Manchester coding

Example of Manchester encoding according to the original article by E.G. Thomas.

Manchester coding, also called biphase-L coding, is a method of electrical coding of a binary signal in which at each bit time there is a transition between two signal levels. It is a self-synchronized coding, since in each bit the clock signal can be obtained, which makes possible a precise synchronization of the data flow. One disadvantage is that it consumes twice the bandwidth of an asynchronous transmission. Today there are numerous encodings (8b/10b) that achieve the same result but consume less bandwidth than Manchester encoding.

Manchester encoding is used in many telecommunications standards, such as 10 Mbit/s variants of the Ethernet standard, for example 10Base5 or 10Base-F.

Description

• Data and clock signals are combined in a single that self-synchronizes data flow.
• Each encoded bit contains a transition in the middle of the duration of the bits.
• A transition from negative to positive represents a 1 and a transition from positive to negative represents a 0.

Manchester coding example, according to Ethernet conventions

Manchester codes have a transition in the middle of the period of each bit. When there are equal and consecutive bits, a transition occurs at the beginning of the second bit, which is not taken into account by the receiver at the time of decoding, only the transitions uniformly separated in time are those that are considered by the receiver. There are some transitions that don't happen mid-bit. These transitions carry no useful information, and are only used to put the signal into the next state where the next transition will take place. Although this allows the signal to self-synchronise, it actually doubles the bandwidth requirement, compared to other codes such as NRZ Codes.

Manchester Coding as Phase Shift Keying

Manchester coding is just a special case of Phase Shift Keying, where the data to be transmitted controls the phase of a rectangular carrier wave. To control the amount of bandwidth consumed, a filter can be used to reduce the bandwidth down to a low value like 1Hz per bit/second, and keep it down so as not to lose information during transmission.

Advantages and Disadvantages of Using Manchester Coding

The following main advantages can be highlighted:

• Manchester encoding or bifase-L encoding is self-synchronized: it provides a simple way to encode bit sequences, even when there are long sequences of periods without level transitions that can mean sync loss, or even errors in bit sequences. That's why it's highly reliable.
• Detection of delays: directly related to the previous feature, at first glance it might seem that a half-bit error period would lead to an inverted output at the receiving end, but a more careful consideration reveals that for typical data this would lead to code violations. The used hardware can detect those code violations, and use this information to properly synchronize in the correct interpretation of the data.
• This encoding also assures us that the continuous component of the signals is zero if positive and negative values are used to represent the signal levels, making it easier to regenerate the signal, and avoiding the energy losses of the signals.

The main associated disadvantages are the following:

• Double bandwidth of the data signal: a consequence of the transitions for each bit is that the bandwidth requirement for the Manchester encoding is the double compared in the asynchronous communications, and the signal spectrum is considerably wider. Most modern communication systems are made with coding lines that pursue the same goals, but better optimize bandwidth, making it smaller.

Data Representation Conventions

There are two contrary conventions in interpreting encoding:

• In the original article of E.G. Thomas of 1949 and in many other authors who follow it, like Andrew S. Tanenbaum, bit 1 is a high-low transition and bit 0 low-high.
• Other authors like Stallings, and IEEE 802.3 standard consider that bit 1 is the low-high transition and bit 0 the opposite.

It should be noted that differential Manchester coding is not a specific interpretation of Manchester coding.

Differential Manchester coding

Differential Manchester Coding (also CDP; Conditional DePhase encoding) is a data encoding method in which the data and clock are combined to form a single, self-synchronizing data stream. It is a differential encoding that uses the presence or absence of transitions to indicate a logical value. This provides some advantages over Manchester Encoding:

• Detecting transitions is often less prone to mistakes than comparing to land in a noisy environment.
• The presence of the transition is important but not polarity. Differential encoding will work exactly the same if the signal is inverted (changed variables).

A bit '1' it is indicated by making the first half of the signal equal to the last half of the previous bit, ie no transition to the beginning of the bit. A bit '0' it is indicated by making the first half of the signal opposite to the last half of the last bit, ie with a transition at the beginning of the bit. In the middle of the bit there is always a transition, either from high to low or vice versa. A reverse configuration is possible, and there would be no downside to using it.

Example of Cudification Manchester Diferential.

A related method is Manchester Encoding in which the significant transitions are mid-bit transitions, encoding the data by its direction (positive-negative is the value '1', negative-positive is the other).

Manchester Differential is specified in the IEEE 802.5 standard for Token Ring Networks, and is used for many other applications, including magnetic and optical storage.

Note: In the Manchester Differential encoding, if the '1 is represented by a transition, then the '0' is represented by 2 transitions and vice versa.

The existence of the guaranteed transitions allows the signal to self-synchronize, and also allows the receiver to align itself correctly; the receiver can identify if it is misaligning by half a bit period, since there will no longer always be a transition during each bit period. The price of these benefits doubles the bandwidth requirement compared to less complex NRZ encoding schemes.