Data Representation Flashcards
1
Q
What is hex used for?
A
- addresses in memory
- error message codes
- machine language instructions
- HTML colour codes
2
Q
Why is hex used?
A
- easy to convert between hex and binary
- easier to understand than a long string of binary
- numeric representation of a value is more compact
- likelihood of making an error when copying hex is lower than binary equivalent
3
Q
What is unsigned binary?
A
Where each bit used represents another binary column
4
Q
What is signed binary?
A
The sign bit is used to show whether the number is positive or negative
5
Q
What is the MSB?
A
“Most Significant Bit” which is the left-most bit, also known as the sign bit
6
Q
What is an overflow error?
A
Result of binary operation is too large to represent within the number of bits available
7
Q
How do you perform 2’s compliment on a number to make it negative?
A
1-Write out number in binary
2-Flip all the bits
3-Add one
8
Q
What is fixed point binary?
A
Consists of a whole part and fractional part separated by predetermined fixed binary point
9
Q
What is a radix point?
A
Character used to separate the integer and fractional parts of a number
10
Q
What two parts do floating point numbers consist of?
A
A mantissa and an exponent
11
Q
What is normalisation of floating point numbers?
A
Process of maximising the precision of values that can be represented with a given number of bits in floating point binary.
12
Q
What are the benefits of normalisation?
A
- Ensure the maximum possible accuracy for a given number of bits
- Ensures only one representation of each number within a given number of bit; simplifies arithmetic operations
13
Q
What is the absolute error for floating point numbers?
A
Difference between number and approximate value of the floating point binary representations of that number. Always positive.
14
Q
What is relative error of a floating point number?
A
Absolute error divided by the actual number, as a percentage. This is often more useful.
15
Q
What is under flow of fractional numbers?
A
Result of a calculation has a fractional part too small to represent in the allocated number of bits.
16
Q
What are the advantages and disadvantages of floating point binary?
A
+ Much wider range of numbers for a given number of bits
+ More relative precision for smaller numbers
- Can result in rounding errors
- More difficult to perform calculations (slower)
17
Q
What are the advantages and disadvantages of fixed-point binary?
A
+ processed more quickly as there is no movement of the binary point
+ Hardware required is much less complicated making it smaller and less energy intensive
- Fixed and limited precision
- Could cause rounding errors
18
Q
What are the causes of overflow?
A
- Multiplying or adding large numbers together
- Dividing numbers by a small number less than 1
19
Q
What are the causes of underflow?
A
- Multiplying by a number by a very small number less than 1
- Dividing a small number by a number greater than 1
20
Q
What are the consequences of underflow or overflow of binary numbers?
A
- Erroneous calculations
- Data corruption
- Program crashes
21
Q
How is underflow or overflow dealt with?
A
Checking the carry at the MSB and LSB (right and left) and setting a “flag” to state an error has occurred.
22
Q
What is a number system?
A
Way of representing numerical values that are used in counting, measuring, and operations
23
Q
What are ordinal numbers?
A
Adjectives that describe the numerical position or order of an item within a list (indexing). There must be an ordered set with at least 2 items.
24
Q
What is a character?
A
Single unit of textual information
25
What is a character set?
Scheme of mapping that maps a set of characters to binary numbers that is recognised by a computer system
26
What is ASCII?
A character set created to encode alphanumeric characters as well as special characters that used 7 bits to store each character
27
What is Unicode?
- A character set developed with the aim of representing every possible character.
- It uses up to 48 bits to store each character
- First 128 characters are the same as ASCII making it backwards-compatible
28
Advantages of parity bits?
- Simplest implementation of error detection
- Very low cost system as it only involves one additional bit.
29
How is a parity bit used by a computer?
1. The value of the parity bit is set to either 1 or 0 to satisfy the parity
2. This is done when character is first generated
3. The number of 1s is counted
4. If there are not the correct number of ones, then an error likely occurred
30
What were the limitations of ASCII?
- Could not represent more than 128 or 256 characters
- Resulted in different version that contained different symbols for each language and were incompatible with each other
31
What are the advantages of the parity bit?
- Relatively easy to implement
- Does not require significant extra data to be included in the transmission or retrieval
32
What are the disadvantages of the parity bit?
- If an even number of transpositional errors occurs the corruption is not detected
- Not possible to identify which bit has been corrupted
- Data correction not possible
33
What is majority voting?
