Network Layer Study Guide - OSI Model | Electrical Engineering

Primary Functions

•Logical Addressing: Assigns IP addresses to identify devices across networks
•Routing: Determines optimal paths for packet delivery between different networks
•Packet Forwarding: Moves packets from input to appropriate output interface
•Fragmentation & Reassembly: Breaks large packets to fit MTU limits

Key Characteristics

PDU (Protocol Data Unit) Packet

Data encapsulated with IP header containing source/destination addresses

Addressing Type Logical (IP)

Hierarchical addressing scheme (Network + Host portions)

Connection Type Connectionless

Each packet routed independently (best-effort delivery)

Network Layer in OSI Context

Application

Data

Layer 7

Transport

Segment

Layer 4

Network

PACKET (IP Header + Data)

Layer 3

Data Link

Frame

Layer 2

Physical

Bits

Layer 1

The Network Layer encapsulates segments from Layer 4 into packets by adding IP headers with logical addressing information

End-to-End Data Flow Through Network Layer

📱

Source Host

Creates packet with destination IP address. Uses ARP to find next-hop MAC address.

Routed Path

↓

🖥️

Destination Host

Receives packet, decapsulates IP header, delivers payload to Transport Layer.

4 IPv4 Address Structure

32-bit Address Format:

■ Network Portion (24 bits) | ■ Host Portion (8 bits)

Notation Dotted Decimal (e.g., 192.168.1.1)

Total Addresses 2³² ≈ 4.3 billion

Classes A (1-126), B (128-191), C (192-223)

Electrical Engineering Note: IPv4 addresses are processed by digital logic in routers using TCAM (Ternary Content Addressable Memory) for high-speed lookups.

6 IPv6 Address Structure

128-bit Address Format:

2001:0db8:85a3:0000:0000:8a2e:0370:7334

■ Global Routing Prefix (48 bits) | ■ Subnet ID (16 bits) | ■ Interface ID (64 bits)

Notation Hexadecimal Colon-separated

Total Addresses 2¹²⁸ ≈ 3.4×10³⁸

Features Auto-configuration, Built-in Security

Advantage: 128-bit addressing eliminates NAT requirements and provides end-to-end connectivity for IoT devices in electrical engineering applications.

IPv4 Address Classes & Subnet Masks

Class	First Octet Range	Default Mask	Networks	Hosts/Net	Use Case
A	1 - 126	255.0.0.0 (/8)	126	16,777,214	Large networks, ISPs
B	128 - 191	255.255.0.0 (/16)	16,384	65,534	Medium-sized organizations
C	192 - 223	255.255.255.0 (/24)	2,097,152	254	Small networks, LANs

CIDR Notation: Classless Inter-Domain Routing (e.g., /24) allows flexible subnet masking beyond traditional classes, optimizing address space allocation in modern networks.

Private Addresses (RFC 1918)

10.0.0.0/8
172.16.0.0/12
192.168.0.0/16

Non-routable on public internet; used for internal networks

Loopback

127.0.0.0/8

Used for local testing and IPC. Packets never leave the host (127.0.0.1 is localhost).

Multicast

224.0.0.0/4

One-to-many communication. Used for streaming, routing protocols (OSPF, RIP).

Routing Algorithm Classification

DV

Distance Vector

▸Bellman-Ford Algorithm
▸Shares entire routing table with neighbors only
▸Metric: Hop count (RIP) or Composite (EIGRP)
▸Slow convergence (Count-to-infinity problem)
▸Low bandwidth, CPU overhead

Examples: RIP, IGRP, EIGRP
Best for: Small networks, limited bandwidth

LS

Link State

▸Dijkstra's Algorithm (SPF)
▸Floods Link State Advertisements (LSAs) to all routers
▸Builds complete topology map (LSDB)
▸Fast convergence, loop-free
▸Higher memory, CPU requirements

Examples: OSPF, IS-IS
Best for: Large enterprise networks, fast convergence needs

Feature	Distance Vector (RIP)	Link State (OSPF)
Algorithm	Bellman-Ford	Dijkstra (Shortest Path First)
Knowledge Scope	Local (neighbors only)	Global (entire network topology)
Update Mechanism	Periodic (every 30s in RIP)	Triggered (on topology change)
Convergence Time	Slow (minutes)	Fast (seconds)
Scalability	Limited (15 hop max in RIP)	Highly scalable (hierarchical areas)
Resource Usage	Low CPU, Low Memory	High CPU, High Memory

Router Forwarding Decision

Routing Table Structure

$ show ip route

Destination	Mask	Next Hop	Interface	Metric
192.168.1.0	255.255.255.0	Connected	Eth0	0
10.0.0.0	255.0.0.0	192.168.1.1	Eth0	1
0.0.0.0	0.0.0.0	192.168.1.254	Eth0	1

Directly Connected: Local network attached to interface

Remote Route: Learned via routing protocol or static config

Default Route (0.0.0.0): Gateway of last resort

Longest Prefix Match

When multiple routes match a destination, the router selects the entry with the longest subnet mask (most specific match).

Destination: 10.1.1.5

10.1.1.0/24 → Eth1 SELECTED (/24)

10.1.0.0/16 → Eth0 Less specific (/16)

0.0.0.0/0 → Eth2 Default route

Routing Protocol Operation Visualization

Link State Operation: Each router floods LSAs to build identical topology databases, then independently runs Dijkstra's algorithm to compute shortest paths.

