A Band Apart?
By Maurie Dobbin
While cdmaOne has long been viewed as the means of relieving 800 MHz operators of overcrowding, it is the new cdmaOne networks at 1.7, and especially at 1.9 GHz, that are likely to shoulder the burden in the Asian region.
A total of 13 Asian countries have awarded commercial or trial CDMA contracts, demonstrating the enormous interest of Asian operators in the benefits of the standard. With China, Japan, South Korea, Cambodia, the Philippines, Myanmar, Thailand, Indonesia, Singapore, India, Hong Kong, Vietnam and Bangladesh all offering or trialing some sort of CDMA service, Asia has one third of all CDMA operators.
The trend of late in the region seems to be towards cdmaOne networks at 1.7 and 1.9 GHz. Already the three new PCS operators in the Korean market have shown startling growth with KT Freetel recording more than 750,000 customers within six months of commencing operation. Rivals Hansol PCS and LG Telecom are not far behind with PCS customers totaling 2.86 million out of the 8.5 million cellular customers in Korea as at the end of April 1998. By 2002 the PCS carriers expect to be supporting almost nine million customers of the 19 million projected for Korea.
The Korean PCS networks operate at the non-standard band of 1.7 GHz, but multi-band handsets expected on the market shortly will open up roaming to international cdmaOne destinations. Already the operators have made arrangements for Korean PCS handsets to be made available on a rental basis in Singapore, Hong Kong and Japan to make it easier for foreign visitors.
Infrastructure sharing has been a key aspect of the development of the Korean PCS market with KT Freetel and Hansol PCS agreeing to cover certain geographical areas and then allow roaming between the carriers for their customers. By agreeing to share the market in this fashion the Korean PCS operators have been able to quickly offer national roaming, thus offsetting the disadvantage of their small cell infrastructure.
Singapores MobileOne (M1) is also looking to PCS cdmaOne to open up new markets. M1 was awarded a license in early 1995 to operate CDMA1900, GSM900 and paging services in Singapore. The M1 GSM900 network commenced operation a little over one year ago and already they have been successful in clawing a major share of the market off SingTel. During April M1 received an unexpected boost when, in an embarrassing fiasco, the government was forced to withdraw a mobile phone license from the P2P consortium due to its failure to secure GTE as a partner. The government has now declared that it will not re-tender the license, leaving the Singapore market to SingTel, M1 and one new competitor, the StarHub consortium. M1 is looking to the two-year period of grace before StarHub commences operation to position its new CDMA network in the market.
The M1 CDMA service was launched on April 1 by Mr. Leong Keng Thai, director general of the Telecommunications Authority of Singapore (TAS). "The introduction of CDMA in Singapore will offer greater consumer choice, wider range of services, more competitive prices and enhanced quality of service standards to consumers," said Mr. Leong.
Analysts Merrill Lynch have stated that they expect that seven per cent of M1s estimated 290,000 customers in 1998 will opt to join the new CDMA service lured by cheaper tariffs and subsidized handsets.
The Singapore mobile market has been stimulated by competition over the past year with subscriber numbers exceeding 800,000 in March, sending penetration climbing towards 30 per cent of the population. But it is not just the numbers of customers that are stressing the networks; it seems that Singaporeans talk and talk and talk. Every month M1 estimates that their customers spend an average of six hours on their mobile phone one of the highest rates in the world. And the time they spend on the phone is increasing; M1 estimates a 17 per cent increase since a year ago.
Some of the other performance objectives that M1 had set included: 95 per cent of call attempts should be successful during the busiest hour of the day; on-street and in-building coverage should be greater than 95 per cent and 85 per cent respectively; and the average call set-up time should be less than five seconds. OKane claimed that M1s initial tests of the CDMA network had shown that the call set-up time was much less than this objective.
One of the challenges facing M1 was meeting the in-building performance objectives which presented some unique problems to their implementation team including:
"As the M1 network expands we will have a new set of challenges," said OKane. "These can be classified into capacity expansion and new feature introduction, As usage increases and customer numbers ramp up there will be a need to increase carriers on individual sites and even, on occasions, to add new sites. Generally, our sectorized sites are running approximately six erlangs/sector on 13 kbps vocoders. There are currently sectors on our GSM network far exceeding this figure".
But the largest challenge lies in China where PCS at
1.9 GHz is regarded as a logical and cost-effective migration path for
operators who expect to have exhausted their currently allocated frequencies
within the millennium. The China cellular market has been growing at 100
per cent per year and by 2002 the cellular subscribers have been predicted
to exceed 70 million.
