Compression, Encryption and hashing
Define 'Sampling'
At a given time point in a sound, a measurement is taken
Give the formula for the size of the sampled resolution
Sampling rate(no of bits) x sample resolution x length of the sound(seconds)
Define 'Compression' and give the two types of compression
The process of reducing a file's size.
Lossy compression
Lossless compression
Give two benefits of compression
- Reduces file size
- Files are quicker to be transferred
What does 'Lossy compression' and name files where it is used on
Reduces file size by permanently removing data from the file.
JPEG
MP3
MPEG-1
MPEG-2
What does 'Lossless comprssion' do and give examples of where it is used
Reduces the size of the file however doesn't remove data from the file.
PNG
ZIP
GIF
Give 2 ways of reducing file size of an image
- Store a lower number of colours
- Store a larger area of pixels as a single colour
Name 2 types of lossless compression
Dictionary coding
Run-length encoding
What does 'Dictionary coding' do
Replaces recurring sequences with shorter unique codes
Where would dictionary coding be useful and where would it not be useful?
Useful: Large text documents with repeating data
Not Useful: text documents with no repeating data, takes up more space
What does 'Run-Length encoding ' do
Condenses identical data into single values with a count
Where would you use run-length encoding?
In bitmap images. Where you have contiguous pixels of the same colour so you store the binary for one of the pixels and multiply it by the number of pixels with that colour
What is the purpose of encryption?
To make the original code impossible to crack without the key
What is the difference between Cipher and Plain text
Plain text - text that is reabdable by anyone
Cipher text - text that can only be read by someone with a decryption key
What is 'Symmetric encryption' and give some drawbacks of it
Where a single key is used to both encrypt and decrypt a message.
- Messages can be cracked by someone intercepting the key
- Messages can be crakced by someone duplicating the production of the key to get a copy
What is 'Asymmetric Encryption'
Where the key used to encrypt the message (public key) is different to the key needed to decrypt (private key) the message (key pair)
Give benefits of 'Asymmetric encryption'
- The sender and recipient are sure no one else can read their message
- You're sure the message hasn't been modified
- You can be sure the message is authentic
How does asymmetric encryption work
The sender gives the recipient their public key. The recipient uses the public key to encrypt a message and send back to the sender. The sender then uses their private key to decrypt the message
Where would you use symmetric and asymmetric encryption
Symmetric: On large files and databases
Asymmetric: Confidential data
Define 'Hashing'
Transforms a string into a fixed length value or a key that represents the input string
Give 2 common hashing algorithms
MDA
SHA-1
SHA-256
SHA-3
Define a 'hash table'
A data structure used to remove & insert key value pairs
What makes hashing a different process to encryption
Hashing is a one-way process, meaning you can't get the original value from the hashed value even if you have the hashing algorithm
Why do hash tables need to be large in size
In order to reduce the chance of a collision occurring
Define a "Collision"
Where two different inputs produce the same hashed value
Give 3 features of a good hashing function
- Be calculated quickly
- Result in as little collisions as possible
- Use as little memory as possible
Define "Open Addressing"
A process which searches for the next available empty space in a hash table and stores the item there.
Define "Linear probing"
Searching for an empty spot by checking the next slot sequentially
Give 2 disadvantages of "linear probing"
- It prevents certain items being stroed in their correct location
- Can cause clustering
Define "Clustering"
Where many inputs have the same hashed value or similar hashed values causing large blocks of occupied slots, which increases search time (reduces efficiency)
Give another way of handling collisions
2D hash tables, this can allow more than one item to be placed in the same index (chaining)
Give a 3rd way of handling collisions
Overflow tables, a second table used to handle collisions when the first table is either full or items can't fit in their expected location