Alternate Mark Inversion. The signalling used on T-1s before it was replaced by a new standard, and still used on some (mostly legacy) equipment.

Data on a T-1 is transmitted as no voltage for a zero-bit, and voltage for a one-bit, where each one-bit has the opposite voltage from the last. The reason for the voltage reversal is simple. If a bi-polar violation is seen (two positives in a row, or two negatives in a row), the equipment knows that at least one bit was dropped.

Since the receiving equipment derives timing from the incoming signal, 8 zero-bits in a row could possibly cause the equipment to go out of sync. Therefore, if 8 zero-bits are to be transmitted in a row, the equipment will subsitute a one-bit as the eighth bit. Doing this limits the data bandwidth on an AMI circuit to 56kb/sec per channel rather than 64kb/sec, since one bit out of eight may be altered by the carrier equipment. This means that 8000 times per second, 7 bits are transmitted from end to end, rather than the actual 8 bits being sent on the T-1 line.

AMI was supplanted to some degree by B8ZS, which uses bi-polar violations to overcome the 8 zero-bit limit.

Because of the difference in handling data, when testing a T-1, certain tests are useful to determine whether equipment on a line is misconfigured for AMI or B8ZS (which is pretty common). Certain tests send data which is changed by equipment configured for AMI. Due to human error, it is possible that in a circuit, one piece of equipment may be misconfigured, to be AMI when the line is supposed to be B8ZS. This is a common problem that must be diagnosed.

Log in or register to write something here or to contact authors.