Ho stated that the State Radio Regulatory Council has assigned a pair of 15 MHz frequency assignments in the band 1865-1880 MHz and 1945-1960 MHz for PCS services. In addition another consecutive pair of frequency assignments has been designated for wireless access applications (1880-1900 MHz and 1960-1980 MHz). Both of these bands are within the assignments for PCS in the US market, enabling standard equipment from that market to be used in China.
But the application for PCS in China is very different
from that in the US. In China, mobile handsets are being used as a substitute
for inadequate telephony services, resulting in very high concentrations
of users with low mobility. This means that networks operating at roughly
50 per cent capacity overall are stressed in hot spots where
capacity is stretched to the limits.
"PCS sites can be deployed as an underlay to the cellular network for hot spot coverage to provide these high traffic concentration areas with additional capacity," said Ho. "Dual band terminals operating at both 800 MHz and 1900 MHz can be used to hand-off between the PCS and the cellular network".
Ho said that based on the frequency allocations at 800 MHz he believed there were two potential CDMA PCS operators in China: China Unicom and Great Wall. Both of these operators would first use CDMA within their existing 800 MHz allocations and then move upwards as capacity became saturated. Great Wall Mobile Communications and China Unicom are deploying 800 MHz CDMA systems in Beijing, Shanghai, Guangzhou, Xian and Tianjin; these networks are expected to be extended to another nine provinces in 1998. However, China Unicom only has four MHz of frequency spectrum, enough for just two CDMA carriers. On this basis Ho thought that China Unicom would need to move up to PCS frequencies by 2001. Great Wall, on the other hand, has 10 MHz of 800 MHz spectrum, which allows a maximum of seven CDMA carriers, making the adoption of PCS as a solution more likely around 2003.
China Telecom, the third and largest operator in China, has a different migration strategy which is likely to involve the use of CDMA at 900 MHz in a similar fashion to that adopted by the Japanese cellular operators DDI and IDO. Introducing 900 MHz CDMA would make the most efficient use of China Telecoms existing TACS infrastructure through the use of dual mode handsets during the migration period.
Ho explained that the TACS networks in China are primarily supplied by Motorola and Ericsson and are interconnected via IS-41, which provides subscribers with roaming capability in either network. The TACS networks cover the entire country with two TACS networks being deployed in each of the nine provinces and autonomous cities.
China Telecoms migration plan involves moving into ETACS spectrum and combining the dual networks in these provinces into one large network. CDMA could then be introduced to increase network capacity in the urban areas with dual mode units used to enable wide area roaming. Ho added his belief that, over time, the natural migration of the urban 900 MHz network would be towards PCS.
Ho admitted that the use of PCS would involve a larger investment, stating that a simplified RF estimate indicated that a 1.9 GHz PCS network would require twice the number of cell sites to cover the same area as an 800 MHz network. There would also be some noticeable difference in signal quality, he added.
"PCS may be superior in urban areas because its signal reflects better than signals at 800 MHz and thus would be able to fill in more coverage gaps", said Ho. "However, in-building coverage for PCS may be worse than 800 MHz and more micro sites need to be deployed to improve the Grade of Service".
Testing indoor coverage was a specific aspect of the 800 MHz CDMA trial system installed by Motorola in Beijing for Great Wall. One quarter of the Beijing downtown area was involved, with a 300 meter grid pattern established and then a building selected close to the center of the cell. Ten locations were then tested on the first floor of each building with 10 calls being made from each test point. According to the Great Wall test procedures, a test location was regarded as having indoor coverage only when origination and termination were successful and there was no drop-out. Under these conditions, the test found that the average coverage level on the first floor in the selected areas was greater than 80 per cent.
The Beijing system currently has 47 cell sites and has
a capacity of 68,000 subscribers. During the next couple of years, Great
Wall has forecast that it intends to add three more carriers increasing
cell sites to 120 and capacity to 540,000. Ninety of those cell sites
will be located in urban areas of Beijing.
Despite this focus, Mr Zong said that the base of the mobile communications industry was very weak and the country was still dependent on joint ventures involving technology transfer. He identified twelve mobile telecommunications manufacturers that were supplying handsets and infrastructure equipment, but commented that the profitability in these enterprises was very low and that they lacked any capability for research and development.
Zong called for the expansion of these joint ventures to enlarge the scale of manufacture and more effectively transfer know-how as the basis for Chinas involvement in 3G development.
"The 3G mobile telecommunications technology is a new industry and it has huge market potential", said Zong. "On the research base of second generation CDMA and the manufacturing base of second generation mobile telecommunications we should propel the preparation of 3G. We should strive to supply products to the market when 3G technologies begin to enter commercial service in five years time."
Zong wanted to gather researchers within China to focus on key techniques in the development of CDMA systems.
The rollout of PCS CDMA in Asia and the involvement of local research organizations in the development of 3G standards fits well with the CDMA Development Groups objective to create a harmonized approach to the next millennium.