2016 is upon us, and as everyone looks ahead to see where the APAC telecoms sector is going, Telecom Asia looks to the past to see how the sector got here in the first place.
In our second annual Rewind feature, we wrap up 2015 with a look back at the telecoms news stories that dominated the headlines and captured our readers’ attention (i.e. generated serious pageviews) on telecomasia.net.
Among the big stories this year: Spectrum wars at WRC-15, standards wars for narrowband LTE, 5G defined, the rise of OTT video services, an M&A feeding frenzy (including one that hasn’t actually happened yet), Project Loon, and the strange and terrible saga of the [alleged] Great Firewall of Thailand.
The Great Firewall of Thailand
Thailand was a major hotbed of telecoms-related news as the military government continued implementing various telecoms reforms and conducted a marathon 4G spectrum auction.
But the story that really raised eyebrows was the reported plan by the government to create a single internet gateway in and out of the country that would be managed by state-owned CAT Telecom.
When the news broke in September, critics instantly blasted the move. Some described the single gateway plan as “the great firewall of Thailand” whose real purpose was to allegedly control the flow of information on the Internet. Others argued that routing the Internet through a single gateway controlled by one entity would result in slower internet speeds and a single point of failure. Yet others complained it heralded a return to the monopoly days when CAT Telecom ran the only commercial Internet gateway in the country.
After the news broke, various government and telecoms parties issued sometimes conflicting statements about whether the plan would go ahead, or ever existed in the first place.
Information and Communication Technology Minister Uttama Savanayana said the single gateway was just a proposal under review, and that the goal of implementing it would be to reduce costs for Thailand’s ISPs. Acting CAT CEO Colonel Sanpachai Huvanandana said CAT was pushing ahead with a rebranded single gateway as a cost-saving measure.
By mid-October, Deputy Prime Minister Somkid Jatusripitak said the plan had been scrapped, according to Reuters. But that didn’t stop the Anonymous hacking collective from declaring war on Thailand’s government - and in particular, CAT Telecom - over the idea a week later.
“The latest project of the Thai military government is to deploy a single gateway in order to control, intercept and arrest any persons not willing to follow the Junta orders and your so called morals,” said a statement from Anonymous.
Shortly afterwards, the websites of the Information and Communication Technology and finance ministries and CAT Telecom were among those targeted in a DDoS attack.
WRC is over (if you want it)
The hottest telecoms-related event of 2015 was WRC-15, in which companies representing the mobile, satellite and broadcast sectors lobbied the ITU hard for new spectrum allocations.
And in the end, everybody won. More or less.
The main battle was between the GSM Association and the satellite industry, who have already been scrapping over use of the extended C-band (3.4GHz-4.2GHz) for mobile broadband services.
For WRC-15, the GSM Association argued in official documents that the mobile sector will require another 600-800MHz worth of spectrum by 2020 to handle traffic demand over the next five to ten years, and asked the ITU to harmonize not only lower frequency bands (470-694/698MHz) and extended C-band (3.4GHz-4.2GHz) for mobile broadband use, but also other satellite bands, including the S-band (2.7-2.9GHz) and L band (between 1300MHz and 1518MHz).
The satellite sector accused the GSMA of taking more spectrum than it needed, claiming that in most markets, less than 50% of the spectrum already earmarked by the ITU for mobile broadband services has actually been licensed - and much of what spectrum has already been licensed isn’t being fully utilized.
In the end, the GSMA got three new globally harmonized bands: the 700 MHz band (694-790 MHz); the lower 200 MHz of the C-band (3.4-3.6 GHz) and L-band spectrum (1427-1518 MHz). What it didn’t get was harmonized spectrum for the sub-700 MHz band (especially 610-694/698 MHz) or the rest of the extended C-band (3.6-4.2 GHz).
The satellite sector was pleased about the latter, and wasn’t too concerned with the L-band allocations, as there are clear measures to protect the satellite users in the 1518-1559 MHz portion of the band. The satellite sector also scored a victory by keeping the harmonized frequencies for the C, Ku or Ka bands off the menu for WRC-19 as possible 5G spectrum candidates.
The GSMA was clearly disappointed with the decision to preserve the sub-700 MHz bands for terrestrial broadcasters in Region 1 (EMEA and central Asia) until at least 2023 (when the topic will be revisited at WRC-23). Chief regulatory officer John Giusti made it clear in public statements that the decision was a mistake, arguing that it was possible for broadcasters and mobile to co-exist in the UHF band, and anyway more people will watch video on mobile.
Meanwhile, everyone is already bracing themselves for WRC-19, which will be focused heavily on global and regional allocations for 5G technologies, particularly millimeter-wave bands above 24 GHz. As Caroline Gabriel noted in the Wireless Watch newsletter published by Rethink Research, the debate regarding millimeter-wave at WRC-19 won’t be just about harmonization and economic benefits: “There are many technical challenges to address too, particularly if these bands are to be used for access as well as backhaul - reducing shadow and loss, deciding on an air interface, working out how to integrate it into mass market devices and to aggregate it with lower bands.”
