, not the throughput available to each user, as is often implied. Users can expect real world rates of 1Mbps [downlink] and 345 K [uplink]. Furthermore these rates are achieved through the use of smart “beamforming” antennas provided by Navini.
The lower power and frequency of operation of WiMAX (802.16e) will necessitate the installation of far more towers, antennas and rooftop sites than is being presently predicted. To that end, 3G/HSPA has the first move advantage of leveraging the extraordinarily large network assets of 3GPP-based networks globally. In Australia, for instance there is already over 15,000 base stations (Telstra, Optus, Vodafone, Hutch, etc) deployed providing coverage to 98% of the population.
Achieving mobility is more than just tweaking the air-interface
To provide mobility support WiMAX 802.16e (Mobile WiMAX) uses a more sophisticated and scalable OFDMA (Orthogonal Frequency Division Multiple access) technique, that improves multi-path performance in a non-line-of-sight environment. OFDMA is similar to OFDM except separate carriers are assigned to separate users at the same time, thus achieving a better link budget.
Whilst, Mobility & Wireless are often used interchangeably, they are not the same thing. Wireless means, as the name suggests, without wires. This may seem obvious, but it’s one of those terms that is not truly or fully understood by the vast majority of the industry — or at least it is understood differently by various factions in the market. Whereas mobility means being able to take something with you. It also implies motion, and in the context of telecommunications, to then be able to use that something to communicate while being on the move.
In our view, full mobility is something that WiMAX cannot yet (and may never) match, even with WiMAX (802.16e). We maintain that achieving mobility is more than simply tweaking the air-interface with a new form of modulation namely OFDMA. There is an entire mobile management infrastructure required.
Both WiMAX (802.16e) and HSPA only define the air-interface (physical) and MAC (Medium Access Control) layer in their specifications. Higher layer protocols are required to manage the user’s attachment to the network as they move around.
When WiMAX 802.16e refers to mobility, it really means that it provides support for fast handover signalling that will allow pedestrian users and slow moving vehicles to seamlessly switch (or roam) between different base stations. But this is not the same mobility in the sense that we have come to know it. That is, realizing the appearance of seamless connections across cell boundaries without latency.
3G/HSPA has evolved from the GSM mobile communications world, drawing from 20 years of experience (and extensive 3GPP/UMTS specifications and network architecture) in providing mobility management. The standards cover all the key protocols and networking interfaces for facilitating the functions and issues that are specific to mobile cellular communications. These include, radio resource allocation, mobility management, numbering plans, call routing and signalling transport protocols, network databases internetworking, paging messaging systems, roaming, authentication and security protection, service definitions and accounting. Most of these functions are not defined by the IEEE and has been left to the WiMAX forum to sort out.
Volume economics
Wireless telecommunications is an R&D-intensive endeavour, where the stakes are high and returns are uncertain. The theory goes that prices fall by increased production volume, giving de facto winning technologies additional advantage. As electronic equipment becomes very cheap, maintenance, service and physical infrastructures as masts, buildings and cables will become the dominant part of costs.
Attempts at harmonisation of spectrum across regions, to try and help wireless standards to get broadly adopted, are critical to the ability of manufacturers to achieve volume economics. That is why GSM operates globally at 900 MHz, 1800 MHz and 1900 MHz; all the while 3G operates globally at 2100 MHz; thus allowing vendors to reap volume economics with minimal changes to devices.
We do not believe that volume economics in silicon for WiMAX have yet manifested itself; due to the small market penetration of the technology. Claimed declines in WiMAX based equipment prices, relative to proprietary BWA systems, are largely due to the lower frequency of operation (<5GHz) and use of inexpensive off-the-shelf components (achieved from other technologies operating in the unlicensed bands in which WiMAX hopes to operate).
Moreover, we maintain that focusing on equipment cost reductions, per se, to draw attention to network deployment advantages is misleading. We have illustrated that equipment costs represent only 10% of the deployment cost of a BTS. The predominant costs are made up of civil works, namely site preparation, towers and antenna systems, cabinets, and power. Unfortunately, volume economics in silicon do not apply to real estate, civil works and labour.
In order to offset pricing pressures and achieve volume economics Wireless Network Equipment Vendors (WNEV) need large market share and therefore prefer to sell into a large installed base, which is more profitable than competing for initial business. We submit that the fertile ground of almost 3 billion users, operating in harmonised spectrum, is incentive enough for WNEV and component vendors to pursue the 3GPP based cellular market, and drive down the cost curve, while simultaneously driving up the capabilities curve.
