5G Won’t Fully Develop Until 2020, Watch For Network Consolidation in 2014
Global 4G/LTE deployments are well underway. We’re even starting to see the first LTE-Advanced (LTE-A) rollouts starting in South Korea. This naturally has wireless vendors thinking about “what’s next?” And “what’s next” is usually referred to as “5G.”
For some perspective, let’s first look at timelines. Typically, new wireless standards have come roughly 10 years apart. NTT DoCoMo originally proposed 4G/LTE around 2004, with mobile device data consumption primarily driving 4G/LTE technology. But Japanese operators and Verizon Wireless in the U.S. being on non-GSM wireless standards, along with the threat of a vanishing ecosystem, also accelerated the 4G/LTE timeline. Extrapolating from this experience, we can expect the wireless industry to define 5G standards over the next four to five years, with deployments beginning no sooner than 2020. Semiconductor suppliers have to be two years ahead of equipment vendors and operators, but regardless, 5G will not have an impact for anyone for quite some time.
Do people or even industry experts know what 5G could or should be? Not at all! The consensus is that more speed is needed, and typically sets 1 Gbps to a wireless device as the target. If we doubled traffic every year, in 10 years we would need more than 1000x the capacity we have today. Obviously, video content can drive the demand for such bandwidth increases, but it’s still hard for an individual consumer today to imagine what to do with 1 Gbps of bandwidth on a smartphone or tablet.
Ultimately, 5G only makes sense if it generates additional revenue for mobile operators and service providers. This initially wasn’t the case for 4G/LTE. But new software-defined radios, and much simpler, flatter, and more cost-effective architecture with significant CapEx and OpEx savings provided significant incentives to upgrade from 3G. Because of readily available LTE handsets, 4G/LTE adoption happened almost twice as fast as the previous wireless standard. Operators who adopted 4G/LTE quickly gained significant market share compared to their slower competitors, and consequently increased their ARPU (average revenue per user).
So, can 5G collapse the service provider and network silos that exist today? I’m talking about mobile, fixed (residential as well as business), and cable networks on one side, and the content providers (classical, as well as over-the-top, or OTT) on the other side. This would not only allow further consolidation, reducing OpEx and CapEx, but also force service providers to rethink how they can generate more revenue from their wireless assets, rather than leaving more and more services to the OTT players. Some early indications hint that this may already be happening, although 5G would need to acquire enough regional reach to make this work.
Even though its commercial impact is still several years away, we can safely make a few predictions about 5G in 2014:
- More Spectrum Will Be Imperative, Millimeter-wave Frequencies May Be The Answer: With spectral efficiency gains reaching diminishing returns, 5G will need a lot more spectrum. LTE today requires 20MHz of spectrum, and LTE-A, about 100 MHz. Based on that, 5G would need hundreds of MHz. We may have to look at millimeter-wave frequencies to find that much available spectrum.
- Spectrum Efficiency Will Be Key: Utilization of spectrum resources will have to become much more efficient. Besides small cells that are sharing spectrum spatially, we may also have to look at dynamic spectrum access, and wireless service providers may have to statistically share spectrum.
- Expect to See RAN Sharing: We will need to use network resources more efficiently (read: RAN sharing). In this scenario, service providers would lease network capacity from the Mobile Network Operator (MNO) with the financial strength to build a RAN infrastructure. This will enable CapEx and OpEx consideration tradeoffs for the MNOs as a whole.
- It’s All About Content Delivery: Operators will have to build 5G wireless networks with content delivery in mind, just as they built 4G with data delivery in mind. In order for MNOs to generate revenue from content distribution, the network needs to get smarter, offer much more differentiated service delivery capabilities than even 4G/LTE, and allow much faster delivery of new services and capabilities than before. Both SDN and Network Function Virtualization (NFV) will likely play a key role in these next-generation networks.
About the Author
Martin Nuss, Ph.D. is Vice President, Technology and Strategy and Chief Technology Officer at Vitesse Semiconductor. Dr. Nuss has over 20 years of technical and management experience. He is a recognized industry expert in timing and synchronization for communications networks. Dr. Nuss serves on the board of directors for the Alliance for Telecommunications Industry Solutions (ATIS) and is a fellow of the Optical Society of America and IEEE member. He holds a doctorate in applied physics from the Technical University in Munich, Germany.