BSc CSIT (TU) Science Computer Networks (BSc CSIT, CSC258) Question Paper 2081 Nepal

Subnetting a Class B Network into 4 Subnets

Given: A Class B network. Take the example network 172.16.0.0 with the default mask 255.255.0.0 (/16).

Step 1: Number of subnet bits

To create 4 subnets we need $n$ borrowed bits such that $2^n \ge 4$.

Step 2: New subnet mask

Class B default is /16. Borrowing 2 bits gives /18.

The third octet borrows the top 2 bits: $11000000_2 = 192$.

Step 3: Block size

So subnets increment by 64 in the third octet.

Step 4: Subnet ID and Broadcast addresses

Step 5: Hosts per subnet

Host bits = $32 - 18 = 14$.

Result: New subnet mask = 255.255.192.0 (/18), with the four subnets, subnet IDs and broadcast addresses listed above.

TCP (Transmission Control Protocol)

TCP is a connection-oriented, reliable, byte-stream transport-layer protocol of the TCP/IP suite. It provides reliable, in-order, error-checked delivery of data between processes using sequence numbers, acknowledgements, retransmission, flow control (sliding window) and congestion control. It is identified by IP protocol number 6.

TCP Segment Structure

The TCP header is a minimum of 20 bytes (without options). Key fields:

Three-Way Handshake (Connection Establishment)

Client                          Server
  |------- SYN (seq=x) -------->|
  |<-- SYN+ACK (seq=y,ack=x+1)--|
  |------- ACK (ack=y+1) ------>|
   Connection ESTABLISHED

1. SYN: Client sends a segment with SYN=1 and an initial sequence number $x$.
2. SYN+ACK: Server replies with SYN=1, ACK=1, its own sequence $y$, and acknowledgement $x+1$.
3. ACK: Client sends ACK=1 with acknowledgement $y+1$. The connection is now open and data transfer begins.

Connection Termination (Four-Way Handshake)

TCP uses a graceful, four-way close:

Client                          Server
  |------- FIN (seq=u) -------->|
  |<------ ACK (ack=u+1) -------|
  |<------ FIN (seq=v) ---------|
  |------- ACK (ack=v+1) ------>|
   Connection CLOSED

1. Client sends FIN to close its side.
2. Server replies with ACK.
3. Server sends its own FIN when ready.
4. Client replies with ACK (then waits in TIME_WAIT before fully closing).

This ensures both directions of the full-duplex connection are closed independently and reliably.

Error Detection and Correction in the Data Link Layer

Transmission over physical media introduces errors (single-bit or burst errors). The data link layer uses redundant bits to detect and sometimes correct them.

A. Error Detection Techniques

1. Parity Check – One redundant bit makes the number of 1s even (even parity) or odd. Detects single-bit errors only; fails on even-numbered bit errors.
2. Two-Dimensional (LRC) Parity – Parity computed per row and column; detects most burst errors.
3. Checksum – Data divided into $k$-bit words, summed using 1's-complement arithmetic; the complemented sum is sent. Receiver re-adds; result of all 1s means no error.
4. Cyclic Redundancy Check (CRC) – Strong polynomial-based detection (below).

B. CRC (Cyclic Redundancy Check) – Example

CRC treats the bit string as a polynomial and divides by a generator using modulo-2 (XOR) division.

Data = 1010, Generator $G = 1011$ (degree 3, so append 3 zeros).

Dividend = 1010000. Perform modulo-2 division by 1011:

1010000 ÷ 1011
1010 XOR 1011 = 0001
 bring down -> 00010 ...
Result: quotient ignored, Remainder (CRC) = 011

Transmitted frame = data + remainder = 1010 + 011 = 1010011.

At the receiver, the frame is divided by $G$ again; a zero remainder means no error, otherwise an error is detected. CRC detects all single-bit, double-bit, odd-numbered, and most burst errors.

C. Hamming Code (Error Correction) – Example

Hamming code adds parity bits at positions $2^0, 2^1, 2^2, \dots$ to detect and correct single-bit errors. For $m$ data bits we need $r$ parity bits where:

Example: 4 data bits 1011 (positions for d). $2^r \ge 4 + r + 1 \Rightarrow r = 3$. Total 7 bits.

Layout (positions 1–7), parity at 1,2,4; data at 3,5,6,7 = 1,0,1,1:

• $P1$ covers bits 1,3,5,7 → 1,0,1 → for even parity $P1 = 0$
• $P2$ covers bits 2,3,6,7 → 1,1,1 → $P2 = 1$
• $P4$ covers bits 4,5,6,7 → 0,1,1 → $P4 = 0$

Codeword = 0110011 (positions 1–7).

