Comprehensive Overview of the TCP/IP Model and Core Network Protocols

1. TCP/IP Model

The TCP/IP protocol suite forms the foundation of the Internet, consisting of four layers: link, network, transport, and application. The model maps to the OSI model and defines how data is encapsulated and transmitted across networks.

Data moves through the layers by adding headers (and sometimes footers) at each step (encapsulation) and removing them at the receiver (decapsulation). This process is illustrated with an HTTP example.

2. Data Link Layer

The data link layer converts raw bit streams into frames, assigns MAC addresses, and provides error detection via CRC. Frames are built by adding source and destination MAC addresses, performing transparent transmission, and optionally ensuring reliable delivery on low‑error links.

Encapsulation into frames with MAC addresses.

Transparent transmission (bit stuffing, padding).

Reliability (rarely needed, used in WLAN).

CRC error detection.

3. Network Layer

The IP protocol is the core of the network layer. It routes packets using IP addresses, which are divided into classes (A, B, C) and consist of network and host portions. The IP header includes fields such as TTL, which limits the number of hops a packet can traverse.

4. ARP and RARP

ARP resolves an IP address to a MAC address by broadcasting a request; the host owning the IP replies with its MAC. RARP performs the opposite operation.

5. ICMP

ICMP provides network‑level error reporting and diagnostics (e.g., destination unreachable). It is used by tools such as ping to test connectivity.

6. Ping

The ping utility sends ICMP echo‑request packets and reports the echo‑reply, allowing users to verify host reachability and measure round‑trip time.

7. Traceroute

Traceroute discovers the path to a destination by sending packets with incrementally increasing TTL values, causing each router along the way to return an ICMP time‑exceeded message.

8. TCP/UDP

TCP and UDP are transport‑layer protocols. TCP provides reliable, connection‑oriented byte‑stream communication with flow and congestion control, while UDP offers connectionless, best‑effort datagram delivery for low‑latency applications.

9. DNS

The Domain Name System maps human‑readable domain names to IP addresses. DNS operates over UDP (port 53) and provides a distributed hierarchical database for name resolution.

10. TCP Connection Establishment and Termination

TCP connections are established via a three‑way handshake (SYN, SYN‑ACK, ACK) to synchronize sequence numbers and exchange window sizes. Termination uses a four‑step handshake (FIN, ACK, FIN, ACK) to ensure both sides close gracefully, with the TIME_WAIT state guaranteeing that delayed packets are discarded.

11. TCP Flow Control

Flow control prevents a fast sender from overwhelming a slow receiver by using a sliding window. The receiver advertises a window size (rwnd) that limits the amount of unacknowledged data the sender may transmit.

12. TCP Congestion Control

Congestion control adjusts the congestion window (cwnd) based on network conditions. The slow‑start algorithm increases cwnd exponentially until loss is detected, after which the algorithm switches to congestion avoidance (linear increase). The threshold (ssthresh) determines the transition point.

13. Fast Retransmit and Fast Recovery

When three duplicate ACKs are received, the fast‑retransmit algorithm immediately resends the missing segment without waiting for a timeout. Fast recovery then halves ssthresh, sets cwnd to ssthresh, and continues with congestion avoidance, avoiding a full slow‑start.

© Original author. Content reproduced with permission; all rights belong to the original source.