retransmit

Retransmit is the act of transmitting data again, typically after an initial transmission fails or a new copy is required to ensure delivery. In communication systems, retransmission can occur at various layers: as a relay in a network to extend coverage or as a remedy for lost or corrupted data by resending the same information.

In digital communications, retransmission is a core mechanism for reliable data transfer. Protocols such as automatic repeat request (ARQ) use acknowledgments and timeouts to detect errors; when a packet is not acknowledged or a negative acknowledgment is received, the sender resends the data. Stop-and-wait, sliding window, and selective repeat are common ARQ variants, differing in how many unacknowledged packets may be in flight and which lost packets are re-sent. In TCP, for example, a packet is retransmitted after a timeout or after receiving duplicate acknowledgments, and this interacts with congestion control to avoid network congestion.

Retransmission is also used in wireless and broadcast networks via relay stations or repeaters that forward signals to extend coverage or improve reliability. In satellite communications and content distribution networks, retransmission can occur at middleware or protocol level to recover from link outages. In streaming and real-time media, scheduled or opportunistic retransmission of missing data may be used alongside forward error correction.

Trade-offs include increased latency and bandwidth use versus improved reliability. Systems may employ forward error correction to reduce retransmissions, or adaptive strategies that adjust retransmission aggressively in good conditions and back off in congested networks. The term is sometimes informally used to describe rebroadcasting content, but in technical contexts retransmission is a targeted resend of data to ensure correct delivery.