5G defined (sort of)
One of the hot tech topics of 2015 was 5G, even though few people in telecoms agreed on what it actually is. They know what it isn’t - it’s not a relatively straightforward network RAN upgrade that would supersede the previous generation, but rather an amalgam of technologies combining everything from legacy 3G, 4G and Wi-Fi to cloud, big data analytics and network virtualization, among others - all for the purpose of delivering ubiquitous, seamless broadband to any device (all 50 billion of them) by 2020. But with different vendors promoting differing visions of what they called “5G”, there’s still a lot of confusion over what counts as 5G and what doesn’t. And operators being operators, what they really needed in 2015 was a benchmark they could get their heads around - how fast is it?
The ITU gave them the answer in June 2015. During a meeting of the ITU-R Working Party in the US, the ITU defined the goals, processes and timeline for 5G development. They even gave it a name: IMT-2020. And according to various internet reports, the ITU defined 5G as networks capable of transmitting data at up to 20 Gbps.
However, that number didn’t actually come from the ITU - not publicly. The ITU statement and links to supporting documents make no mention of data speeds or other performance benchmarks. It merely outlined the next steps in establishing “detailed technical performance requirements for the radio systems to support 5G, taking into account the needs of a wide portfolio of future scenarios and use cases, and then to specify the evaluation criteria for assessment of candidate radio interface technologies to join the IMT-2020 family. These new systems, set to become available in 2020, will usher in new paradigms in connectivity in mobile broadband wireless systems to support, for example, extremely high definition video services, real time low latency applications and the expanding realm of the Internet of Things (IoT).”
The 20 Gbps number came from the Korea Times, which cited claims form South Korea’s Ministry of Science, ICT and Future Planning that IMT-2020 networks must support data speeds up to 20 Gbps, and have a capacity to provide over 100 Mbps data rates to over 1 million IoT devices within a square kilometer.
Sister publication FierceWireless Europe contacted ITU spokesman Sanjay Acharya for clarification. Acharya said: “As of now, I understand the peak data rate of IMT-2020 for enhanced Mobile Broadband is expected to reach 10 Gbps. However, under certain conditions and scenarios, IMT-2020 would support up to 20 Gbps peak data rate.”
So between 10 Gbps and 20 Gbps, then.
LTE officially goes narrowband
All eyes were on the 3GPP in September as it proposed a narrowband version of LTE for inclusion in Release 13 that will allow LTE networks to support IoT applications with very low data rate and power consumption requirements. The technology, billed as “Narrowband IoT LTE,” is expected to be finalized in early 2016 - along with other IoT-oriented low-rate, lower-cost versions of LTE like LTE Cat 1, Cat 0 and LTE MTC.
One reason the decision raised so much interest - apart from a general eagerness by cellcos to use their LTE assets (and associated frequency bands) to cash in on the IoT - was a last-minute standards war for Narrowband IoT LTE.
Actually a number of companies have been clashing over narrowband LTE technologies for some time. But the week before the 3GPP meeting, Nokia Networks, Ericsson and Intel announced they had teamed up to back Narrow-Band Long-Term Evolution (NB-LTE), a technology seen as a direct challenge to Huawei Technologies, who backed its own Narrowband Cellular IoT (NB-CIoT), which had already gained operator support from heavy-hitters like Vodafone and China Unicom.
The main difference between NB-LTE and NB-CIoT came down to how much of existing LTE networks can be repurposed for IoT. Critics of NB-CIoT’s approach claimed that it requires new chipsets and wasn’t backwards compatible with any LTE network older than Release 13.
NB-LTE, by contrast, “can be fully integrated into existing LTE networks, works within current LTE bands and does not need an overlay network,” according to Nokia. In other words, NB-LTE uses more of the existing ecosystem and thus promises better economies of scale.
It’s a moot point now, of course - Narrowband IoT LTE in its final form is expected to use specs from both technologies.
According to Chris Taylor, director of RF and wireless components at Strategy Analytics, Narrowband IoT LTE will have a significant advantage over proprietary air interfaces for low-power wide area networking (LPWAN) technologies for M2M and IoT like SIGFOX, LoRaWAN (Long Range WAN) and Ultra Narrow Band (UNB) because of the huge base of LTE networks that will support it. Potential apps include metering, environmental and industrial monitoring, object location tracking, e-health, wearables and sensors.
“Narrowband IoT LTE would compete with other LPWAN air interfaces, but could complement local area wireless mesh networks using ZigBee, Bluetooth Smart or Z-Wave for example,” Taylor wrote in a blog post.
But while Narrowband IoT LTE is scheduled to be frozen in early 2016, there’s still a chance it may not make Release 13, Taylor added.
“The only concern we have heard so far is that including Narrowband IoT LTE in 3GPP Rel. 13 could delay the 3GPP release. Eager chip suppliers note that LTE MTC is already well defined, and provides low cost, low power consumption and coverage gains compared to LTE Cat. 0 and above, so perhaps the cellular industry should move Narrowband IoT LTE to Rel. 14.”