Each bit is sent multiple times (odd number greater than 2). The receiver checks the bits and if they are not all the same it assumes the one to received the most copies of is the correct value.
34
What are the advantages of majority voting?
- Allows for error correction as it knows exactly where the error occurred
- No need for retransmission
35
What are the disadvantages of majority voting?
- Triples the amount of data that is sent or stored
- Is not entirely reliable as 2 transpositional errors can cause an incorrect correction
36
What is a check digit?
Extra digit appended to a number in order to confirm that the number has been transmitted or retrieved correctly
37
What is a checksum?
Similar to a check digit except a larger sum is transmitted with the data rather than a single digit
38
What makes a good checksum or check digit algorithm?
It will produce very different values even if there is a small change to the original data
39
What are the uses of a checksum?
- Confirm integrity of whole files or applications
- Checksum is provided by the owners so that the checksums of applications can be checked to ensure correct download
40
What is a pixel?
The smallest unit of a bitmap image
41
What is the maximum number of colours that that can be be represented given the colour depth?
2colour depth
42
What is the colour depth of an image?
The number of bits that are used to encode each pixel
43
What is the size of an image?
The width x the height of the image in pixels
44
What is the resolution of an image?
The number of data (pixels) in each square inch (PPI)
45
What is a bitmap image?
An image file format made up of pixels which represent a small square of colour
46
How do you calculate the file size of a bitmap image?
Image size x Colour depth
47
What is a vector image?
Way of representing images using a series of geometric **objects** each of which have properties
48
What are the advantages of bitmap images?
- Can represent photorealistic images
- Each individual pixel can be manipulated and edited
49
What are the disadvantages of bitmap images?
- Often have a larger file size
- As images size is increased the resolution decreases leading to pixelation
50
What are the advantages of vector images?
- Typically have smaller file sizes
- Image can be infinitely scaled
- Each shape or path can be individually altered
51
What are the disadvantages of vector images?
- Are not suitable for complex images such as photographs
- Require to be rendered which requires processing power and time especially for complex images
52
What is an analogue signal?
A continuous signal that varies with time that can take an infinite set of values
53
What is a digital signal?
A signal that must always take one of a specified range of values and can only change at specified intervals
54
What is analogue data?
Data that has no limits to the values that it can take. Represents information in the real world
55
What is digital data?
Data that can only take particular values
56
What is the sample resolution of a recording?
Number of bits used to represent each sample
57
What is the sample rate of a recording?
Number of samples taken per second
58
What is quantisation?
- The process of assigning an integer value to the amplitude of a signal at a specific point in time
OR
- mapping of continuous infinite values to a smaller set of discrete finite values
59
What is a quantisation error?
The difference between the analogue signal and the closest available digital value at each sampling instant
60
What does metadata typically contain?
- encoding format
- colour depth
- size
- time and data the image was taken
61
What is Nyquist’s theorem?
The sampling rate must be at least twice the highest frequency of the signal being sampled such that it can be accurately reproduced.
62
What frequency are digital recordings stored in?
44.1KHz
63
What is a MIDI file?
A set of stored instructions that can be used to tell musical instruments and synthesisers how to produce a piece of music.
64
What does a MIDI event message contain?
- when the note should be played
- pitch, volume and duration
- The instrument that should play that note
- Clock signals to coordinate multiple instruments
- Other control data (tempo, special effects)
65
Advantages of MIDI
- Musical content can be easily edited
- More compact representation of audio
- Simple method to compose algorithmically
- No data is lost through sampling
66
Disadvantages of MIDI?
What the file actually sounded like depends entirely on the setup so the same file can sound totally different.
67
What are the main benefits of compression?
- Amount of data required to be stored is reduced so more files can be stored
- Amount of data required to transmit data is reduced so less bandwidth is required, reduced congestion on the network, increased transmission speeds
68
What are the other benefits of compressions?
- Using compressed images in web pages reduces loading times
- Improves playback of media streams with less buffering
69
What is lossy compressions?
Form of compression where unnecessary detail is removed permanently from the original data.
70
What are the +ive and -ive of lossy compression?
+Greatest reduction in file size
-Permanent loss of quality
Ideal for image and audio
71
What is lossless compression?
Form of compression that uses algorithms to efficiently represent repeating patterns within the original data.
