Encoding is the process of taking a signal and converting it into a form that can be transmitted over a communication medium.
Data can either be analog or digital. Signals can either be analog or digital. Encoding must be done between the 2 types, thus 4 variations. For this module, digital data to digital signals is focused most.
One logic state represented by positive voltage and other by negative voltage.
Unipolar
Section titled “Unipolar”All signal elements have the same sign.
Length of a bit
Section titled “Length of a bit”Aka. duration of a bit. Time taken for transmitter to emit a bit. Larger the bit length, lesser the data rate but convenient for the devices. If the bit length is too short, it might be difficult for the receiver to detect the signal elements accurately.
Baseband
Section titled “Baseband”For digital signals: entire medium carries only one data signal at a time. Example: ethernet.
For analog signals: original frequency range of an analog signal before it is modulated.
Broadband
Section titled “Broadband”Any type of signal transmission technique that carries two or more different data in separate channels.
Encoding Types
Section titled “Encoding Types”Level encoding
Section titled “Level encoding”Data is represented in voltage levels. Less reliable due to noise and interference.
Differential encoding
Section titled “Differential encoding”Data is represented by changes rather than levels. More reliable detection. More complicated than level encoding.
Examples:
Bipolar encoding
Section titled “Bipolar encoding”A return-to-zero type encoding. 2 non-zero levels and zero level are used. Standard bipolar encodings are designed to be DC-balanced.
Examples:
Bipolar violation
Section titled “Bipolar violation”A violation of the bipolar encoding. Where 2 pulses of the same polarity occur without an intervening pulse of the opposite polarity. Indicates an error in transmission.
Biphase encoding
Section titled “Biphase encoding”Uses transitions to represent bits. Combines clock and data. Fixes synchronization issues due to prolonged constant voltages.
Pros:
- Self-clocking
- No DC component
- Error detection using absent mandatory transition
Cons:
- At least 1 transition per bit (possibly 2)
- Requires more bandwidth
- Maximum modulation is 2 times of NRZ
Examples: