IPv6 Technology for IoT
The Internet of Things means IoT is the network of physical devices which are supported with the sensors, software, and other technologies to connect and exchange the data with the other devices and systems over the internet. IoT is the network of these devices which can range from an ordinary household object to ultra-modern industry level devices. Today, there are more than 40 billion IoT devices, and this number can grow up to 125 billion by the end of 2030.
At the time of IoT’s first conceptual framework in 1990, few expected the immense impact it would have on traditional and industrial applications. The idea of a fully networked deft of objects connected to the internet was certainly likable, but the massive growth of the technology outpaced these visionary expectations. While this massive increase certainly indicates well for the IoT marketplace, there are major concerns about whether the current internet protocol, IPv4, will be able to withstand the fresh flood of new devices.
The core internetworking method called the Internet Protocol system used by the current world internet is in charge of defining unique identifiers, called IP addresses, to every connected device over the internet. The system is also automated with delivering packets of information from one source to another, which allows the data to flow between devices with only a minor risk of confusing the deliberated destination. So how might these billions of IoT devices cause problems for this networking method? Here is the answer to this question!
Limitations of the IPv4 Technology
After a lot of discussions over the Internet of Things (IoT), it is assumed that IP address space is inexhaustible that will scale as the IoT scales to unanticipated proportions. But the IP address space is not infinite. In fact, the IPv4 address space has been used up since February 2011. And this is one of the reasons for considering IPv6 (Internet Protocol version 6 ) for the future of IoT.
The IPv4 technology which is used today was created by The United States Department of Defense early in the 1980s. This protocol was mainly deployed for the part of the Advanced Research Projects Agency Network. This agency later became the foundation for how the modern internet operates. The issue, however, is that IPv4 relies on 32-bit addresses; this limits the number of unique identifiers (devices) to around 4 billion, according to the Federal Communications Commission. This is the main hurdle for large-scale IoT deployments, as there simply aren’t enough IP addresses for all the devices planned for deployment. In simple words, 4 billion addresses are not enough to serve for nearly 40 billion devices.
This limitation has been solved by many of the organizations by implementing extra layers of technology, such as network address translation(NAT), which allows multiple connected devices locally to share a single public IP address. But with a spate of new M2M(machine-to-machine) applications on the rise, this method will not remain livable much longer.
The IETF RFC 791 standard is the first publication of IPv4 specification published in 1981 when the internet was known as ARPANET. Lack of scalability and limited addresses of IPv4 has become the issue in 1992. Changes within the routing architecture and therefore the allocation of address space called CIDR (Classless Inter-Domain Routing), which were standardized in 1993 (RFC 1518, RFC 1519), have significantly bogged down the consumption of addresses. In today’s rapidly developed internet, the world has crossed a point where most IPv4 addresses are already distributed and in use. Now, the situation has pretty much changed from when CIDR technology was introduced in the mid of 90s, which allowed blocks of addresses to be allocated according to real needs. At the same time, Regional Internet Registries (RIRs) were formed, responsible for distributing addresses and numbers of autonomous networks according to the rules developed by the regional Internet community. In this list, first were RIPE NCC (1992) and APNIC (1993), second ARIN (1995), third LACNIC (2002), and last was AfriNIC (2005).
Some of the limitations of IPv4 are as follows:
- The lack of address space as the size of the address space is quickly depleted.
- Weak protocol extensibility as the insufficient size of the IPv4 header does not have the space for the additional parameters.
- The problem of security of communications as no solutions are given to limit access to information hosted on the network. IPv4 has never been designed for security, instead, it is created for an isolated military network, later, it has been adopted for public education and research networks.
- Quality lacks in-service support about bandwidth, delays required for the smooth operation of some network applications are not supported.
- Geographical restrictions as this protocol were created in the USA, this country is involved in the distribution of IP addresses as well. Nearly 50% of all addresses are reserved for the USA.
The Revolution of IPv6 Technology
In contrast to IPv4, the IPv6 system is based on 128-bit addresses and can facilitate close to 340 undecillion unique IP identifiers. This massive increase in capability promises to supercharge the IoT revolution, but that’s not all the new system improves upon. IPv6 also supports auto-configuration, integrated security, and various new mobility features, enabling a higher degree of network complexity. While the IPv6 system is not backward compatible with IPv4, both protocols can work in parallel without significant disruption. The protocols like stateless transport layer protocol TCP and stateless application layer protocol HTTP function in the same way for both systems.
With the partnership of several large countries, the Internet Society formally launched IPv6 in 2012 and has continued its acquisition over the past 7 years. At the time of 2018, the Internet Society reported that around 25 % of all internet-connected networks possess IPv6 connectivity, with 49 countries delivering at least 5 percent of their traffic over the system. According to the survey made by Google, 24 countries out of these 49 countries have IPv6 traffic that already exceeded the 15 % of total network usage of the current case, which is quite an achievement in such a small time frame. Broadband internet service providers have been a driving force in the transition from IPv4, both in the U.S. and abroad. Comcast leads the charge in the U.S. with IPv6 deployment exceeding 66 percent, whereas British Sky Broadcasting has reached an impressive 86 percent in the UK.
