The Internet World Stats recently published their numbers for June 2015. There are now over 7 Billion people on this planet, but only about 45% have Internet access. Only 6 months ago, just over 42% of the population had Internet access. The Internet is continuing to grow as more network infrastructure is deployed and more communities are connected.
While there is a high percentage of Internet access in Europe and North America, there are communities in Asia, Africa, and South America that have yet to come online. From their statistics, we can see that in North America, Internet penetration is almost 90%, while in Africa, it is almost 30%.
Much has been discussed about the socioeconomic inequality in the world and this is clearly evident on the Internet. The “digital divide” separates those who live in well-connected communities with better infrastructure from those who have limited access to necessary services. Communities that have limited Internet access are also likely to have challenges with basic human needs like obtaining adequate food, drinking water, sanitation, public safety, electricity, healthcare, and education.
For some people, having continuous access to the Internet may seem like a basic human right. These hyper-connected individuals may have a jokingly revised version of Maslow’s Hierarchy of Needs that adds two new lower layers. One wonders how these hyper-connected people would survive and find food if they didn’t have GPS-enabled smart-phones!
Seriously though, many countries have established broadband initiatives to ensure their citizens have access to wired and/or wireless Internet infrastructure. However, as rural communities move forward with their mobile broadband Internet initiatives, they can utilize high-speed wireless technologies and avoid costly terrestrial wired network infrastructure. These emerging communities are leapfrogging the legacy buried or aerial copper networks that have historically been used for broadband Internet access. These rural communities can utilize wireless technologies to cover their broad geographies and challenging terrain with newer 4G technologies and receive high bandwidth data service.
An additional challenge these emerging communities encounter as they embark on these greenfield Internet access deployments is availability of public Internet addresses. Now that IPv4 address exhaustion has occurred in many parts of the world, there may not be sufficient available IPv4 addresses for these new buildouts. Alternatively, it would be easy for these networks to obtain IPv6 addresses as they will likely want to design these new networks to use dual-protocols from the outset. When building a new system, it is best to make sure it supports IPv4 and IPv6 to have the broadest connectivity options for users.
Using a broadband wireless network that covers these communities could bring other societal benefits. For example, an IPv6-enabled sensor-based system that detects soil moisture content could intelligently use limited water resources only when crops need watering. Internet of Things (IoT) technologies could help with frugal electric energy utilization in homes and buildings, monitor precious food storage and distribution, or even rapid detection and notification of earthquakes and tsunamis. Medical devices could monitor vital signs of geographically sparse populations and make healthcare only one digital step away. IPv6 would facilitate these applications that require end-to-end communications using global addresses.
The issue is that even these very modern networks will still need to support IPv4. Even though IPv6 usage on the Internet continues to double every year, there is a majority of content accessible over only IPv4. Even though these new networks may be deploying IPv6, they still need a strategy to continue to use IPv4 for the foreseeable future. These new networks for rural communities will still be burdened with the legacy IPv4 transport. There are options for prolonging the legacy of IPv4 and some of these techniques rely on using multiple levels of NAT. These service providers could use Carrier Grade NAT (CGN) methods and only have to allocate private IPv4 addresses to the new Internet user population. There are many downsides to CGN and a new carrier would only want to avoid CGN if possible or perform the NAT function only once and not twice.
One approach that should be considered is one that leverages IPv6 to avoid double-NAT architectures. This option involves tunneling the private IPv4 traffic within native IPv6 to a provider gateway that will NAT the IPv4 traffic when it exists to the Internet (see RFC 6144 and RFC 6145). When encapsulating IPv4 packets with IPv6, next-header 41 per RFC 2473 is used. Since the outer IPv6 encapsulated packet is uniquely identified by its global IPv6 addresses, the inner IPv4 packets could use private addresses that might overlap between subscribers. If the subscriber can still reach their IPv4-only Internet sites, they would be unaware that they are using IPv6. Furthermore, their IPv6 connections going to IPv6-enabled Internet sites would not be encapsulated and would be transported natively to the Internet.
Following is a picture from Mark Townsley, Cisco Fellow, presented at the RIPE 65 meeting on September 24, 2012 that shows how IPv4-in-IPv6 tunneling can support subscribers with legacy IPv4 over an IPv6-only core service provider network.
This form of this encapsulating method is called Mapping of Address and Port (MAP). This is a service provider technique that tunnels legacy IPv4 packets within IPv6 packets and leverages an IPv6 backbone. The MAP method uses extra port range identifiers within the IPv6 address to reflect the IPv4 address plus a port number. The encapsulated form of MAP, as described above, is called Mapping of Address and Port with Encapsulation (MAP-E) (RFC 7597). The translation for of MAP, is called Mapping of Address and Port with Translation (MAP-T) (RFC 7599). Since MAP is a stateless technique, it is quite scalable because it avoids maintaining NAT state the gateways. If you are interested in learning more, Cisco has a Technical Guide to Mapping of Address and Port (MAP).
With this MAP technique, these new broadband Internet networks can build out the network infrastructure with IPv6-only and avoid having to obtain public IPv4 addresses for the core network infrastructure. The service provider would only need to obtain sufficient public IPv4 addresses for the NAT pools for the MAP Border Relay (BR) at the upstream Internet access points.
These new broadband Internet deployments could benefit from new CPE that will likely start out being dual-protocol capable and enabled right from the start. These networks would not have to support legacy CPE that burden today’s broadband Internet providers that service urban areas. Having subscribers with newer CPE and mobile devices means that the service providers are able to avoid supporting legacy devices and avoid those backward compatibility costs.
Having data networking to remote tribes and indigenous people is also essential if public safety services are to be provided to them. Here in the United States, the Department of Homeland Security (DHS) recognizes that Emergency Services is one of the sixteen critical infrastructure sectors and they recognize the importance of having a nation-wide incident management capability. This initiative’s plan includes providing these services to rural and tribal populations across a large geography. An example of this is the First Responder Network Authority (FirstNet) in the United States. FirstNet is a Radio Access Network (RAN) initiative supported by the National Telecommunications and Information Administration (NTIA) which is to be deployed by the states to provide public safety services to those in rural areas. Public safety communications systems like Project 25, sometimes referred to as P25, operates in the 700MHz frequency Band Class 14. These systems are new digital narrow-banding systems that provide for voice and integrated data and foster cross-agency collaboration during times of crises. These LTE systems follow the same standards set forth by the 3rd Generation Partnership Project (3GPP) and they require the use of IPv6.
Another consideration is that tribal populations and indigenous people are often leery about losing their culture and concern about Western influences with Internet connectivity. However, tribal nations and indigenous people have a lot to share with the world at large. We want them to maintain their rich cultures while leveraging the Internet to learn and share with the world. Tribes can use the Internet to do this. The world would also benefit when details of their native languages and cultures are preserved to share with future generations.
As new populations obtain Internet access, they should strive for dual-protocol connectivity initially. This will give them a head start by avoiding having to deploy an IPv4-only system and later having to add IPv6. Having an IPv6-enabled backbone and dual-protocol access network will allow these new communities to communicate with the largest possible Internet population and share their rich cultures and heritages.