Correction: At the receiver, recompute the three parity checks; the binary value of the failing parity positions gives the position of the erroneous bit, which is then flipped to correct it.

Summary

• Detection: Parity, 2-D parity, Checksum, CRC.
• Correction: Hamming code (and FEC codes) correct single-bit errors using the error-position syndrome.

Connection-Oriented vs Connectionless Services

Summary: Connection-oriented service is like a telephone call — a dedicated logical connection is set up, used, then released, with guaranteed reliable in-order delivery. Connectionless service is like the postal system — each datagram is independent, requires no setup, and is delivered on a best-effort basis with lower overhead.

UDP (User Datagram Protocol)

UDP is a connectionless, unreliable transport-layer protocol (IP protocol number 17). It provides no handshake, no acknowledgement, no retransmission, no flow/congestion control and no guaranteed ordering. It simply adds port-based multiplexing and an optional checksum on top of IP, making it fast and low-overhead. It suits real-time and query/response applications such as DNS, DHCP, TFTP, SNMP, VoIP, online gaming and video streaming.

UDP Header Format

The UDP header is a fixed 8 bytes (64 bits) with four 16-bit fields:

 0                   16                  31
 +--------------------+--------------------+
 |    Source Port     |  Destination Port  |
 +--------------------+--------------------+
 |      Length        |     Checksum       |
 +--------------------+--------------------+
 |             Data (payload)              |
 +-----------------------------------------+

Key point: Because the header is only 8 bytes versus TCP's 20 bytes, UDP has much lower overhead but offers no reliability guarantees.

CIDR (Classless Inter-Domain Routing)

CIDR is an IP addressing and routing scheme (RFC 1519) that replaces the rigid Class A/B/C boundaries with a variable-length prefix. An address is written as IP/prefix-length (slash notation), where the prefix length states how many leading bits form the network portion. CIDR enables efficient use of the IPv4 address space and route aggregation (supernetting), which shrinks routing tables.

Notation

where $n$ = number of network bits, so host bits $= 32 - n$ and usable hosts $= 2^{32-n} - 2$.

Example

Consider 192.168.1.0/26:

• Prefix length $n = 26 \Rightarrow$ subnet mask = 255.255.255.192
• Host bits $= 32 - 26 = 6$
• Usable hosts $= 2^6 - 2 = 62$
• Network address: 192.168.1.0, Broadcast: 192.168.1.63
• Host range: 192.168.1.1 – 192.168.1.62

Supernetting Example

Four networks 200.1.0.0/24, 200.1.1.0/24, 200.1.2.0/24, 200.1.3.0/24 can be aggregated into a single CIDR route 200.1.0.0/22, advertised as one entry instead of four.

Advantages: flexible subnet sizing, reduced address wastage, smaller routing tables through aggregation.

Hub vs Switch vs Router

Summary:

• A hub is a dumb Layer-1 repeater that floods incoming bits to every port (one collision domain).
• A switch is an intelligent Layer-2 device that learns MAC addresses and forwards frames only to the destination port (separate collision domain per port).
• A router is a Layer-3 device that forwards packets between different IP networks using routing tables and separates broadcast domains, connecting LANs to the Internet.

ARP (Address Resolution Protocol)

ARP is a protocol that maps a known logical (IP) address to its corresponding physical (MAC) address on a local network. It operates between the network and data-link layers, because frames on a LAN are delivered using MAC addresses, while hosts are addressed logically by IP.

How ARP Resolves IP to MAC

1. Check ARP cache: When host A wants to send a packet to IP address $B$ on the same LAN, it first checks its ARP cache (table of IP→MAC mappings). If found, it uses it directly.
2. ARP Request (broadcast): If not cached, A broadcasts an ARP Request frame to the MAC broadcast address FF:FF:FF:FF:FF:FF, asking "Who has IP $B$? Tell A." This reaches every host on the LAN.
3. ARP Reply (unicast): Only the host owning IP $B$ responds with an ARP Reply (unicast back to A) containing its MAC address.
4. Update cache: Host A stores the IP→MAC mapping in its ARP cache (with a timeout) and uses the MAC address to build and send the frame.

A --- broadcast ---> "Who has 192.168.1.5?"  (ARP Request)
B --- unicast -----> "192.168.1.5 is at 00:1A:2B:3C:4D:5E" (ARP Reply)

Related: RARP / DHCP do the reverse (MAC→IP). The cache avoids repeating ARP for every packet. If the destination is on a different network, ARP resolves the default gateway's IP instead.

