Question $2$

$($a$)$ Consider transmission of two control protocol messages: the Internet Control Message

Protocol $($ICMP$)$ and the Address Resolution Protocol $($ARP$)$ message. The formats of

these messages are shown in Figure $2\mathrm{.}1$ below:

Figure $2\mathrm{.}1$

$($i$)$ Which layer is responsible for processing these messages? Does this mean that

ICMP and ARP provide service to this layer, i$.$e$.$ that they belong to the lower layer?

$($ii$)$ Explain how the ICMP and ARP messages are transmitted. What is similar and what

is different for each of the messages?

$($iii$)$ Consider two messages: $1.$ An ICMP message of a minimum possible size sent by a

host. $2.$ An ARP reply message sent by a gateway router. Messages are transmitted

by a classical Ethernet. Determine the size of the Ethernet frame in each case.

$($b$)$ Four networks N$1$ N$4,$ which use Classless Inter$-$Domain Routing $($CIDR$)$ addressing

protocol, have the following IP base addresses for each network, respectively:

N$1-75\mathrm{.}16\mathrm{.}96\mathrm{.}0/21$

N$2-75\mathrm{.}16\mathrm{.}104\mathrm{.}0/21$

N$3-75\mathrm{.}16\mathrm{.}112\mathrm{.}0/21$

N$4-75\mathrm{.}16\mathrm{.}120\mathrm{.}0/21$

The index $/21$ indicates the length of the corresponding mask.estion $2($continued$)$

$($i$)$ If all these networks are connected to the same router, can this router advertise

outside to the other routers the aggregated address? If so$,$ derive the aggregated

address and the corresponding mask. If no$,$ explain why.

$($ii$)$ The address aggregation technique helps routers to speed up processing of the

received packets. Explain why the aggregation speeds up processing of the received

packets.

$($iii$)$ If a packet with the IP destination address IPX $=75\mathrm{.}16\mathrm{.}125\mathrm{.}15$ arrives at the router, to

which of the four networks will the packet be routed? Explain in detail why.

$($iv$)$ Determine the range of the IP addresses in each of the networks, i$.$e$.$ find the smallest

and the biggest address in each network domain.

$($c$)$ Suppose that host A communicates via router R with hosts B and C in the IPv$4$ network

shown in Figure $2\mathrm{.}2.$ Suppose that a TCP message that contains $1200$ bytes of data and $20$

bytes of TCP header is passed to the network layer of the host A$.$ Host A sends two

separate IP packets for delivery to B and C$,$ respectively. The link R $-$ B has MTU$1=600$

bytes, the link R $-$ C has MTU$2=1220$ bytes.

Figure $2\mathrm{.}2$

$($i$)$ Is fragmentation of the packet required for each transmission? Explain your answer.

$($ii$)$ In case if either or both of the packets require fragmentation, in each case, calculate:

the payload, the value of fragment offset, and show the value of the M$-$bit for all

fragments.

$($iii$)$ If the TCP message in host A contained $530$ bytes of data and $20$ bytes of TCP

header, would the IP packet be fragmented in any IPv$4$ network with MTUs which

are different from the MTUs we use in this problem? Explain your answer.