By Garrett Seeley
People can be confused by the TCP/IP because it is a number system. The meaning of numbers are not as clear as words explaining them. However, math is the language of machines and therefore networks. It is essential for a technician to understand these numbers.
In TCP/IP, an IP address is used to identify a machine and identify a network. That may seem odd, but it is like something people use daily: a phone number. Think of a phone as a way to communicate over a network. Phone numbers identify a group of people with an area code and an individual with a local phone number. In this way, phones follow the main goals of any network. All networks identify a group of devices and then identify the individual device in that group. Computers, phones and medical devices follow this pattern even if they do not use TCP/IP to do it. Consider a phone as an example of a network.
In the phone system, the area code is a three-digit number in parenthesis before a seven-digit local number. Similarly, IP addresses, or just IPs, are divided into four groups of three digits from 000 to 255 i.e., 255.255.255.255. This is because each group is an 8-bit number, and the largest number is 1111 1111 or 255 in decimal form. If this is not clear, please review binary math and hexadecimal math. This article will reference, but not focus on data conversions.
This article will consider the numbers pertaining to Local Area Networks, or LANs. It will focus on the IP version 4 addressing model, or IPv4, when discussing IP addresses. IPs are structured in what is called an Octet and there are four of them. The octets explain where a network, a subnet and a host number are in an IP. However, it does not do this alone. There is a subnet number that accompanies an IP address. This subnet tells a computer how to make sense of the IP. An IP by itself will not make sense to the computer, nor does a subnet by itself. They must be used together.
Consider the graphic above: An IP address 192.168.000.001 with a subnet 255.255.255.000. This is a common IP address and subnet. Notice how the subnet number changed from 255 to 0 at the third to fourth octet. That means that the change from network to host occurs in the last octet of the IP address. The subnet tells that the first three octets are the network part of the IP, and the last octet is the unique host number part of the IP. The rule is this: The network part must match for two devices to be networked, and the host number must be different from all other host numbers. If these conditions are met, the devices will communicate. This must be exact, or the network will not function. That is just like the phone system, a phone can be in the same area code, but cannot use the exact same number as another phone. The IPs follow the same pattern.
In an IP address of 192.168.0.1 with a subnet of 255.255.255.0, the network address is 192.168.0.XXX, where XXX is the individual host number of a device. The total number of individuals in this network about 250 unique hosts. That is plenty of devices for a home or a clinic, but not nearly enough for a full hospital. Therefore, bigger networks change the subnet for more hosts. Changing the Subnet to 255.255.0.0 moves the network and host part of the IP. It is no longer at the last octet, but now is between the second and third octets. The first and second octets are now the network portion of the IP, and the last two octets are now the host portion. Our available numbers expand to 250 times 250, or over 64,000 different unique host numbers.
In this way, IPs are expandable. If more IPs are needed, just change a number in the first two octets for a separate network. In this way, IP addresses are also scalable. This is how hospitals can have several different networks connected to the same switches. The devices listen for their networks and communicate only to the other hosts in their network. This is called network segmentation and is one of the most essential concepts of TCP/IP.
It is important to note that these IP numbers are not compatible with the Internet without a router. LAN IPs do not appear on the Internet. A router understands its internal LAN IP numbers and knows how to switch packets between networks. Routers know if a packet’s destination is inside the LAN or over the Internet (a Wide Area Network or WAN). An example of an Internet number would be 1.1.1.1, which is a server over the Internet that we can use in troubleshooting. Most LAN would not use a host number like this one. If this is seen, a router transfers the related packets to an Internet service provider for Internet access. For a device using the Internet, two more settings must be used; they are the Gateway and the Domain Naming Service or DNS. A gateway is just what it sounds like, an exit to leave a network. A gateway IP is the number a device must use on its LAN to communicate on another network. A DNS is a server that converts a webpage name to an IP. Think of it like the phonebook for the Internet. Typing 142.251.45.46 in a web browser may lead to a Google website, but it is easier to remember to type Google.com in the browser. The DNS server makes the connection between webpage names, called Uniform Resource Locators or URLs, and the web IP addresses. In this way, people use the Internet referring to a webpage by name. Keep in mind that it is still just a set of IP numbers in the background.
In this way, medical devices use IPs to connect with other devices and bring the hospital together into one network. In concept, all networks accomplish the same tasks of identifying groups of devices and identifying each device as a unique number in that network. The settings required for TCP/IP to work on a LAN are its IP address and Subnet. Optionally, a Gateway and a DNS can bring Internet connectivity. The wonderful thing about TCP/IP is that anything using it uses the same settings. Therefore, understanding it will help a technician work on many devices. Following this information will help a technician gain more familiarity on networking and increase their value to the field. Networking is, after all, the future of HTM.
Garrett Seeley is a Biomedical Equipment Support Specialist-Imaging at the Veterans Administration Healthcare System-North Texas.