HTTP (HyperText Transfer Protocol)

HTTP is the application-layer, request-response protocol used by the World Wide Web to transfer hypertext (web pages, images, etc.) between a client (browser) and a web server. It runs over TCP, typically on port 80 (HTTPS uses TLS on port 443), and is stateless — each request is independent (state is maintained via cookies/sessions).

Working of HTTP

1. Connection: The client opens a TCP connection to the server (three-way handshake) on port 80.
2. Request: The browser sends an HTTP request consisting of:
   • a request line — method + URL + version, e.g. GET /index.html HTTP/1.1
   • headers (Host, User-Agent, Accept, etc.)
   • an optional body (for POST/PUT).
3. Server processing: The server locates/generates the requested resource.
4. Response: The server returns an HTTP response with:
   • a status line — version + status code, e.g. HTTP/1.1 200 OK
   • headers (Content-Type, Content-Length, Date)
   • the body (the requested HTML/data).
5. Render/Close: The browser renders the content. In HTTP/1.0 the connection closes per request; HTTP/1.1 uses persistent (keep-alive) connections for multiple requests.

Common Methods & Status Codes

• Methods: GET, POST, PUT, DELETE, HEAD.
• Status codes: 200 OK, 301 Moved, 404 Not Found, 500 Internal Server Error.

Client: GET /index.html HTTP/1.1
        Host: www.example.com
Server: HTTP/1.1 200 OK
        Content-Type: text/html
        <html>...</html>

NAT (Network Address Translation)

NAT is a technique implemented on a router/firewall that translates private IP addresses (used inside a LAN) into one or more public IP addresses (used on the Internet), and vice versa. It conserves the limited IPv4 public address space and hides the internal network structure, adding a layer of security.

Types of NAT

1. Static NAT (one-to-one): A fixed, permanent mapping of one private IP to one public IP. Used when an internal server must be reachable from outside.
2. Dynamic NAT: Maps private IPs to public IPs from a pool on a first-come, first-served basis. Mappings are temporary; the number of simultaneous users is limited by the pool size.
3. PAT / NAT Overload (many-to-one): Many private IPs share one public IP, distinguished by different port numbers. This is the most common form (e.g., home routers).

Uses / Advantages

• Conserves public IPv4 addresses — many hosts share a few public IPs.
• Security/privacy — internal addresses are hidden from the outside.
• Flexibility — internal addressing can change without affecting public addresses.
• Enables multiple devices on a home/office LAN to access the Internet through a single public IP.

Disadvantage: Breaks end-to-end connectivity and complicates some protocols (e.g., requires ALGs for FTP/SIP).

FTP (File Transfer Protocol)

FTP is an application-layer, client-server protocol used to transfer files between a client and a server over a TCP/IP network. It runs over TCP and is reliable. A distinctive feature is that FTP uses two separate connections:

• Control connection (port 21): carries commands and replies; stays open for the whole session.
• Data connection (port 20 in active mode): carries the actual file data; opened/closed for each transfer.

Working of FTP

1. Connection & login: The client opens a control connection to the server on port 21 and authenticates with a username and password (anonymous FTP allows guest login).
2. Commands: Over the control channel the client sends commands such as USER, PASS, LIST, RETR (download), STOR (upload), CWD, QUIT.
3. Data transfer modes:
   • Active mode: server initiates the data connection from port 20 to a client-specified port.
   • Passive mode: server opens a port and the client initiates the data connection (firewall-friendly).
4. File transfer: Data is sent over the separate data connection in ASCII or binary mode; the control connection remains open to issue further commands.
5. Termination: The client sends QUIT and the connections close.

Client --(control, port 21)--> Server : USER, PASS, RETR file.txt
Client <==(data, port 20)===== Server : file contents

Note: FTP transmits credentials in plaintext, so secure variants FTPS (FTP over TLS) and SFTP (over SSH) are preferred for security.

ICMP (Internet Control Message Protocol)

ICMP is a network-layer protocol of the TCP/IP suite (IP protocol number 1) used to send error-reporting and control/diagnostic messages about the operation of IP. Because IP is unreliable and connectionless, ICMP feeds back information when packets cannot be delivered or when network conditions need to be reported. ICMP messages are encapsulated inside IP datagrams.

Purpose

• Report errors (e.g., destination unreachable, time exceeded).
• Provide diagnostics and reachability testing — used by tools like ping (Echo) and traceroute (Time Exceeded).
• Flow/redirect control between routers and hosts.

Common ICMP Message Types

In short: ICMP is the troubleshooting and error-signalling companion of IP; without it, hosts would have no standard way to learn why a packet failed to reach its destination.