Data Communications Flashcards
Name some examples of transmission media.
- Wire (copper)
- Glass fibre
- Infrared
- Laser
- EM radiation
What are some pros and cons of copper wire as a transmission media?
- common, inexpensive and easy to install
- risk of interference
How does optical fibre work as a transmission media?
- thin glass tube coated in plastic
- reflects light internally
- transmitter uses LED or laser and receiver uses a light sensitive transistor
What are the pros and cons of Optical fibre as a transmission media?
- no electrical interference
- signal can be carried further than by copper wire
- can transfer more information at one time than copper
- BUT hard to find breaks
What is a main example of infrared as a transmission media?
remote controls
What are some pros and cons of infrared as a transmission media?
- portable and needs no antennae
- cheap
BUT - limited range
- sensitive to orientation of transmitter to receiver and being blocked
How does use of EM signals to transmit data vary between microwave and Radio?
Radio:
- small computer antennae
- can penetrate walls
Microwave:
- higher frequency
- carries more info than RF
- cannot penetrate metal so needs very tall towers and direct line of sight
What are six categories of trade-offs between media transmission types?
Cost
Data rate
Delay
Affect on signal
Environment interaction
Security
What does a ‘communication system’ do?
Accepts input from one or more sources and delivers information from a given source to a specified destination
What’s the difference between analogue and digital signals?
Analogue:
changes from one value to another via all possible intermediate levels (curvy)
Digital:
has fixed valid levels (straight flat lines with jumps between)
Most signals are ____ signals
composite (can be decomposed into a set of simple sine waves)
Why are signals typically modulated? What does this mean?
Signals are modulated so data can be carried by a carrier wave.
What energy type do each of these methods use?
- Wire (copper)
- Glass fibre
- Infrared
- Laser
- Radio / satellite
Copper wire: electrical
Glass fibre (fibre optic): light
Infrared: light
Laser: light
Radio / satellite : EM
What are the 3 categories of error?
Interference
Distortion
Attenuation
What is a distortion error?
- Physical systems distorting signals
- wires have capacitance and inductance and will block signals at some frequencies
What is an attenuation error?
Weakening of signal with distance
What is a single bit error?
A single bit in a block of bits is changed - normally due to short duration interference
What is a burst error?
Multiple bits in a block are changed
What is an erasure error?
The the signal that arrives at the receiver is to ambiguous to be a 1 or 0
- can be caused by distortion or interference
What are the trade-off considerations for error detection and correction?
- Error detection adds overhead cost
- single bit error is hugely important for some transmissions (bank transfer) but not others (pictures)
What is forward error correction (FEC)?
Adding additional information to the data to allow the receiver to check if it’s transmitted correctly
What are some examples of FEC (forward error correction)?
Single parity bit checking
Row and column parity
Checksums
What are Automatic Repeat reQuest mechanisms (ARQ)?
Using acknowledgment messages to check whether data has arrived at its destination
What is single parity bit checking?
The number of 1s is counted and an extra ‘parity’ bit is added to make this even or odd
- check at the other end that it’s still either even or odd as intended
What are some limitations of single parity bit checking?
- introduces additional costs
- only detects limited types of errors
- cannot be used to correct errors
What is row and column parity checking?
An even or odd parity bit is added to each row and column so that locations of errors can be identified and potentially resolved
What are some pros and cons of row and column parity checking?
- identifies exact error location
- receiver can correct a single bit error
But can only correct single-bit errors (can detect an odd number of errors if more than one bit changed)
What are some factors used to compare error detection/correction methods?
- amount of extra data needed to be corrected
- amount of extra computation needed
- types of error detected
- whether detected errors can be fixed
What are checksums?
Data is interpreted as a sequence of integers and added together to get a ‘checksum’
What are cyclic redundancy checks (CRCs)? What are they especially good for?
Bits are cycled through registers several times in several places
- detects more errors than checksums and especially good for burst errors
What are ARQs good for?
When an underlying system detects error but doesn’t correct it. Can guarantee delivery
What does resistance in wires cause (to data)?