72
What are the properties of lossless compression?
+No original data is lost so can be perfectly reconstructed
-Less reduction in file size
This is the only option where data cannot be lost such as text and program files
73
What is the JPEG image file?
- Form of lossy compression that averages similar colours across adjacent pixels
- Suitable for photographs which do not have sharp edges
74
What is a PNG image file?
- Lossless compression that efficiently encodes repeating patterns
- Higher quality images with sharp edges such as logos, icons and texts.
- Larger file sizes
75
What is an MP3 file?
- Uses lossy compression to remove detail from audio recordings that we cannot perceive then other algorithms are used
- Results in files that are around 10 times smaller
76
How is text compressed?
Text can only be compressed using lossless compression. Otherwise text files such as source code could fail to interpret/compile otherwise.
Therefore we use Dictionary Compression or Huffman Tree Encoding.
77
What is Run Length Encoding?
Simple form of lossless compression that represents repeating patterns of data more efficiently. E.g. 6 white pixels in a row could be encoded into 6w
78
What is Dictionary Based Compression?
A form of lossless compression where frequently occurring pieces of data are replaced by an index of a dictionary.
It works well on large documents with many repeating words.
79
How does Dictionary Based Compression work?
1. Each word in a piece of text is stored in a dictionary along with an index number
2. Message is then encoded using the dictionary index values in place of the original data
3. The resultant message will be significantly smaller, although the dictionary must also be sent/stored with the file.
80
Why was Unicode introduced?
- To enable the representation of a greater range of characters
- All modern languages can be represented
- Each character has a unique representation in Unicode which wasn’t the case for ASCII
81
What is encryption?
The process of making information unreadable to unauthorised parties through the use of an encryption algorithm.
82
How does encryption work?
A cipher is applied to plaintext using a key to transform it into ciphertext.
83
What is computational security?
A cipher is considered computationally secure if it cannot be broken using current technologies within a reasonable timeframe.
84
What are methods for cracking ciphers?
- Frequency analysis
- Looking for other patterns
- Dictionary attacks
85
What is frequency analysis?
It involves looking for frequent symbols within the cipher text and mapping them to common letters in the language of the original message.
86
What is a dictionary attack?
A list of common words or phrases that can be compared to the cipher text to see if any patterns arise
87
What is a Caesar cipher and how does it work?
A basic substitution cipher where each letter is substituted for another. A numeric **key** is used to determine the distance that each letter should be shifted by.
88
Why is a Caesar cipher not computationally secure?
- It can be cracked almost instantly using basic brute force attacks as there are only 25 possible keys
- Frequency analysis is very effective
- Each character is always converted to the same cipher text letter
- As it is only a shift cipher you only need to map on of the letters between plaintext and cipher text to deduce mapping for all letters
89
When is a Vernam cipher perfectly secure?
- key used is truly random (e.g. using atmospheric radiation)
- key is at least as long as the plaintext
- key used only once and then discarded
- key itself must be transferred securely
90
How do you use a Vernam cipher?
To encrypt a message,each character in the plaintext and key are encoded into their binary representations in a given character set and bitwise XOR them together to get another character. Same process for decryption.
91
Why is a Vernam cipher perfectly secure?
- Cipher text contains no useful information about the plaintext as characters are randomly distributed
- Mapping of plaintext to cipher text is different for each letter in plaintext
- Brute forcing cannot reveal plaintext and has too many possible keys
- Frequency analysis doesn’t work as multiple plaintext letters can map onto the same cipher text letter
92
How does a check sum work?
1. Checksum is calculated from the binary data before it is transmitted
2. Checksum is recalculated when the packet is received
3. If the checksum received in packet matches recalculated checksum then data is correct.
93
What does a vector object contain?
- Coordinates
- Fill colour
- Edge thickness
- Edge colour
- Object type
94
How can data become corrupted?
- Cosmic rays
- Magnets changing magnetic media
- Scratches on a CD
- Cables become damaged
95
How does Run-Length-Encoding work?
A run is a series of pixels of the same colour. Pairs of values would be stored which consist of the run length and the colour of the pixel run.
96
How does an ADC work?
- Takes an analogue signal as an input
- Samples the amplitude at regular intervals
- Rounds amplitude to closest discrete value
- Range and number of sample values controlled by sample resolution and rate
