Wireless Hotspots

Back to Contents of Issue: April 2002


Speedy broadband Internet access is winning subscribers hand over fist in Japan, mostly through ADSL and fiber services that are hard-wired to your house. Thanks to innovative experiments in the wireless broadband market, that's all about to change radically, presenting the consumer with an even better range of options for that fast online fix.

by Michal Thuresson
Major fixed-line telecommunication carriers and ISPs are attempting to make wireless 'hot spots' the next big Internet thing in Japan. Hot spots -- high speed, short-distance wireless access points centered in highly populated areas -- are conceptually well suited to crowded Tokyo. For fixed-line Internet service operators, they are also relatively cheap to deploy and have the potential to be lucrative add-on services that steal the spotlight back from mobile operators and their much-hyped 3G offerings.

THE IDEA OF HOT spots is a service provider provides a wired, high-speed connection to a popular area such as a hotel or airport, and wireless LAN technology extends the reach to users in the area. In theory, it's a much cheaper and faster way to realize wireless multimedia applications than cellular networks. This theory has been tested extensively in public experiments in Tokyo over the past year at sites such as fast-food restaurants, coffee shops, hotels, convenience stores, offices and retail stores. While these experiments have been able to create minor public interest, a sound business plan has been more elusive. "We have not yet decided on the business model for providing a public hot spot service,"says Tadao Kobayashi, senior manager of Corporate Strategy Planning at NTT East, which conducted an extensive public hot spot experiment called 'Biportable' in Shibuya last spring.

In the background, the Ministry of Public Management, Home Affairs, Posts and Telecommunications has been deliberating how to create a suitable radio spectrum environment for high speed wireless LAN to develop. In an October 2001 press release, the ministry states: "With the increased use of the Internet in a wide variety of social fields and the demand for broadband access, there has been a huge increase in user needs for wireless access systems." The ministry sees latent demand for broadband services and believes that alternative solutions using wireless technology are needed. "In Japan, there are many apartments that cannot use DSL solutions because they are already connected to fiber or the apartment is too far from the local NTT switching office,"says Masato Muto, deputy manager of Corporate Planning for SpeedNet, a wireless solution provider focusing on apartments in the outer Tokyo areas. "DSL providers have a very, very big problem: It's impossible to service places that are connected to fiber. NTT started the process of installing fiber 10 years ago and its coverage is growing."

Since Japanese law restricts the commercial use of outdoor high-speed wireless access to the 2.4-2.497-GHz radio spectrum, most ISPs have been forced to use this radio band for their hot spot experiments. This is a severe limitation because this is an unlicensed spectrum with significant interference. This bandwidth is noisy because of interference from other home equipment. For example, microwave ovens and cellular phones using Bluetooth technology can disturb the transmission.

Therefore, the ministry decided in late 2001 to open up certain parts of the 5-GHz radio spectrum for outdoor use from March 2002. "As the need for an outdoor-based 5-GHz wireless access system remains as strong as ever, in the future, this is likely to be divided among different mobile industries," says the October ministry report. This is an extremely important development for hot spots, as both the 4.9-5.0-GHz and 5.03-5.09-GHz bands are open for outdoor use from March. Most interested parties expect not only the 4.9-5.0-GHz band to be open for outdoor use, but also the 5.03-5.09 band. This happening just as hot spots are gaining momentum has created new optimism. Scaling wireless access service to the higher data speeds of 5GHz makes a long-term business strategy more attainable, though doing this isn't without its challenges. According to NTT East's Kobayashi, even the 4.9-5.0-GHz band is noisy. "Many fixed microwave systems occupy that band. I think it's impossible to use outdoors in the Tokyo or Osaka areas. This is why the ministry added the 5.03-5.09-GHz band in early February. Nobody uses this band presently, so it is very pure," he says. It is also very fast, with data throughput up to 36Mbps. Transitioning users from 2.4GHz to the next generation 5-GHz solution is the kind of scalability service providers need. "We intend to use the 5-GHz bands to scale up our service for faster speeds. We realize the unlicensed 2.4GHz band has many problems because we have to share it with so many products," says SpeedNet's Muto.

