 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
|
ARTICLE
By Adlane Fellah, CEO and founder
It is often said that spectrum is a “must have” asset for wireless operators who want to scale their service offering. The lower the frequency band, the better the propagation characteristics, with sub-1GHz sometimes referred to as “beach-front property”. Having beach-front property is great, but having a large property is even nicer. Regardless of the frequency band, the amount of spectrum does matter. In our Clearspectrum™ database, we have accounted for more than 1,000 spectrum owners in the 700MHz, 2.5, 3.5GHz bands. This is obviously a large number compared to the 300+ 3G operators in the market. But the question is how many have sufficient spectrum to provide profitable fixed/portable/mobile broadband services? And consequently how much spectrum is needed to build a profitable and scalable network? Spectrum characteristics Our first step is to breakdown the operators from our BWA/WiMAX spectrum database by the amount of spectrum they own/lease. Exhibit 1 below shows the results of a sample of the top 200 BWA/WiMAX licensees worldwide that own at least 20MHz of spectrum and less than 50 MHz. A big proportion owns more than 50MHz, which is surprisingly high. Exhibit 1: Operators Split by Spectrum Owned
Those are of course global numbers that do not show the regional differences and particularities. The following provides the regional particularities:
Exhibit 2: Operators Split by Spectrum Owned by Region
CALA is the region with the highest percentage of operators who own “fat” amounts of spectrum, which should help these operators deploy BWA and WiMAX services with more leverage than their Asian counterparts, such as the Indian operators who have the lowest amount of spectrum. How much spectrum is enough? Thus we have broken down the spectrum ownership by region. But how much spectrum is enough? There is probably not a single answer. Five and 10MHz are the most popular channels in the mobile WiMAX profiles, while 3.5MHz channels are characteristic of 802.16-2004 radios. According to Kevin Suitor, VP Business Development at Redline Communications, “All deployment scenarios we’ve analyzed in the past indicate that 30MHz is the minimum required to provide decent residential broadband service, assuming that we are talking about providing continuous (ubiquitous) coverage in urban areas. This is based on a reuse of 1/3/3. Obviously in low subscriber density scenarios, 3 x 5MHz may work as well from a quality of service perspective, but the business case may not look so good (depending on ARPU).” Haig Sarkissian from WiMAX 2020, a provider of business and financial planning tools for WiMAX, concurs. “The amount of spectrum needed depends highly on the amount of traffic that needs to be supported on a given cell site. This is a factor of the geographic coverage area of a cell, the number of users, the oversubscription rate, the types of applications used and their throughput requirements. The ideal amount of bandwidth needed depends on the traffic model on a regional basis.” For rural deployments with low population densities, 15MHz of spectrum would usually be sufficient. For urban and suburban deployments with high population densities, 30MHz would be necessary. Based on the traffic generated on the network, it would be desirable to have 60MHz or even 90MHz available in order to be able to expand the network as traffic demands dictate the need for additional bandwidth. Further, the minimum amount of spectrum can also depend on whether a given operator is expanding its offering to include mobile services. “Assuming that the target is to achieve the same quality of service (in terms of achievable peak data rate), SNR requirements for mobile application will be higher. For example, just the difference between pedestrian scenarios and full mobile (60km/h) is around 2 – 3dB in required SNR. Consequently, a higher reuse factor may be required. There are other not so obvious reasons why additional capacity may be required in a mobile scenarios (for example, a certain amount of resources may need to be reserved for handoff and load balancing). However, at this point I don’t have a quantitative view on how this will impact spectrum requirements,” said Redline’s Kevin Suitor. Is a lack of spectrum causing quality of service issues? According to Redline, the lack of spectrum will dictate a need for “tighter” frequency reuse and hence lower quality of service. Segmentation can be used to mitigate this problem, but then the achievable per sector capacity will be reduced and this will obviously impact the overall business case. Haig from WiMAX 2020 added, “The lack of spectrum will start causing network capacity issues as the traffic in a network increases over time. Unfortunately, for those operators who do not have enough spectrum, the only way to mitigate these issues is by adding more base stations, which is an expensive proposition. So what is the situation globally? Exhibit 3 clearly shows that the more spectrum owned, the more likely an operator is to deploy commercially. For instance, 66% of those who own more than 50 MHz have deployed a commercial network, while a large proportion of those who own 20 MHz or less of spectrum are either doing nothing with it or are in a trial phase to test the robustness of the network under tight spectrum conditions. Exhibit 3: Deployment Status by spectrum ownership
It will be interesting to see how much additional spectrum is acquired by 3G operators to expand their services in the next two years. I want to thank the WiMAXCounts™ team and in particular Cintia Garza for putting this unique hard data together. For more information you can contact the author. afellah@maravedis-bwa.com Copyright © 2008 by Maravedis Inc. All Rights Reserved.
|
|
|
|
|||
| Created by: Virage>2.0 |