Availability of networks and devices – A chicken-and egg-problem
The availability of networks and devices, and the rate of user adoption is the tipping point which will ultimately determine the success or failure of the two competing technologies. The proponents of WiMAX (chief amongst them being Intel) have somewhat of a chicken-and-egg problem in terms of supporting WiMAX, particularly in the mobile and portable/nomadic form, in that they need networks to drive interest in the chipsets they plan to include in their notebooks, PDAs and potentially mobile phones. Conversely, without these end-user devices available and being competitively priced, operators are reluctant to deploy network infrastructure.
Intel the main driving force behind WiMAX has staked the future of its telecommunications effort on the standard. To that end, they have headed standards bodies, built chipsets, and subsidised a lot of early network development. Intel has invested into dozens of start-up companies including Navini (equipment – $17.5M), Clearwire (WISP USD$600M), local Australian WISP Unwired (UNW) around AUD $50M, to name but a few. By investing heavily in start-ups, Intel is hoping to ensure that WiMAX networks will be deployed. In the process it is putting mobile network operators – who have not been great supporters of the technology – on notice that Intel (and WiMAX for that matter) is not their partner.
Despite almost $1 billion of investments there are still very few, fully-compliant, commercial WiMAX networks in the world today. Almost all of these networks are based on WiMAX (802.16-2004), supporting mostly fixed applications and/or portability at best. Despite what may be read in the press there are currently no wide-scale deployments of WiMAX, only planned or trial deployments.
US carrier Sprint Nextel has placed a $3 billion bet on the nascent mobile WiMAX (802.16e) standard and plans to build a nationwide WiMAX -based BWA (Broadband Wireless Access) network to provide broadband services to customers across the United States.
However, the lack of compliant WiMAX 802.16e equipment and end-user devices (not scheduled till 2009), will result in vast areas of the globe being covered by cellular networks (GSM/EDGE, W-CDMA, EVDO, 3G/HSPA, etc) that can efficiently provide “real world” data speeds, of many hundreds of kilobits (comparable to that of DSL), with reliable mobile wireless access.
End-user devices will be a key driver
The availability, form, function and price of end-user devices will be a key driver and influencing factor in the success, or failure, of the competing technologies and future telecom services. New, advanced broadband multimedia services require additional functionalities to be included into end-user devices. Service usage will depend on the capabilities of these end-user devices. It is here, we believe, the tipping point for the competing mobile wireless technologies lies.
From the end-user point-of-view, comparison between WiMAX and 3G/HSPA does not reveal any major differences between the two technologies: both are aimed at similar uses. The WiMAX fraternity believes that the initial target for both technologies is to provide broadband connectivity to the road warrior. Supposedly it is he/she who needs to gain access to the corporate intranet and internet, for which their WiMAX technology is better suited. WiMAX vendors also believe they can, at a later stage, and through scale, extend the market towards consumers by embedding the technology in PDAs and mobile phones.
Although we concur that the market consists of both business users and consumers willing to have broadband connectivity to their notebooks and mobile phones, we argue that it is not sufficient to match only the end-user needs with specific technical characteristics of the standards; the technology has to find acceptance on all levels of the (CEVEN) system.
The battle for the bandwidth hungry road warrior
While ostensibly a philosophical argument, we believe the world is going increasingly mobile-centric. Although we might concur that the market consists of both business users and consumers willing to have broadband connectivity to their notebooks and mobile phones, we differ on the nature of its use.
We feel the market for the notebook carrying road warrior, requiring multi-megabits per second while on the move, has been overstated. We cite commercial networks such as IP Wireless, iBurst and EVDO, which support mobile broadband connectivity to notebooks today, have yet to get user “traction” for these services. Something is missing!
This leads us to deduce that the market need for mobile broadband access is not just simply replicating the PC centric use of broadband access by extending it into the wireless domain. The need is providing broadband connectivity to users while on the move, but enabling access to new multimedia content and services in a highly personalised and controlled way, combining the best of both (Fixed and Mobile) worlds. Some define it as FMC; we like to define this as the concept of personal broadband mobility.
The WiMAX -enabled, notebook-carrying road warrior (or nomadic user) vision, while needed, is symptomatic of a PC centric view of how content and services are currently accessed. We believe that HSPA will act as the catalyst that extends personal communications far beyond the formerly separate worlds of voice, data, mobility and the Internet. That is: towards the mobile information society, if you will, that moves & changes the way we live, work and play.