SpeedNet, like many other ISPs, uses its 2.4-GHz wireless solution to connect the user to a high-speed backbone. The venture, started in September 1999, is majority-owned by Tokyo Electric Power Co. (Tepco), with major investments by Microsoft and Softbank. The solution uses Tepco's extensive fiber network and provides a wireless link for apartment building tenants. The company is notable not only for its big-name investors, but also for being the first major carrier affiliate to launch a 2.4-GHz wireless LAN solution commercially.

SpeedNet uses an 802.11b (an international 2.4-GHz protocol) antenna and installs it atop any of Tokyo's ubiquitous Tepco-owned power poles. This is a crucial advantage as one of the few things matching the level of NTT's control of local phone switches (which it uses to dominate the ADSL market) is Tepco's control of Tokyo-area utility poles. Tepco also owns over 50,000 kilometers of fiber optic pipe in the Tokyo area, and each utility pole base station connects the apartment user to this fiber backbone. The antenna range is about 300 to 500 meters from the utility pole, enough to cover a significant number of buildings in crowded Tokyo, and the maximum data throughput is 1.5Mbps, which is competitive with current ADSL speeds. The current monthly price of JPY2,450 per month is also competitive with ADSL. "Our best advantage is the ease of installation, which usually takes about four days. Other broadband services require negotiation with the apartment landlord and often take several weeks, but our solution is contracted with the individual tenant," says Muto.

SpeedNet began service in Saitama City in May 2001 and is carefully going after areas it sees as easy targets. According to Muto, early research showed that around 14 percent of the apartment buildings in Saitama City could not be fitted with ADSL. From there, coverage expanded to the entire Saitama prefecture in September, then to over a dozen suburban Tokyo wards, parts of Kawasaki City, Kanagawa Prefecture and Yokohama by December, and then adding Urayasu and Ichikawa cities and sections of Chiba prefecture in February 2002. As of early February the company had just over 3,000 subscribers but it expects this number to grow rapidly as the message gets out. Ads promoting the service have been running on the packed JR Yamanote train line since December; the company plans to expand into more Tokyo wards during 2002.

Accessing the central Tokyo area, though, is a potential source of problems, as it could expose the weaknesses of the 2.4-GHz spectrum. SpeedNet claims the interference with other devices has not been a problem, but until now it has deliberately avoided the densely populated regions of Tokyo. Muto admits that their solution has experienced significant problems with structural interference. Only 60 percent of inspected apartments as of February could receive the wireless signal from the utility pole antenna. The antenna needs to have an unobstructed line of sight to the apartment complex, but the closer to central Tokyo you get, the more obstructed the skyline gets. Other line-of-sight wireless solution providers are dubious on this area's potential as well. "The Tokyo skyline is very crowded and is a difficult place to have a line-of-sight solution," says NTT Communication's Tetsuro Okamoto, assistant manager at NTT Communications' Corporate Planning department, a section offering enterprises a fixed wireless solution that competes with fiber services.

The development of a 5-GHz product, which does not require a line-of-sight connection, is therefore a huge factor in SpeedNet continuing its expansion into more populated areas. Even though parts of this spectrum have been cleared by the Japanese government, antenna manufacturers have yet to commit large-scale to making products for this band, making the hardware expensive, according to SpeedNet's Muto. "There are a number of companies considering a 5-GHz service, however none is currently available. 802.11a products only began shipping at the end of 2001," says Pete Fowler, vice president of marketing, sales and operations at Magis Networks, a US wireless LAN hardware maker.

Despite the concerns over interference, the lure of capturing a huge audience within a small service area has led to several investments in hot spot services in the Tokyo area. Long-distance carrier Japan Telecom and East Japan Railway tested a 2.4-GHz wireless LAN using 802.11b at Tokyo Station from last September through December. Through ODN, Japan Telecom's ISP, the companies collected 300 trial applicants and distributed laptops and PDAs with wireless LAN cards for use in the station. "We never heard of any significant interference problems from our users," says Atsushi Kobayashi, in charge of marketing in the Network Products division of Melco, a wireless LAN hardware maker that supplied its wireless LAN hub 'Airstation' for the Tokyo station trial. Japan Telecom is now set to run the trial in other major train stations in Tokyo, but it has no definite plans to commercialize the service yet.

