SY110: Intro to the TCP/IP Stack: Physical Layer



Intro to the TCP/IP Stack: Physical Layer

Learning Outcomes

After completing these activities, you should be able to:


Introduction to the Physical Layer

As we've learned so far about the TCP/IP Stack, data primarily originate from the user through an application, thus starting at the top of the stack. The Physical layer governs the physical transmission and reception of signals across various media. It deals with electrical, mechanical, and procedural aspects of interfacing with physical media, ensuring the reliable transmission and receipt of data between devices. This includes:

NRZ Encoding
Non-Return to Zero (NRZ) Physical voltage is translated into logical 1s or 0s.
Wired Infrastructure: Cables (copper, fiber-optic, coaxial) establish physical connections between devices, data centers, and across continents. Undersea cables, in particular, are crucial for global data traffic.

Wireless Technologies: Short-range (Bluetooth, Wi-Fi, etc.), mid-range (mobile phone LTE, 5G, etc.), and long-range (microwave, satellite communications (SATCOM), and HF and lower-band military data communications, etc.) wireless communication technologies expand network accessibility without the constraints of physical cables, enabling mobility and flexibility in connectivity.

Signal Encoding and Multiplexing: Methods like Non-Return to Zero (NRZ) and Time-Division, Frequency-Division, and Code-Division Multiplexing (TDM, FDM, and CDM) optimize data transmission efficiency, crucial for both wired and wireless communications.

Global Map of Undersea Cables. Map of submarine cables used in providing global telecommunications
data, as of Sept 2023. Courtesy of Telegeography.

The first lab in this class introduced you to Computer Architecture, disassembling and reassembling a computer. This lab intended to expose you to the physical components of a computer system to make the connection with the observable properties, materials, and components that are not often associated when primarily working within the logical layer of the cyberspace model. This class intends to reveal the physical infrastructure required to support data networks but to also gain an appreciation for the physical aspects required for interconnecting billions of people and systems globally.

Image of cable tracks above data racks
The Physical layer is the most fundamental layer and is sometimes combined in other networking models with link-layer because it is often associated with one another. Other physical items that enable this layer to operate will include the medium and signaling that allow the transmission of data. This can entail the circuits, processors, repeaters, antennas, fiber, copper, wiring, power supplies, and fans that regulate and maintain the optimal temperature of the hardware equipment. Without physical structures to allow electrons to flow, systems would not be operational to send the bits across wires. In the case of wireless networks, RF propagation and signaling allow conveying data through the physical media of the aerial environment and space.

Encoding methods used to translate signaling, such as Non-Return to Zero (NRZ), where voltage is used for representing logical 1's or 0's as depicted in the diagram on the right. In addition, multiple access techniques rely on timing as a crucial component to be able to leverage efficient use of resources while allowing multiple systems to transmit and receive. Military communications systems, such as Link-16, uses Time-Division Multiple Access (TDMA) to allow platforms to simultaneously communicate at the same time by allocating time slots to systems.

Types of Wired Cabling

Media

Diagram of EIA/TIA T568A Standards for Wiring an RJ-45
Ethernet Cable
.

Wires. One of the obvious signs of physical connectivity consists of the wires and cables used to transport data. Many organizations have to maintain and organize tens of thousands of cables, if not more, that feed data centers. Those cables likely originate from a desktop computer, plugged into a wall that goes to a data closet containing several network devices. The data closet, often referred to as an Intermediate Distribution Feed (IDF), may then go into a room within the building that is a central location that connects multiple floors and feeds into a data center somewhere on campus that connects multiple buildings. While most of the cables are likely copper, fiber can be used to increase bandwidth between buildings that require larger throughput. Before twisted-pair cable became popular (decades ago), you would find Ethernet primarily on coaxial cable in a bus topology. In the United States, cable television connects into homes typically using coaxial cables; some newer installations use fiber optic cables.