The big advantage of IPv6 is its highly efficient multicast communication feature, which eliminates the need for routine broadcast messaging. According to the article from Network World, this improvement can help preserve the battery life of IoT devices by reducing the number of network-wide packets each device must process. Take an example, an Address Resolution Protocol that is used for binding MAC addresses to IPv4 addresses. It is done by sending the same message to every node present in a network, irrespective of their involvement in the exchange. Once the message is received, the ARP packet is processed on the device level, which can unnecessarily drain battery power and lead to more frequent replacements.
Difference between the IPv4 and IPv6
Both the IPv4 and IPv6 are connectionless datagram protocols. They both are used for addressing and routing packets among the different host nodes. “Connectionless” means that there is no need to establish a connection before exchanging data over the internet. “Unreliable” means there is no guarantee of delivering the packet. IPv6 is not only the extension of IPv4 addressing but is way more than that. In RFC 2460, IPv6 is first defined as it is a complete implementation of the network layer of the TCP/IP protocol stack and it covers a lot more than simply address space. It is the expansion from 32 bits to 128 bits. This increases the ability of IPv6 to allocate almost unlimited addresses to smart devices in the upcoming years.
Some of the key points where IPv6 provides improvements over IPv4 are:
- Due to the removal of fragment packets, overhead in the routing process has been removed. This made routing efficient and faster in IPv6.
- In the IPv4, there is no way to differentiate delay-sensitive packets from bulky data transferred. It made the workload extensive. But, IPv6 has the built-in Quality of Service (QoS) for solving this issue.
- IPv6 removed the concept of NAT to increase the address space.
- IPv4 is never made for security purposes and has no security. But IPv6 has an inbuilt Network Security Layer (IPsec).
- IPv6 has a stateless address autoconfiguration mechanism. It makes network administration easier such that IPv6 deals with installation parts in an automated fashion whereas, in IPv4, it is manual and complicated.
- With the improvement in the header structure of IPv6, it is removing processing overheads. It has only essential fields in the header which are used frequently. In this case of IPv6, options are handled differently.
The IPv6 packet contains its own frame Ethertype value, which is 0x86dd. It makes for the receiving side to differentiate the frame content on the same interface.
The following table shows the different fields of IPv4 Header (Fig 1) and IPv6 Header (Fig 2):
Network bandwidths are significantly used as IPv6 uses multicast addressing mode over broadcast addressing mode (Table 3). IPv6 uses a hierarchical address allocation mechanism for efficient routing and Maximum Transmission Unit (MTU) for path allocation.
Challenges for the IPv6 Technology
Although IPv6 technology has various advantages over IPv4 technology, there are some challenges for IPv6 protocol, such as -
- You have to enable the IPv6 routing which is dependent on the type of the system. Also, if it is enabled manually, then in that case the IP needs to be typed, and also it needs to be remembered. This is not an easy task to remember IP since it contains a large number of letters and numerics.
- It is challenging to fix the prefixes on most of the topological drawings when you use the IPv6 protocol due to its extra space used.
- Upgrading the devices is also a challenging part of the IPv6 protocol as many of the internet devices by default are not designed on the basis of IPv6 adoption. Even upgrading the device is no simple task since it requires expertise for seamless transmission.
- Due to the lack of strong backward compatibility in the process of transition from IPv4 to IPv6, it becomes slow and complex. Some of the money is needed to be paid by ISP (Internet Service Provider) for switching from IPv4 to IPv6 protocol and vice versa.
- Due to these conversion complexities, the communication between IPv4 and IPv6 is not easy and direct.
- There is a need for Local Network Management for assigning these IPv6 addresses to devices. It will reduce the manual process which is complicated.
Conclusion
Today, security is a major challenge in this world and also, a major concern in this IoT technology. Security involves many parameters which are important to be taken care of such as -
- Authentication: to make sure the identity of the end-user.
- Authority: to make sure that the user has the proper rights to perform actions.
- Integrity: Data that is present in the communication should not be tampered with and need to be received in the correct form. Data Integrity should be maintained.
- Confidentiality: Data communication must be done in a secure form to avoid compromising confidentiality.
However, the IPv6 protocol fulfills these security requirements because it is created in a way that it can carry out IPSec security. IPSec contains Authentication Headers (AH). This header has inbuilt methods into the network firewall for authentication purposes.
Another important parameter is the mobility of IoT devices since they need to be always connected over the internet irrespective of the physical location. It creates various challenges to the protocols and also to the networks of devices because mobility protocols must tackle the built-in features of IoT, like hard duty cycles and resource constraints. IPv6 protocol has this mobility because it provides smooth and flawless connections when smart devices are moved from one location to another. To achieve this, avoid triangular routing. The triangular routing is the technique in which connections to the proxies are sent even before transferring to the host. Thus, without any renumbering, all the subnets are moved to a new router. With the help of automatic configuration by removing the concept of static IP addresses, the subnetting problems are completely avoided in IPv6 protocol.
Due to the 128-bit address scheme, the IPv6 technology provides a large address space which is indeed enough to serve around 125 billion smart devices in the future.
References
Coauthors: Pradunya Maladhari, Shantanu Sontakke, Talib Hussain, and Aadarsh Kandewar
