Error detection and correction system
AssigneeBell Telephone Laboratories, Incorporated
InventorsDavid Joel Goodman, Raymond Steele
Life of this patent
Estimating the term…
Abstract
An error reducing arrangement is provided for a signal receiver wherein a sequence of received signals is divided into groups and each group is partitioned into subgroups. A signal representative of the deviations among the signals of each group and a signal representative of the deviations among the signals of each subgroup of said group are generated. Responsive to a subgroup deviation signal exceeding its group deviation signal, the subgroup signals are altered to reduce the subgroup deviations. In this manner errors are reduced without affecting error free signals.
Claims13
An error reducing circuit comprising means for receiving an input signal, means for dividing said signal into a plurality of segments, means for partitioning each signal segment into subsegments, means responsive to each signal segment for generating a first signal representative of the deviations of said input signal within said signal segment, means responsive to each signal subsegment of said signal segment for generating a second signal representative of the deviations of said input signal within said subsegment, means for generating a third signal for each subsegment corresponding to the difference between said subsegment second signal and said segment first signal, means for generating a predetermined threshold signal, and means responsive to said difference signal exceeding said threshold signal for forming a predetermined alteration of said subsegment signal.
An error reducing circuit comprising means for receiving a sequence of signals, means for dividing said sequence of signals into a plurality of groups, means for dividing each group of signals into a plurality of subgroups, means responsive to the signals of each group for generating a first signal representative of the deviations among the signals of said group, means responsive to the signal of each subgroup for generating a second signal representative of the deviations among the signals of said subgroup, means for generating a third signal for each subgroup corresponding to the difference of the magnitude of the second signal of said subgroup and the magnitude of the first signal of the group including said subgroup, means for producing a predetermined threshold signal, and correcting means jointly responsive to said third signal and said threshold signal for forming a predetermined alteration of said subgroup signals.
An error reducing circuit comprising means for receiving a sequence of signal codes, means for dividing said sequence of signal codes into a plurality of groups, means for dividing each group of signal codes into a plurality of subgroups, means responsive to the signal codes of each group for generating a first signal representative of the deviations among the signal codes of said group, means responsive to the signal codes of each subgroup for generating a second signal representative of the deviations among the signal codes of said subgroup, means for generating a third signal for each subgroup corresponding to the difference between the second signal of said subgroup and the first signal of the group including said subgroup, means for producing a predetermined threshold signal, and correcting means responsive to said third signal being greater than said threshold signal for replacing the corresponding subgroup signal codes with one of its adjacent subgroup signal codes.
An error detection and correction apparatus for use with quantized information receivers which comprises:
- means for receiving a series of quantized signals;
- means for partitioning said series of quantized signals into a sequence of signal blocks;
- means for generating a first signal representative of the successive difference of adjacent quantized signals;
- means for generating a second signal representative of the root mean square value of said sucessive differences of said adjacent quantized signals over each signal block;
- means for scaling said root mean square value signal;
- means for comparing each of said successive differences of adjacent quantized signals of one of said signal blocks with said scaled root mean square value corresponding to said one of said signal blocks;
- and
- means for replacing one of said quantized signals with one of its adjacent quantized signals whenever one of said first signals exceeds said second signal for one of said signal blocks, whereby the adjacent quantized signal chosen corresponds to the middle value of said one quantized signal and its two adjacent quantized signals.
An error detection and correction apparatus for use with quantized information receivers which comprises:
- means for receiving a series of quantized signals;
- means for partitioning said series of quantized signals into a sequence of signal blocks;
- means for partitioning said signal blocks into a plurality of subblocks;
- means for generating a block correlation signal corresponding to the correlation of said quantized signals within each of said signal blocks;
- means for producing a scaled block correlation signal;
- means for generating a subblock correlation signal corresponding to the correlation of said quantized signals within each of said subblocks;
- means for comparing each of said subblock correlation signals with its corresponding block correlation signal;
- and
- means for altering said quantized signals within said subblocks whenever said subblock correlation signal is less than said corresponding scaled block correlation signal, whereby said alteration causes said quantized signals within said subblock to be replaced by the quantized signals within an adjacent subblock of said block.
A method for reducing the error content of a received sequence of signals which comprises:
- dividing said sequence of signals into a plurality of groups;
- partitioning each of said groups into subgroups;
- generating a first signal for each group representative of the deviations among the signals of said group;
- multiplying said first signals by a constant to produce a threshold signal for each group;
- generating a second signal for each subgroup representative of the deviations among the signals in said subgroups;
- generating a third signal for each subgroup corresponding to the difference of said second signal of said subgroup and said first signal of the group including said subgroup;
- and
- selectively altering said subgroup signals responsive to said third signal exceeding said threshold signal to produce a second signal which is closer in value to said first signal.
Dependencies are inferred from the claim text.