In telecommunications, a frame is a packet which has been encoded for transmission over a particular link. This process involves, at a minimum, adding delimiters to distinguish the packet from dead air, address and control fields specific to the link, and checksums to detect errors. (Sometimes the address, control, and checksum fields from the higher-level protocol are used directly.)

Frame may also refer to the way a multiplexer divides the underlying communication channel so that it can be used simultaneously for more than one transmission. Notionally, each frame is a slot which could be filled by a transmitted packet. In these schemes, not all frames are necessarily in use at once.

  • In the multiplex structure of pulse-code modulation (PCM) systems, a frame is a set of consecutive time slots in which the position of each digit can be identified by reference to a frame-alignment signal. (This signal does not necessarily occur in each frame.)

  • In a time-division multiplexing (TDM) system, a frame is a repetitive group of signals resulting from a single sampling of all channels. The term in-frame is used to indicate that a time-division multiplexer is properly synchronized with the demultiplexer on the other end of the link, so that (barring in-flight data corruption) packets will be properly received.