In December 2000, Japan Telecom set up a planning and research company called Air-Catch together with NCS, an affiliate of Ryuseki, a major oil company in Okinawa. The company's main purpose is to develop a wireless LAN product that connects to Japan Telecom's fiber-optic network. "The two businesses will not compete, but rather complement each other," says Melco's Kobayashi.

Another high-speed fixed-line operator, Usen Broad Network, funded a wireless LAN carrier called Mobile Internet Service. For the last half of 2001, Mobile Internet Service set up various 802.11b access points in Tokyo, offered wireless LAN cards for laptops and PDAs, and experimented with the system. Though Mobile Internet Service has connected these access points to Usen's fiber network, it also had not announced any plans to commercialize the service as of this writing.

Most likely, the first major fixed-line provider to commercialize hot spots will be NTT Communications. The long-distance carrier arm of NTT currently has been running a trial hot spot service called 'HI-FIBE' since last July, and it plans to commercialize it sometime this spring. The service uses Melco's 802.11b Airstation in various Tokyo locations such as Mos Burger fast-food shops, Mini-Stop convenience stores, Blenz Coffee shops and the Shinagawa Prince Hotel. Customers at these locations can boot up their laptops and register at the Web site of NTT Communications' ISP OCN; they then connect to the LAN via a wireless LAN card.

Neither NTT Communications nor Mos Burger would comment on the results of the trial, but in a February 2002 Reuters article, an NTT Communications representative expected to get "at least 5,000 subscribers" at the HI-FIBE commercial launch. Some significant questions remain about how this can be a profitable and useful service. "The stores want to be the exclusive 'food shop' or 'hotel' partner for the service, and likewise, OCN and NTT Communications want exclusivity with all the shops they set up," say industry sources.

Security is also a major concern for the HI-FIBE service. "Services utilizing unlicensed spectrum through 802.11b or other technologies have great promise, provided security is managed carefully," says Ken Zita, president of Network Dynamics Associates, a global telecom management consultancy based in New York. "Networks that rely on the Wired Equivalent Privacy protocol (WEP), the encryption and access control aspect of the 802.11 standard, are at risk as WEP is easily hacked."

NTT East, in the end, could be the first company to conquer all these challenges and deploy a profitable hot spot service. To get an initial customer base, the company can cross-market hot spots with its market-leading FLET's ADSL service. NTT East also has extensive R&D dedicated to developing wireless access devices, the ear of many content providers looking to deploy high-speed wireless applications, not to mention the ability to outmuscle anyone standing in its way.

The Biportable hot spot experiment, featuring data speeds of 36Mbps, was conducted in Shibuya from March to July last year and included over 30 content companies. The involvement of content companies is a differentiating aspect of the Biportable hot spot. The lure of cool content is an important key to getting hot spots in the black. In a March 2001 news release announcing the experiment, NTT East said it was able to launch Biportable because the planned content was gathered smoothly and that this would yield appropriate data to use for commercialization. The content providers included: Yomiuri Shimbun (news content), Tower Records (live event video), Digital Hollywood (educational content), Gaga Communications (movie content) and Recruit (entertainment content). Contents and applications that have been tested include: television conferencing, music videos, a face-to-face help desk, video on demand, short movies and town guides.

NTT East provided its Advanced Wireless Access (AWA) technology and fiberoptic IP network, participating companies offered their content, and then NTT handled the integration. "We don't know much about wireless content formats and things like that. NTT East handled it -- they brought cables, cameras, mixers and everything else that was needed," says Tatsuro Yagawa, director of the Business Development Group at Tower Records. "We just dealt with copyright and license matters, and we arranged to have the pop singer Sakura do a live show for Biportable debuting her new CD."

