US 5,859,857 A · Method for the integration of additional data in digital data packets

Abstract

In digital data packets comprising a first datum from which a second datum can be unequivocally computed, e.g. an error correction value, one or a plurality of additional data can be introduced by manipulating said second datum without altering the structure of the data packet. For this purpose, one or a plurality of bits of the second datum are modified in an unequivocally reversible manner, e.g. inverted, according to the additional datum which is also digital. The recuperation of the additional data is effected by comparing comparative values which are obtained by reversing all possible combinations of additional data in a received data packet to a value which is computed from the first datum of the data packet: The presence of the corresponding combination of additional data results from the consistency of the latter with one of said comparative values. Another possibility is to limit the inquiry to a fraction of the possible combination. A preferred application is the recognition of endlessly circulating ATM cells in a ring-shaped communication network. The one-bit additional datum is inserted in the HEC byte by a monitor node, and each ATM cell which reaches said monitor node and includes said additional datum is recognized as being undeliverable.

Claims6

Dependent claims

A method of integrating additional data in a digital data packet, said packet is capable of being represented by a sequence of bits and comprises a first datum as well as a second datum which is unequivocally computed from said first datum, wherein at least one bit of said digital data packet is manipulated as a function of said additional data in an unequivocally reversible manner in order to encode said additional data in said data packet without altering a structure of said data racket;

and
obtaining said additional data from said digital data packet, wherein:
first provisional data are derived from said second datum by reversing, in a remover, the corresponding manipulation for at least one additional value of at least one additional datum of said second datum;
an error check procedure is applied by an interpreter to said first provisional data and to the unaltered data packet, the result of said procedure being an indication of probability at least with respect to whether said first datum corresponds to its original value;
the highest value of probability is determined by a second interpreter;
and
the combination of additional data yielding the highest value of probability is output by the interpreter.

4 more elements

and applying said method for the recognition of faulty and/or multiply circulating data packets in a ring-shaped communication network,
said communication network comprising a plurality of nodes which are serially connected by a transmission line having a unique transmitting direction and one of which serves as a monitor node;
an error correction byte being provided in a header portion of said data packets for the purpose of recognizing transmission errors, and each node computing, from the bits of said header portion of a given data packet, a decision value which indicates whether an error is present in said header portion or not;
a monitor node being provided for the purpose of recognizing multiply circulating data packets, which marks each transiting data packet in a recognizable manner, and monitors whether the concerned data packet is already marked, and
said error correction byte of the data packet representing said second datum and that part of said header portion which is protected by said error correction byte representing said first datum,
wherein
said monitor node inverts at least one of the bits of said error correction byte in order to mark the transiting data packets,
for the purpose of recognizing transmission errors in the header of the data packets, each one of said nodes additionally inverts those bits of the header of the respective data packet which are also inverted in said monitor node, a second decision value is computed from the total of bits resulting therefrom, and the combination of said first and said second decision value indicates whether an error is present in said header, and
for the purpose of recognizing multiply circulating data packets, said monitor node computes a decision value from the bits of said header of the concerned data packet which indicates whether those bits which are inverted for marking purpose are being inverted or not.

6 more elements

Dependencies are inferred from the claim text.

Method for the integration of additional data in digital data packets

Life of this patent

Method for the integration of additional data in digital data packets

Life of this patent

Abstract

Claims6