Input & Output Devices

• Barcodes are printed monochrome diagrams with light and darks parts. A barcode reader can read information from them. There are two main types, 1D and 2D
• 1D barcodes, or linear, are commonly used to identify products and medical samples
• 2D barcodes, or QR (quick response), can store more information in the same space, but they require more processing. They can serve as tickets, URLs or other information
• In pen-style readers, a light source and photo diode are placed in the tip of a pen. As it is dragged slowly over the barcode, the light source shines onto it and the diode measures the intensity of light shining back, and forms a waveform. Dark bars absorb light and white spaces reflect it, so the voltage waveform should be a duplicate of the barcode in a digital format. Then this can be used to look up, say, a product
• Pen-style readers are the most durable, but applications are limited since they must be in direct contact with a barcode to read it. Best suited for portable laptops or low volumes of barcodes.
• Laser scanners use a laser for a light source. It reflects off a moving mirror allowing barcodes to be read from a variety of angles. Economical and reliable.
• Charge-Coupled Device (CCD) readers use an array of tiny light sensors. Each sensor reads the light intensity in front of it. They are so small they can accurately read a barcode in one go.
• A camera-based imaging scanner uses image processing to decode a barcode as seen through the camera. This could be at any orientation and works even if it is blurry or damaged. Commonly used to read 2D barcodes
• Digital cameras use either a CCD sensor or a CMOS sensor. CCD produces higher quality images, but consumes significantly more power than CMOS.
• A digital camera sensor consists of millions of smaller sensors arranged in a grid.

  -Shutter opens and light enters the camera

  -Image is projected onto sensor at back of lens

  -Tiny sensors each measure brightness and turn it into an electrical signal

  -Resulting image is greyscale
• For coloured images, sensors are placed behind coloured filters
• Method 1: Incoming light is split into three paths that are each passed through one colour filter and picked up by its own set of sensors. Higher quality but creates bulkier cameras
• Method 2: Sensors are placed under a mosaic of RGB filters, eg the Bayer colour filter which has more green to better represent what we see. Software makes approximations of colours based on values of neighbouring pixels. Can be very compact
• In a laser printer, entire pages are printed at once.

  -The drum in the printer is negatively charged

  -A laser draws a negative image onto the drum according to a bitmap

  -Where the laser hits, the drum loses negative charge

  -When the drum rotates past the toner hoppers, the toner is attracted to parts that weren't hit by the laser because it is positively charged

  -Particles are bonded to the page through heat and pressure
• For coloured printing, a laser printer must complete the process 4 times, once for each toner - cyan, magenta, yellow, black (CMYK)
• A laser printer has a maximum quality of about 1200 dpi, so it cannot print photorealistic images. This is better suited to an inkjet printer
• RFID tags use radio waves to communicate with a reader. They do not require a line of sight to be read, and some can be read up to 300m away.
• An RFID tag has a chip with memory for storing data, and a coil of wire that serves as an antenna
• RFID tags can be tiny (eg a grain of rice) - this is why they can be injected under the skin of a pet for identification, for example
• Active tags are larger and have a battery for power to constantly transmit a signal. This is good for tracking items at greater distances
• Passive tags are smaller and cheaper, and rely on the radio waves from a reader for electromagnetic power. These are often used in cards like the Oyster card or bank cards
• Primary storage includes RAM and ROM, which are volatile or unable to be written to respectively. Thus, for saving and storing files, we must use secondary storage
• Optical disks include CDs, DVDs and Blu-Rays. They can be readable, recordable or re-writable.
• An optical disk is burnt by a laser to form dents in the surface, called pits. The areas surrounding them are called lands. These form one continuous track that spirals from the centre out
• When a laser is directed at an optical disk, light will either be reflected straight back when over pits and lands (reflective), or scattered when at a transition between the two (non-reflective). This can be converted into a binary signal.
• Writable optical disks use dye that can become opaque or transparent via a reversible chemical reaction. This allows a disk to be written to, and over, multiple times
• Optical storage is cheap, portable and lightweight, but the data is slow to access and it is very prone to scratches and other damage
• Blu-Ray disks use a smaller laser, so the pits and lands can be closer together, allowing for more data to be stored on one disk
• A hard disk drive uses iron particles polarised in a north or south state to hold information. A read/write head floats just above the surface and can read or write data.
• A hard disk can spin extremely quickly, up to 10,000RPM
• A hard disk can store more data at the same size by having multiple platters or decreasing the width of tracks
• A hard disk's tracks are concentric and are split up further into sectors
• Due to the moving parts used in hard disks, they aren't very portable, and have slow access times similar to optical disks. However, they are cheap and can have very high capacity for data
• In the past, magnetic storage was used to store backups of data
• Hard disks work better if they are defragmented
• Solid-state drives consist of NAND flash memory cells and a controller that manages pages and blocks of data
• In SSDs, data is stored in small amounts called pages, organised into larger amounts called blocks
• NAND flash memory cells work by forcing electrons across an insulated oxide layer into a floating gate transistor. Once the current is off, these electrons are trapped, and the charge is stored. No charge is a 1, some charge is a 0.
• NAND flash memory cannot overwrite existing data. An entire block must be erased before a page can be written to again. A new block that mirrors the original must be made to store the updated data - the original is marked as invalid, and is erased when it must be used again
• The oxide layers in SSD deteriorate over time as they are used until they are no longer reliable, so SSD technically has a finite amount of read/writes. However with modern technology this number is so large it is unlikely to be an issue