IP

Internet Protocol

Connectionless, unreliable delivery

IPv4 Header Format (20-60 bytes):

Version (4)

IHL (4)

TOS/DSCP (8)

Total Length (16)

Identification (16)

Flags (3)

Fragment Offset (13)

TTL (8)

Protocol (8)

Header Checksum (16)

Source IP Address (32)

Destination IP Address (32)

Options + Padding (0-40 bytes)

•TTL (Time To Live): Prevents infinite loops; decremented at each hop
•Protocol Field: Identifies upper layer (6=TCP, 17=UDP, 1=ICMP)
•Fragmentation: Handles MTU differences across networks

!

ICMP

Internet Control Message Protocol

Used for error reporting and diagnostic functions. ICMP messages are encapsulated in IP datagrams (Protocol = 1).

Type 8 / 0

Echo Request / Reply

PING

Type 3

Destination Unreachable

Error

Type 11

Time Exceeded (TTL=0)

TRACEROUTE

Type 5

Redirect

Routing

Address Resolution Protocol (ARP)

ARP bridges Layer 3 (Network) and Layer 2 (Data Link) by mapping IP addresses to MAC addresses on local networks.

1

Host A wants to send to 192.168.1.10

Checks ARP cache - no entry found

2

Broadcasts ARP Request

"Who has 192.168.1.10? Tell 00:1A:2B:3C:4D:5E"

3

Target Host Responds

"192.168.1.10 is at 00:5E:4D:3C:2B:1A"

4

Cache Entry Created

Entry stored for 4-20 minutes (configurable)

$ arp -a

? (192.168.1.1) at 00:50:56:c0:00:08 [ether] on eth0

? (192.168.1.10) at 00:5e:4d:3c:2b:1a [ether] on eth0

? (192.168.1.15) at 00:1a:2b:3c:4d:5e [ether] on eth0

ARP Packet Structure:

Hardware Type: Ethernet (1)

Protocol Type: IPv4 (0x0800)

Operation: Request (1) / Reply (2)

Sender MAC/IP → Target MAC/IP

Network Layer Protocol Stack

Application Layer

HTTP, FTP, DNS, SMTP

↓

Transport Layer

TCP, UDP (Segments)

↓

NETWORK LAYER

IP

Addressing, Routing

ICMP

Error Reporting

ARP

Address Resolution

Packets with IP Headers

↓

Data Link Layer

Ethernet, Wi-Fi (Frames)

🔄

Routers

Layer 3 devices that forward packets between networks based on IP addresses.

• Maintain routing tables
• Run routing protocols (OSPF, BGP)
• Perform packet switching
• Implement ACLs/firewall rules

⚡

Layer 3 Switches

Hardware that combines switching speed with routing intelligence.

• ASIC-based packet forwarding
• VLAN routing (Inter-VLAN)
• Wire-speed performance
• Lower latency than routers

🛡️

Firewalls

Security appliances operating at Layer 3/4 filtering traffic.

• Stateful packet inspection
• NAT/PAT implementation
• VPN termination (IPsec)
• DoS protection

Router Internal Architecture

Functional Components

1

Input Ports

Physical layer termination, data link processing, lookup/determine output port using forwarding table

2

Switching Fabric

Connects input to output ports. Types: Memory, Bus, Crossbar. Determines router capacity

3

Output Ports

Queueing (buffer management), data link processing, physical layer transmission

4

Routing Processor

Executes routing protocols, maintains routing tables, calculates forwarding tables

Hardware Implementation

TCAM (Ternary CAM)

Specialized memory for longest prefix matching in IP lookups. Stores 0, 1, and X (don't care) states.

ASICs

Application-Specific Integrated Circuits for wire-speed forwarding at line rate (Gbps/Tbps).

NPUs

Network Processing Units - programmable processors optimized for packet processing tasks.

FPGAs

Field Programmable Gate Arrays for reconfigurable high-speed packet processing logic.

Packet Processing Pipeline

1. Line Termination

PHY layer reception
Bit synchronization

→

↓

2. Data Link

MAC processing
Frame check (CRC)

→

↓

3. Network Layer

IP header validation
TTL decrement
Checksum recalc
Route lookup (TCAM)

→

↓

4. Queueing

Buffer management
Scheduling (QoS)

→

↓

5. Transmission

MAC framing
Physical transmission

Electrical Engineering Perspective: Modern routers implement the forwarding plane in hardware (ASICs/FPGAs) to achieve terabit-per-second throughput, while the control plane (routing protocols) runs on general-purpose CPUs. This separation is crucial for meeting strict latency requirements in carrier-grade networks.

Interactive IPv4 Subnet Calculator

IP Address CIDR Prefix (/n) or Subnet Mask

Calculation Results:

Network Address:192.168.1.0

Broadcast Address:192.168.1.255

Usable Host Range:192.168.1.1 - 192.168.1.254

Total Hosts:256

Usable Hosts:254

Subnet Mask:255.255.255.0

Wildcard Mask:0.0.0.255

Binary Network:11000000.10101000.00000001.00000000

Visual Binary Representation

IP:

Mask:

Network Portion (24 bits) Host Portion (8 bits)

Formula Reference:

• Hosts per subnet = 2^(32 - CIDR) - 2
• Subnets = 2^(CIDR - DefaultClassMask)
• Block size = 256 - MaskOctetValue

Quick Tips:

• /30 gives 2 usable hosts (point-to-point links)
• /24 is standard for Class C networks (254 hosts)
• Network address has all host bits = 0
• Broadcast address has all host bits = 1