From this viewpoint, we cite the great strength of cellular mobile communications networks; in that the context that you are in (location) as known by the network; and so that changing information about an individual’s location, environment, and social situation can be used to deliver new context-aware communications services and applications.
Smartphones subsuming a lot of the notebook function
We are not advocating the ditching of notebook PC’s in favour of smartphones just yet. Both are needed and are complementary in the sense that each is preferred by users in different contexts and for different applications.
From the end-users perspective, whether a technology is complementary or substitutive depends on how well it provides function, additional value and possibly enhances the user experience. If we assume the initial market opportunity is broadband wireless connectivity to notebooks, then this raises the question of what air-interfaces should be accommodated on the device.
Many of today’s notebooks already have Infrared, USB, Ethernet, Bluetooth and now WiFi. There is a limited space for all interface types to be accommodated. Add to this, the complicated requirement of adding WWAN air-interfaces, it begs the question: Which ones will be selected?
Since both mobile WiMAX and HSPA are substitutes, then we believe that the vendors of these devices will take into account the end-user demand, market size and network availability. Further, that as there are presently no WiMAX 802.16e devices available today, and chipsets won’t be available for some time, this will lead Notebook and PDA vendors to favour 3GPP-based WWAN air-interfaces.
Already, smartphones (equipped with email/messaging) of the BlackBerry type and their ilk, have addressed the need of the road warrior, and other mobile users, for retrieving and sending email while out of the office environment. Because of their expanded storage capacities, one can usually store documents, contact information and other applications on them. You can also synchronise them with your office intranet or desktop PC environment, as well as browse the Web.
We believe that the convergence of cellular mobile telephony, palm-sized computers, location-based information and other sensor data into a mobile handset (packaged into a form factor that is now more aesthetically pleasing, sharper, more colourful, and with a lot more capabilities than its predecessors) will threaten the utility of notebook PCs and standalone PDAs.
Of the estimated 1 billion mobile phones sold globally per year, 25% will be smartphones. Compare this to about 100 – 120 million desktop PC’s, and 50-70 million laptop/notebook computers. More and more of those hundreds of millions of phones are becoming ‘smart’; tracking or replacing the capabilities traditionally found in PCs.
We see HSPA driving the adoption of a new breed of personal smartphones; embedded with even more features such as multi-megapixel image/video cameras/recorders, music players, email/messaging, gaming, WiFi, SIP, etc which, in turn, will be far more enticing to end-users. This will be especially true when coupled with the ability to seamlessly traverse across a vast array of heterogeneous global networks, thereby increasing the choice of personalised converged services.
Everything needs to be in the right context
One of the main problems with the PC centric view is that it presupposes that the manner by which we currently access the internet (that is, via a PC), will be the manner of the future. In our introduction we stated that the world is evolving from one in which almost all access to the Internet comes via the PC; to one in which small ‘smart’ mobile computing/communications devices are expected to make up a growing share of the end-user equipment.
Although we are not foolish enough to suggest which of the myriad of applications will succeed, we will suggest that they will have attributes of ease of use, rich experience; addictive, impulsive, task oriented and be instantly gratifying.
3G/HSPA smartphones, in partnership with an architecture known as IMS, will open up a broad new menu of personal mobile communications. IMS will facilitate context-based services to work in concert with emerging multimedia technologies, to deliver information in a more compelling manner. For instance, the statement “I’ll meet you for lunch at that seafood restaurant opposite the Opera House” defines context with surrounding location to convey useful information. Through IMS, based on the users’ location, they can: learn what restaurants are around them; get a listing of the restaurants in the area; watch a video streamed trailer of the inside of a restaurant and make a reservation either by phone, email or the web.
While much of the attention seems to focus on the consumer sector, new enterprise-targeted applications -traditionally, the foundation of the Context or Location Based Services market – are in progress as well. For instance; a different set of services related to mobile resource management are being developed. Tracking systems for police, ambulances and taxis have been around for years, but may be expanded to people like sales force members.
Although IMS is, in itself, technology agnostic, it applies equally to all access technologies, including WiMAX. The extensive reach and resources of 3GPP based networks, with their existing subscriber management, session control and signaling systems, will enable the tracking of all subscriber information and movements and locations, and places its technology at a distinct advantage over, say, mobile WiMAX.