Signals in wired communications can take one of two forms. If in a conductive medium, like copper wires, the Physical layer encodes 0s and 1s as varying voltages. If in a fiber optic medium, the Physical Layer encodes 0s and 1s as variations in the wavelength, amplitude, and/or frequency of the light (usually visible or ultraviolet) waves. In both cases, attenuation (loss of signal strength as it travels through a medium) can be a concern for reliably receiving the transmitted signal.

Specifications for the use of Ethernet identify wiring diagrams and use of the four-pair when terminating RJ-45 standard connections as part of the Physical Layer. Over time, standards have improved data rates from original Ethernet (10Mbps) to Fast Ethernet (100Mbps) and more recently with advancements in Gigabit Ethernet (GbE) (1Gbps) up to speeds as fast as 100 GbE. This is possible through improvements and standards using copper cabling from Category (Cat) 4 types that supported Fast Ethernet to Cat8 that can go up to 40Gbps. The Content Addressable Memory (CAM) tables stored within network devices and communications using MAC addresses use Ethernet Framing standards for IP and Address Resolution Protocol (ARP) when interacting with the Networking Layer.

Wireless. Standards that identify short range wireless communications may be some of the familiar everyday technologies that are used, such as cellular, Near Field Communications (NFCs), Bluetooth, Wireless Fidelity (Wi-Fi), and Radio Frequency Identity (RFID) with other not-so-common ones like Z-Wave, ZigBee, Ultra Wide Band (UWB), and IPv6 over Low-Power Wireless Personal Area Networks (6LoWPAN). The Institute of Electrical and Electronics Engineers (IEEE) outlines standards for the development and specifications for manufactures to adhere to, publishing Wireless Personal Area Networks (WPANs) under the 802 series like Bluetooth (802.15.1), Wi-Fi (802.11), and ZigBee (802.15.4), while 6LoWPAN is published under the Internet Engineering Task Force (IETF) Request for Comment (RFC) 8025. You can thank the IEEE for the Bluetooth enabled headphones and speakers you're working out to in the gym or the track! Characteristics of short range wireless communications include low power output, use of Industrial, Scientific, and Medical (ISM) non-licensed frequency ranges, and compact components that are low cost and convenient for applications across household, commercial, and industrial use.

Traditional Long range wireless communication technologies consist of RF, microwave, and laser-based communications but the forefront of implementing a truly interconnected world is dependent on deploying end-to-end solutions to the masses. Long-Term Evolution (LTE) Machine (LTE-M) Type Communications, Narrowband Internet of Things (NB-IoT), Fifth Generation (5G) mobile network technology, and the Low Power, Wide Area (LPWA) networking protocols will enable endpoint nodes that will build and interconnect smart cities, connected health, smart metering, automotive transportation, vending monitoring, and environmental remote sensing applications at cost and scale.

Although many of the newly established standards for enabling Internet of Things (IoT) and mesh networks are important, the older infrastructure still provides vital services, such as the Global Positioning System (GPS) for timing, weather satellites for identifying and tracking natural disasters, Search and Rescue (SAR) communications, and other on-orbit sensors that provide for national security and defense. Threats in the space domain have been heightened with increased access to commercial and privatization of space launch capabilities, which is a great benefit for reducing cost and providing advancements in technological opportunities but has also created intentional and unintentional conflicts throughout the international community. The RF communications space systems are dependent on are increasingly vulnerable to interference and disruption, impacting availability to services leveraged by global economies.

Devices

Some of the most basic physical layer devices include repeaters, transceivers, media converters, and hubs.

Repeaters reconstruct, amplify, and rebroadcast electrical signals, which can attenuate (e.g., lose signal strength and fidelity) as they travel across long distances.
Transceivers are physical devices capable of receiving and transmitting a signal.
Media converters convert raw data between physical mediums, such as converting fiber optic communications (a light signal) into electrical signals.
Hubs are physical layer devices with multiple ports, and will redistribute a signal from one port to all other ports. While they have practical uses, by blindly retransmitting signals to all other ports, hubs have no confidentiality and are inefficient, reducing availability.

