Repeaters operate within the physical layer.
Effect of cable segments on signal quality
There is a loss of signal strength as a signal propagates along the length of any cable. This is a natural phenomenon that occurs for any type of transmission: e.g., electrical power or network data.
There is also a loss of signal at any connection points along the cable such as connectors. If a cable segment were allowed to exceed the maximum length or the maximum number of attached systems to the segment, the signal quality would deteriorate. (If unchecked this would ultimately lead to errors in the data.) The longer the cable, the greater the loss of signal (expressed in decibels, dB) and the more probable it is that the represented bits are corrupted when they are finally received at the end of the cable segment.
To provide a link with a low probability of error, the received signal needs to be above the receiver threshold for reliable communications. Repeaters are a physical layer equipment that regenerates the digital signal to compensate for the transmission loss along a cable segment. When a signal passes along a cable segment and through a repeater, it is expected to be restored to a perfect copy of the signal that was originally sent. Hubs and repeaters may be used between a pair of segments (see below) to provide signal amplification and regeneration to restore a good signal level before sending it from one cable segment to another.
Ethernet Repeaters
Ethernet hubs and repeaters operate at the Physical Layer and are defined by IEEE 802.3c/d. They are used to connect together one or more 10 Mbps Ethernet cable segments of any media type (10B5, 10B2, 10BF or 10BT).
A simple repeater might have two ports - one connecting to each of two cable segments. Many Ethernet repeaters provide more than two ports are known as a "multi-port" repeater, or Ethernet Hub. There is no functional difference between a hub or a repeater and we will use the two terms interchangeably.
If an Ethernet segment were allowed to exceed the maximum length or the maximum number of attached systems to the segment, the signal quality would deteriorate. (If unchecked this would ultimately lead to errors in the data.) Hubs and repeaters may be used between a pair of segments (see below) to provide signal amplification and regeneration to restore a good signal level before sending it from one cable segment to another.
By allowing two or more LAN segments to be connected, they allow the network to span a larger distance. They also provide electrical isolation from failures in the cable or attached systems, protecting equipment on other LAN segments from the effect of the fault. A key fact about hubs and repeaters is that they allow users to share an Ethernet LAN. A network of repeaters and hubs is therefore called a "Shared Ethernet" or a "Collision Domain". The various systems sharing the Ethernet all compete for access using the CSMA/CD access protocol. This means that only one system is allowed to proceed with a transmission of a frame within a Collision Domain at any one time. Each system has to share a proportion of the available network capacity.
Some key uses of repeaters are:
Effects of Increasing the Maximum Length of Cable between Two End Systems
The various systems sharing the Ethernet all compete for access using the CSMA/CD access protocol. This means that only one system is allowed to proceed with a transmission of a frame within a Collision Domain at any one time. A very important fact about hubs and repeaters is that they allow systems to share use of an Ethernet LAN. In a network of repeaters and hubs is there is therefore only allowed to be one sender active at a time. The capacity of the media of the entire LAN is shared and is known as a "Collision Domain". The timing requirements of CSMA/CD require a maximum slot time across any collision domain. This leads to the limit on the maximum number of in-series cable segments that are allowed by the 5-4-3 rule when repeaters are used to build a larger LAN.
Need to Regenerate Preambles and Repeating Runt Frames
The DPLL at each receiver needs to lock onto the clock embeded in the Manchester encoded data. This takes a a longer time when the clock has a different frequency of phase to the previously received frame, which will always be the case when the frame is sent from a different source. This means it takes a time for the receiver clock in a reepater to lock on to the timing of (completely synchonise with) of a received frame's preamble. This means the initial part of the preamble is never really reeeived by a repeater. If this shortened preamble were to be sent it may be insufficient to allow a NIC to acquire a lock before the start of the frame. This is not allowed to happen because if a shortened preamble were to be sent it may be insufficient to allow a NIC to acquire a lock before the start of the frame. The repeater must therefore reconstruct the full preamble before the frame is sent out of the repeater.
An important job of a repeater is therefore to reconstruct the full preamble as a frame is sent out of the repeater. In reconstructing the preamble, a repeater will necessarily delay the signal. Note that since the repeater is responsible for changing the bit timing,it can not guarantee that the Inter-Frame Gap (IFG) generated by the sender is preserved in all cases.
Note: a repeater might introduce additional delays in processing the preamble, it therefore can not guarantee that the Inter-Frame Gap (IFG) generated by the sender.
A repeater design therefore needs to minimise any additional delay from regenerating the preamble. Therefore, there is a limit to the number of repeaters that can be placed in series when using CSMA/CD. This is reflected in the maximum number of repeaters permitted by the 5-4-3 rule. This limits delay to avoid the inter frame gap (IFG) reducing below 47 bit times.
Regenerating JAM signals and Repeating Runt Frames
Each repeater monitors the signal it receives on each active port. If a collision (or JAM) is detected on any port, the repeater implements a part of the CSMA/CD algorithm. It stops transmission and sends a 4 byte JAM sequence, effectively forcing any receivers on the connected cable sgments to abort their frame reception. At the same time, the repeater regenerates the JAM signal to all active output ports - ensuring that all connected cable segments also experience a JAM signal, and any transmitting NIC will be aware of the collision, even when the original colision was not on a directly connected cable segment.
Repeaters can also be designed to avoid repeating Runt Frames - frame that are too small.
Partitioning Cable Segments.
Some repeaters can detect other cable faults. When any fault is detected, the repeater temporarilly inhibits the port where the fault was observed, and does not repeat data received on this port, this isolates problems to a single cable segment, allowing the remainder of the LAN to continue working. Usually a fault light illuminates to show that this partition mode has been enabled. If the fault is rectified, the repeater restores fowarding of data from and to the port.