This chance to trial streaming video contents did not cost Tower Records anything, and more importantly, it gave it a chance to determine the value of offering wireless content to in-store customers. The company currently does not offer content for any of Japan's three wireless carriers, but has received inquiries from NTT DoCoMo about becoming an official content provider. Not many companies are in the enviable position of being asked by DoCoMo to provide content for i-mode, and even fewer would consider developing hot spot content first. One big reason for this could be that Tower Records' eight-story Shibuya mega-store has the distinction of being the world's busiest record store -- a clear indication of why Tokyo hot spots have unrivaled potential.

Biportable was limited in scope to indoor venues because at that time there was no spectrum in the 5-GHz band that could be used outdoors. Harnessing the potential of outdoor use will lead to many more public-oriented applications than the experiment tested. Being able to handle the massive user demand on an outdoor hot spot network is one reason NTT East uses AWA technology, a European standard with different functionality than 802.11. According to Kobayashi, IEEE 802.11 is not suitable for public use because there's a mass collision of data and interference problems. Also, he explains, AWA was originally developed to connect to a fiber backbone and is superior to IEEE in the areas of security and performance.

AWA for indoor use occupies the 5.15-5.25-GHz band, and Kobayashi's job is to develop AWA devices that can be commercialized for both indoor and outdoor use. It's no wonder then that the opening up of the 5.03-5.09-GHz band for outdoor use has him excited. "We will probably be able to use this band from this autumn or winter, and we intend to launch an AWA card for PDAs by then," says Kobayashi. AWA's 36-Mbps transmission speed, more than a thousand times faster than current PHS wireless service for PDAs, would be an incredible leap forward in PDA functionality.

The results of the Biportable experiment, which featured PDA-like terminals co-developed by NTT East and Matsushita, provide a glimpse into the PDA's future. According to NTT East's report, which is based on user surveys, "Image quality achieved the expected level, but the screen size for motion pictures has to be improved." The content providers concluded that the images were good enough to be sold and that content which is targeted at a narrow range of customers, and which is highly specialized, will be accepted even if the price is higher than general content. Perhaps the most encouraging statistic was that only 3 percent of users thought regular Internet-based video was of a higher quality than Biportable's, with 62 percent feeling Biportable was better. Another user survey came up with a rough idea of what consumers would pay for hot spot content: JPY773 for movies, JPY281 for music videos and JPY228 for music downloads.

Of course, things would have to run differently for the commercial version of Biportable. "NTT East can only provide the backbone and extended wireless connection. An ISP must provide the service and contract with the content companies," says Isaharu Kikushima of NTT East's Corporate Strategy Planning department. Daiji Hirata, chief research manager at Neoteny, an IT business incubator that organized Biportable, says, "The challenge is creating a profit-sharing model like the larger cellular market in Japan and getting the cost down for the consumer."

NTT East plans to eventually tie hot spots to cellular networks, which would expand the user base exponentially and allow for seamless roaming across different networks using one handheld device. The AWA technology used by NTT East was designed to be compatible with both 3G cellular and fixed-line networks. Also, a prototype device capable of accessing both PHS and wireless LAN networks was developed in late 2001 by NTT Network Innovation Laboratories, a separate NTT entity. The technology, called software-defined radio (SDR), uses onboard device software to download communication specifications and would allow a single terminal to use multiple networks. "Technically speaking, we believe it is possible to develop SDR devices that are affordable and small enough to be portable. Commercial success will depend on the SDR vendor's strategy," says Masahiro Umehira, director of engineering for Wireless Systems Innovation at NTT Network Innovation Laboratories. "These devices will be like magic boxes we can use as cellphones, wireless LAN terminals, radios, TVs, PDAs and more."

NTT Network Innovation Laboratories is in close communication with NTT East on developing and deploying the SDR devices. In fact, Umehira was on Kobayashi's staff two years ago and began developing SDR there. "Dr. Umehira's system is ideal for a wireless service but is at least a couple of years away from being commercialized," says Kobayashi. "But the ability to use handsets seamlessly across different radio bands will be mainstream in five years." @

Note: The function "email this page" is currently not supported for this page.