The Internet backbone consists of many devices that enable the reliable transmission of signals across physical territory. Signal loss and degradation due to environmental conditions, attenuation, thermal conditions generated from energy, and network distribution all contribute to impacts to the physical layer. Devices that reconstruct an electrical signal due to attenuation may include wired and wireless repeaters. For example, many consumer household products are available to extend WiFi signals throughout the home. For commercial use, 100 meters is the maximum distance for copper wiring and repeaters are necessary to extend signals.

Termination points within data closets may have to convert one medium to another using media converters and transceivers. Fiber optic cables allow long-distance communications, eliminating the need for repeaters because signal attenuation is not as impacted when using lasers or Light Emitting Diodes (LEDs); however, computer systems containing Integrated Circuits are made of metals and not fiber optics, therefore, transceivers need to convert a light signal to an electrical signal at the termination point. Transceivers are also used in wireless communications, containing both transmitter and receiver in a single device.

Expanding network connections can be possible with the use of hubs, which are physical layer devices that distribute signals across all other ports. Hubs will have a limited number of ports because of increased likelihood of collisions, which occur when two devices try to transmit at the same time on the same shared medium. For example, five devices connected through a five-port hub must take turns transmitting because they all share the same medium. There is no confidentiality, like a switch would provide, and network efficiency is bad compared to a switch that separates these collision domains.

Circuit Diagram of a Hub - Physical Layer. Network traffic is sent across all
ports, regardless of the destination address. There is no processor that makes
decisions on network traffic passing through hubs.
Circuit Diagram of a Switch - Data Link Layer. ASICs are responsible for
processing frames for a certain number of ports, allowing for the management
of layer 2 data and isolating network traffic through collision domains, allowing
hosts to only communicate with those identified as the destination address.

Operation Ivy Bells

Undersea cables have long since been a target for espionage with the use of wiretapping throughout the Cold War. The unprotected marine cables provided a gold mine of data, allowing the United States a significant information advantage over the USSR and targets of foreign intelligence from 1972 until its discovery in 1981.
A 2021 Atlantic Council report identified increasing risks to undersea cables’ security and resilience, focusing on authoritarian government's impact, especially in Beijing, to control Internet infrastructure and route data more favorably, potentially gaining better control of internet chokepoints and gaining espionage advantage; centralizing control over components and introducing new levels of operational security risk; and the increasing volume and sensitivity of cloud computing data crossing these cables. Concerns over undersea cable security have only grown since then.
Image courtesy of SIGINT CHATTER.

Physical Layer Cybersecurity Concerns

In theory, wired communications provide some inherent confidentiality because we can physically monitor who connects to a network. In practice, this can be difficult - consider both historical incidents and modern concerns over tapping submarine cables, which carry over 99% of intercontinental Internet traffic. Additionally, as mentioned earlier, some physical layer devices (hubs) have no inherent confidentiality.

Controlling who can receive a wireless signal is even harder. In some scenarios, particularly in military ones, techniques such as frequency hopping, spread spectrum communications, and directional antennas may reduce the chances an adversary intercepts a digital signal.

In both cases, confidentiality concerns can be mitigated with data encryption at higher levels of the TCP/IP Stack. In an everyday scenario, you use a password to connect to a home Wi-Fi router, and the password helps generate a key to encrypt the RF Wi-Fi communications. Though eavesdroppers may be able to detect the RF emissions, without the key, decrypting the conversation to make sense of the data becomes very difficult. Note: Wi-Fi security continues to evolve; the current standard, WiFi Protected Access 3 (WPA3), was developed in 2018 after previous standards were shown to be broken (WPA2 was broken in 2011, WPA in 2008, and Wired Equivalent Privacy (WEP) in 2001).

As for availability, physical infrastructure is susceptible to disruption, whether purposeful (sabotage) or accidental (such as a ship severing an underwater cable with its anchor…or even squirrels). With wireless communications, electronic warfare techniques (such as jamming and spoofing) can further impede availability and integrity).