Signal loss
What type of signal will propagate further than other signals?
A continuous oscillating signal
What type of signals is the term modulation used for? What is the equivalent for the other type of signal?
Analogue signals
- called shift keying for digital signals
What are the two types of modulation?
Frequency modulation (FM)
Amplitude modulation (AM)
What is modulation in a general sense?
An oscillating carrier wave is sent and modulated in some way - the receiver picks this up and discards the carrier (which is known).
What is shift keying?
The modulation of a carrier wave to encode a digital signal
What is amplitude modulation?
The amplitude of a carrier wave is adjusted to match the amplitude of a wave to be transmitted
What are ASK and FSK?
Amplitude shift keying
Frequency shift keying
Why do we never reach the theoretical limit of max data speed?
because of noise (caused by interference)
What is Nyquist’s theorem?
Suggests if you are more clever in encoding data you can send more data per unit time e.g. changing phase of a signal
What is PSK?
Phase shift keying
- timing of the carrier is changed so several bits can be sent per cycle
How many wave cycles are needed to send a bit using AM and FM?
at least one
How does PSK carry more bits?
Each phase shift represents more than one bit.
e.g. 00 - no shift
01 - 1/4
10 - 1/2
11 - 3/4
What is QAM?
phase and amplitude are changed to encode even more data
What is a modem?
A combined modulator-demodulator
(can apply modulation and extract bits from a modulated wave)
What is a dial-up modem? What method is used to increase data rate?
A modem using audio tone on a telephone
Uses QAM to increase data rate
When are optical modems vs RF (radio frequency modems) used?
optical: for optical fibres
RF: for radio e.g. wifi networks
What is multiplexing? What is demultiplexing?
multiplexing: combining information streams from several sources to transmit on a shared medium
demultiplexing: separating the combined stream
What are the 4 types of multiplexing?
FDM (frequency division)
- widely used
WDM (wavelength division)
- form of FDM
TDM (time division)
- uses time slots
CDM (code division)
- mathematical approach
When is WD multiplexing used?
With optical fibres
What are the requirements to use FDM?
Requires high bandwidth and coaxial cables as it’s sensitive to interference
Using FDM, data is sent on _______ frequencies
several
Explain WDM.
Application of FDM to optical fibre
prisms are used to combine beams of light of different wavelengths into a single beam
What is TDM?
Alternate to FDM where sources sharing the medium take turns
What differentiates statistical TDM?
A source with nothing to send is skipped
What is the most important basic component of a WAN? What does it do?
Packet switches that connect local computers (LANs) to other computers and other packet switches
Briefly describe the concept of store and forward.
Each packet switch in a WAN receives packets, queues them in its memory and sends out when possible (available destination)
How does addressing work in a WAN?
A WAN defines it’s own frame format and assigns an address to each computer
How does hierarchical addressing work in a WAN ?
The first part identifies which packet switch the computer is connected to and the second part defines the actual computer
Explain next hop forwarding in a WAN.
Each packet only has the information for the next packet switch it can send to, so the frame hops from packet to packet til it reaches the packet its destination is connected to.
e.g. london to NY to Dallas to San Francisco to computer
What does source independence mean in a WAN?
The next hop in a forwarding path only depends on the destination not on source address
What is routing (in WAN context)?
Routing is the process of forwarding a packet to its next hop
What must WAN routing tables guarantee?
- Universal routing (next hop for every possible destination)
- Optimal routes - next hop must point to the shortest path to the destination
What are the two types of routing for a WAN?
Static routing - program computes and installs routes when a switch boots and they don’t change
Dynamic routing - program builds an internal table which alters as condition changes
Describe static routing.
program computes and installs routes when a switch boots and they don’t change
Describe dynamic routing.
program builds an internal table which alters as condition changes
What does QAM stand for?
Quadrature Amplitude Modulation
How do you encode even more data than amplitude modulation or shift keying alone would?
Combine them - QAM
What types of errors can row and column parity bit checks correct?
Single bit errors
