[next] [prev] [prev-tail] [tail] [up]
Receiver needs to determine bit information from the signal level, for instance, by sampling the
signal level at the middle if the time interval and comparing the value to a threshold.
- Receiver needs to know the timing of each bit in transmitter through a
- A common clock (expensive)
- Mechanism based on the encoding of the transmitted signal
- Interference and distortion can be reduced by
- Avoiding direct current (dc) components
- Concentrating the transmission power at the center of the bandwidth
- Bit error rate increases with: Data rate, and (noise strength) / (signal strength) ratio
- Data rate can be increased with: Bandwidth and Encoding
Positive and negative voltage levels for binary digits.
- Easy to engineer
- Make efficient use of bandwidth
- Suffers from the presence of dc component
- Lack of synchronization capabilities due to potential of long runs of unchanged voltage levels.
- Attractive for digital magnetic recording, but not for signal transmissions.
- The spectral density graph shows that most of the energy spent between dc and half the bit rate
NRZI is a differential encoding in which the signal is decoded by comparing the polarity of adjacent signal levels: 1 encoded by transition between levels and 0 encoded by a lack of
transition.
Similar advantages and disadvantages as NRZ-L.
A multilevel binary approach in which binary 0 is represented by a lack of pulse, and a binary 1 is
represented by a positive or a negative pulse. The binary 1 pulses must alternate in
polarity.
AMI stands for alternate mark inversion, whereas mark and space historical references to binary
digits 1 and 0.
A multilevel binary encoding that complements the bipolar-AMI encoding: binary 1 is represented
by a lack of pulse, and a binary 0 is represented by a positive or a negative pulse. The binary 0
pulses must alternate in polarity.
Manchester encoding is a biphase encoding in which the transition takes place in the
middle of the bit period: a low-to-high transition for 1, and a high-to-low transition for
0.
- Allows for clocking mechanism for both kinds of bits.
- Modulation rate twice than that of NRZ, implying a greater bandwidth
- The spectral density graph shows that there is not a dc component and the bandwidth is relatively
narrow
- Noise on the line has to invert the signal before and after the inverted bit to avoid detection.
- Has been specified for the IEEE 802.3 standard for baseband coaxial cable and twisted-pair
CSMA/CD bus LAN's.
- Require high signaling rate relative to data rate, making it too costly for long-distance
applications.
A biphase encoding in which transition at the start of the bit period represents 0, and a lack of
transition at the start of the bit represents 1. In addition, a transition occurs at the middle of each
bit period just for the purpose of clocking.
- Has been specified for the IEEE 802.5 token ring LAN, using shielded twisted pair.
The bipolar-AMI encoding supplemented with a scrambling scheme, which uses two code violations
to ensure synchronization in runs of 0's.
- Replace `00000000' with `000+-0-+', if the preceding voltage pulse was positive
- Replace `00000000' with `000-+0+-', if the preceding voltage pulse was not positive
- The amount of data remains unchanged.
- The spectrum graph shows that there is no dc component, with most of the energy concentrating
in a relative sharp spectrum. Making the encoding suitable for high-rate transmissions.
- Used mainly in North America.
The bipolar-AMI encoding supplemented with the following substitution scheme for `0000'
runs.
| Number of bipolar pulses (ones) since last substitution |
| Polarity of preceding pulse | Odd | even |
| - | 000- | +00+ |
| + | 000+ | -00- |
|
- Used in Europe and Japan
- Successive violations are of alternate polarity to avoid dc component.
[next] [prev] [prev-tail] [front] [up]