Uplink and downlink jamming. Two forms of electronic counter space attack that targets the electromagnetic spectrum
through which space systems transmit and receive data (CSIS, 2023). Image courtesy of CSIS.

Cybersecurity physical requirements extend beyond the physical components that allow devices to operate and intercommunicate with other systems. This includes protection from the physical elements of the external environment, such as weather effects, natural disasters, and Ultra-Violet (UV) degradation from the sun's rays. Fencing, gates, vehicle barriers, and controls for people and wildlife need to be taken into consideration. To defend against many of the external elements, hardening of external structures is just as important when considering internal structures. Keeping the natural elements out requires the proper control of the environment inside, from Heating, Ventilation, and Air Conditioning (HVAC), to fire suppression, grounding, and more.

The 2016 United States Frequency Allocation Table. This chart depicts NTIA
frequency allocation from 0kHz to 300GHz.

Policy Considerations

Implementing the conceptual designs of the stack requires a consideration of government policy, even if the layers in the TCP/IP themselves never discuss this. A prime example in the physical layer concerns the division of the electromagnetic spectrum, which is used in wireless communications (see below). Without proper segmentation and regulation, wireless communications would likely suffer from significant interference, reducing availability and overall utility. In the U.S., frequency allocation is jointly regulated by the Federal Communications Commission (FCC) and the National Telecommunications and Information Administration (NTIA), while international wireless telecommunications - vital for space communications - are governed by the International Telecommunications Union (ITU). Additional policy considerations include the location of physical infrastructure, which may create geo-political tension or legal complexities during operations.

Concluding the Gmail Scenario

Everything we've covered so far in the Gmail scenario occurs inside devices. We've covered the how the Application Layer provides services, the Transport Layer connects these services end-to-end, the Network Layer provides connectivity from one network device to another across multiple network hops, and the Data Link layer connects devices within a local-area network (LAN). Now, we take one step deeper into the TCP/IP Stack by exploring how the Physical Layer conveys data.

Unlike the previous descriptions that have described each step at each layer in moderate detail, we take a different approach for the Physical Layer. First, know that encoding the 0s and 1s onto each of physical links in the diagram follow the conventions described above. Because we do not know the exact physical characteristics (and they may change occasionally) of the many devices in the paths shown, we can abstract and generalize them. (Remember, this is a benefit of the TCP/IP Stack, that each layer has some independence from the other layers.) Second, we follow the physical paths of the application-layer data rather than accounting for what happens at each step. Because our diagram is relatively wide, you may need to scroll to see its right end.

  1. The orange double arrow shows the path of your laptop's DNS query (#1) to the DNS server and the DNS server's DNS response (#6) back to your laptop.
  2. The purple arrow shows the path of your laptop's HTTPS request (#11) to the Gmail web server.
  3. Some important aspects to remember about how this all works:

Note that we've gotten to the point at which the Gmail web server has received our first HTTPS request. Now, it's the Gmail web server's turn to respond to your laptop's HTTPS request. Can you figure out how the HTTPS response gets from the Gmail web server's application layer to your laptop's application layer?

As you learned when building your own web sites, your browser will need to send an additional HTTPS request for each additional element, and for each of those additional elements that are not at mail.google.com, your laptop will need to send another DNS query to resolve each other server's IP address. Amazingly, all this happens in a fraction of a second!


Supplemental Media:

Cyber Squirrel

Gamification. Wreak havoc on the physical infrastructure as a squirrel!

How does the Internet work?


Review Questions:

  1. What are the different types of wired connections and wireless technologies used across network architectures?
  2. What are some of the critical aspects of undersea cables?
  3. What are some of the physical characteristics that are important for protecting computing networks and infrastructures?
  4. What are the advantages and disadvantages of different networking devices?
  5. How do all of the TCP/IP Stack layers work together?


References

  1. FCC. (2015). "FCC Circuit Status Report." [Online]. Available: https://www.fcc.gov/circuit-status-report