Voice still an essential means of communication
What has been missing in the discussion of broadband wireless technologies is the role of voice in the broadband world. While “broadband over whatever” is a one of the fastest growing market segments in the ICT industry, conventional person-to-person voice is still an essential means of human communications, and is still a growing mega-billion dollar industry.
The WiMAX proponents argue that, with the advent of PC-based VoIP (Voice over Internet Protocol) programs such as Skype, and their ilk, road warriors can now make calls from their notebooks to anywhere in the world. So, it seems that any lack of intrinsic voice support in the technology is not an issue. Voice, they argue, is just an (IP) application on the broadband network. While this may be true, the fly in this ointment is that VoIP coverage isn’t ubiquitous. The quality still leaves a lot to be desired and it is only free between users on-net; off-net users incur charges, emergency services may be non-existent, and the application still requires many of the facilities (interconnect, numbering, directory, etc), to name but a few, provided by the PSTN.
In contrast, mobile phones have become the de facto voice communicator, removing physical boundaries and empowering people to communicate and conduct their affairs free from the constraints of physical proximity and spatial immobility – you can take them with you almost anywhere.
It is estimated that 25% of all voice calls originate from mobile phones and 60% of calls within the enterprise are made by mobile phones; a source of lucrative mobile network operator income. But MNO’s have recognised the threat of losing voice minutes going over private intranets and the public internet (as VoIP) rather than their cellular networks, and have accelerated their foray into the fixed indoor world and IP.
While 2G/2.5G bandwidths were too narrow to ever support VoIP upgraded, HSPA-enabled 3G networks will have the capacity and latency to support VoIP. MNO’s have already committed to deploying it, but not as rapidly as to introduce services which may commoditise voice.
For that reason, we believe that VoIP will make its way into the 3G/UMTS RAN (Radio Access network) in some fashion in the not-too-distant future. Either its implementation will come through the IMS architecture discussed earlier, or through the device convergence effort (such as UMA), leading to interoperability between Mobile and IP (internet protocol) networks.
A word about global roaming
The ability to roam, or use one’s mobile seamlessly across network borders, is almost the raison d’être of the GSM system from which 3G/UMTS has evolved. Roaming users account for 25% of the overall 2.5B global subscriber base for voice, with global roaming revenues accounting for over EUR80 billion.
Voice has been the main focus of functionality (in terms of roaming services), and contributes the largest proportion of roaming traffic and revenues. To date, data usage remains within SMS or messaging-based services, with limited full data access. However, we believe this will change, as new revenue-generating services are developed to increase roaming usage; including context-based destination content, presence, MMS postcards and call assistance.
The ability to provide roaming support across different administrative domains is a key success factor to the contending broadband wireless technologies. But roaming is not just a matter of establishing commercial agreements between operators – there are many technical and commercial challenges to overcome!
HSPA (as part of the 3GPP community) has by far the greater advantage over Mobile WiMAX. This is not only due to the numerous regional and global alliances, but also its experience and extensive palette of Authentication, Authorisation, and Accounting (AAA) principles already embedded in the standards that it can bring to bear. For instance, the SIM (just one) function of network authentication to protect against fraud has been of undoubted significance to the GSM mobile industry. Its equivalent in 3G the USIM brings an extension in performance, security coupled with capabilities for multiple transactional applications.
Which road will lead us to the 4th Generation?
Both camps have developed roadmaps to improve and enhance the data carrying and spectral efficiency of their respective air-interfaces to deliver 100s of megabits per second.
There are some who believe the current W-CDMA technology standard has fundamental capacity limitations for high user loads. Accordingly, they see the next generation wireless standard evolving into an entirely new modulation scheme based on OFDM (Orthogonal Frequency Division Multiplexing), As WiMAX, is already based on OFDM, it is focusing its efforts on tweaking it mobility management capabilities to deliver ever-increasing data rates to mobile users.
While 3G/W-CDMA essentially comes to a standards pause after HSDPA/HSUPA, leaving mobile operators’ anxiously wondering what will get their networks up to, say, 100 Mbps. Enter what is now dubbed “Super 3G”, which is more formerly given the rather long-winded name of UMTS Terrestrial Radio Access Node Long Term Evolution (UTRAN LTE), and sometimes called 3.99G, or Evolved UMTS. The Third Generation Partnership Project (3GPP) has started to look at the required specifications for Super 3G /LTE (Long Term Evolution) that is ambitiously intended to come up with new standards by mid-2007.
