1. SS7

1a  Which layer(s) have error correction functionality? Explain briefly which errors 
are detected and how they are corrected. (3p)

The error detection is handled by the layer MTP-2, i.e. the link level protocol. A 
checksum is calculated for the entire packet. The packets are enumerated with sequence 
numbers, and the recepient acknowledges the recieved packets. If the packet  is lost, 
the  sender never recieves an acknowledgement and the data is retransmitted. 
If the checksum does not match the data, i.e. the data is corrupted, the recipient 
acknowledges the damaged packet, but sets BIB to indicate a negative acknowledgement 
(NAK). The sender retransmits the damaged packet and sets the FIB bit to indicate a 
retransmission.

1b Which layer handles the flow control? Explain briefly how it is implemented.(3p)

The flow control is handled by layer MTP-2. The sequence number (FSN) is a 7-bit number
and thus 128 packets can be sent before the sender has to wait for an acknowledgement. The
acknowledgement is done by copying the FSN of the latest correctly recieved packet to the
BSN field. The acnowledgements can be piggybacked TCP-style on packets containing data
(MSSU:s), but the FISU:s and LSSU:s carry acknowledgements as well.

1c Explain the addressing scheme used in the MTP-layers of SS7. Why is there no lack of 
addresses even though the address field is shorter than in IP? (3p)
 
In Europe the networks use ITU-T point codes as addresses. They are 14 bits long and are 
divided into three parts: Zone (3 bits), Network (8 bits) and signaling point 
identification number (3 bits). There is no lack of addresses since the addresses only
need to be unique within the network.

2. Structure of the telephone network (6p)

SSP1 -> SSP3: IAM
SSP3 -> SSP1: ACM
SSP3 -> SSP2: IAM
SSP2 -> SSP3: ACM
[The phone rings and B picks up the phone]
SSP2 -> SSP1: ANM

IAM: Initial Address Message -- reserves a trunk between two SSP:s.
ACM: Address Complete Message -- indicates that a trunk is reserved
ANM: ANswer Message -- indicates that the recepient picks up the phone.

3. Data transmission technologies (5b)

3a: Explain the major differences between negative and positive byte justification 
in SDH and Sonet. (2p)

Negative byte justification is needed if the sender transmits data too fast. The incoming
data is buffered by the recipient, and the buffer is cleared when the sender 
adds an extra SPE byte from time to time. The SPE pointer is updated in subsequent frames
to indicate that the start of the SPE has moved one byte forward.

Positive byte justification is needed if the incoming stream runs too slow. The sender
adds extra dummy bytes as "stuffing", and updates the SPE pointer to show that the next
SPE starts a few bytes later.

3b: Can an ADSL modem change its transmission rate if the properties of the transmission 
path (copper pair) change?

To be completed later.