scispace - formally typeset
Search or ask a question

Showing papers on "Data Corruption published in 1999"


Patent
19 Apr 1999
TL;DR: In this paper, a negative acknowledgement (NACK) signal is generated in the event that message corruption is detected by a node on the network to flag the message corruption and generates a signal which is guaranteed to corrupt the message in progress by overwriting the stop bit of the UART character frame.
Abstract: A method for communicating information between networked UARTs to achieve reliable peer to peer communication. A negative acknowledgement (NACK) signal is generated in the event that message corruption is detected by a node on the network to flag the message corruption and generates a signal which is guaranteed to corrupt the message in progress by overwriting the stop bit of the UART character frame. This causes all receivers of the message in progress to cancel reception of a corrupted message and causes transmitters to retry transmission of the message. The invention does not impose any limits on the number of nodes on the network and does not require any special data content other than a unique source address that is assigned to each node on the network.

22 citations


Patent
Hideo Sunaga1, Nobuyuki Nemoto1
04 Mar 1999
TL;DR: In this paper, the authors propose a mechanism to keep an adequate distance between read and write address pointers to prevent data corruption and subsequent system disruption from happening by using dual port data buffer storage.
Abstract: An apparatus for resynchronizing data signals by using dual port data buffer storage, which prevents data corruption and subsequent system disruption from happening by employing a mechanism to keep an adequate distance between read and write address pointers. An input unit receives an incoming data stream having a cyclic data structure of N bytes. A data writing unit sequentially writes each data word of the received data stream into a storage unit with a capacity of 2N bytes in synchronization with a first clock. A data reading unit sequentially reads out each data word from the storage unit in synchronization with a second clock. A detection unit tests whether the write and read address pointers have come within a predetermined threshold distance. A relocation unit moves the read address pointer by N bytes to increase the distance between the read and write address pointers, when the detection unit has detected that the write and read address pointers have come within the threshold distance. An outputting unit transmits to an external destination the data words read out by the data reading unit.

15 citations