Super 3G plans to reduce the cost-per-bit even further, while undergoing a vast spectral-efficiency improvement to the air-interface. By increasing the bandwidth from 5MHz up to 10MHz, Super 3G plans to offer throughput of 30Mbps (downlink) in the wide area and more than 100Mbps (downlink) in the local area (10 times the current speed of 3G).
Although Super 3G does not specify any particular technologies (a robust form of OFDM has been nominated), but instead indicates a need for identifying methods for greater bandwidth that: maximize the use of the radio spectrum; and that offer increased flexibility for the delivery of future services.
One goal of Super3G is to enable an easy migration into all spectrum (including 900 MHz) and offer seamless mobility and hand-off between existing WCDMA/HSDPA networks.
But 4G is not a single standardised air interface and network infrastructure
Going forward, the progress towards 4G is not only about achieving higher data rates, lower latency and increased user capacity, as many would have us believe. Rather, it is also about the incorporation of the user to a much higher degree (almost as the centre point) than previous generations of communication networks.
In this sense, user centricity means applications and services will be developed with the end-user as a person and not some anonymous entity that will have to use whatever the technology is capable of offering. Convergence is a word that is bandied around a lot; but convergence at both the end-user device and at the network (fixed and mobile) is a necessary part for achieving 4G.
The user’s need for a display device, which may be TV, notebook screen or smartphone, will drive device convergence….but not as it is today, where one can show pictures or videos taken by a Smartphone on the TV. The vision is that any content can be shown, at any time, and from anywhere. It involves all kinds of enhancements, both software and hardware, for interacting with the network and the service providing device. The question that it raises is: Will this device evolve from a notebook/PDA, embedded with some WWAN air-interface, or from a mobile phone, with the added functionalities of a notebook and a landscape TV screen?
From a user’s point of view, a common network fabric is beneficial; since this basically enables communication between various devices, over short as well as long distances; through whatever communication means is available. This is where a common platform, such as IP, comes in to play; making software development much easier, not only for new network and application components, but also for services.
In the end, if you believe in the paradigm that adaptively supports traffic coming from either fixed or mobile terminals; and that the services are provided independently of the underlying network infrastructure, but depending on the context (location, terminal capabilities, personal preferences, etc.) interacting with services and applications at home, in the office and around us, then you concur with the evolutionary view of 4G. One that will be the horizontal integration of heterogeneous infrastructure comprising different wireless access systems on a common packet based platform to compliment each other in an optimum way for different service requirements. This is commonly referred to as always best connected (ABC). As such we can see a place for many broadband wireless access technologies such as WiMAX and others fulfilling unique roles in the evolving wireless technologies landscape.
But if it comes down to a battle between WiMAX versus 3G et al our analysis suggests that the value proposition of 3G/HSPA is more compelling and more capable of providing a full personal broadband mobile service to the market, consisting of networks, terminals, and applications and content. Despite the emergence of alternatives such as WiMAX it is expected most mobile operators will continue down the 3G standards path to HSPA, as the business case for alternative wireless technologies is not yet compelling enough to switch.
(See also Part one)
About Investech
Investech Technology Consultants is an independent, dedicated consultancy catering specifically to the unique requirements of the international financial and investment community. Our clients include Institutional Funds Managers, Hedge Funds, Venture Capitalists, Investment Banks and Brokers.
With over 40 years of combined experience as working practioners in Telecommunications, Information Technology our consultants use our background, knowledge and experience to make technology understandable. Moreover, we provide the relevant detailed, timely assistance necessary to help the financial community make better informed investment decisions.
To this end, we offer a range of Masterclasses and Seminars to the public for a limited time.
To learn more about this topic, the details of our Masterclasses are:
Series 1 Masterclasses – Next-Generation Wireless
This series of Masterclasses gives a comprehensive overview of the Broadband Wireless Phenomenon.
Time / Date Location
6th & 7th June 2007
9:00AM-5:00pmAdelaide – Stamford Grand
14th & 15th June 2007
9:00AM-5:00pmBrisbane -Marriott Hotel
20th & 21st June 2007
9:00AM-5:00pmMelbourne Rendezvous Hotel
26th & 27th June 2007
9:00AM-5:00pmSydney Marriott Hotel
3rd & 4th July 2007
9:00AM-5:00pmPerth Medina Executive Barrack Plaza
11th & 12th July 2007
9:00AM-5:00pmAuckland Stamford Plaza
To book your place at one of these Masterclasses go here: www.informa.com.au/nextgenwireless