Error detection/correction and fault detection/recovery – Pulse or data error handling – Data formatting to improve error detection correction...
Reexamination Certificate
1999-03-29
2001-03-13
Baker, Stephen M. (Department: 2133)
Error detection/correction and fault detection/recovery
Pulse or data error handling
Data formatting to improve error detection correction...
Reexamination Certificate
active
06202178
ABSTRACT:
The present invention relates to a method for interleaving data frames as described in the preamble of claim
1
, a forward error correcting device which performs this method as described in the preamble of claim
6
, and a modulator including such a forward error correcting device as described in the preamble of claim
7
.
Such a method and equipment to perform such a method are already known in the art, e.g. from the contribution to assist the Standards Committee T1-Telecommunications with reference T1E1.4/95-075. This contribution is entitled ‘How to use ADSL for more than 8 Mbps data?’, and is filed by Orckit Communications Ltd. Therein, a method for interleaving data frames, called interleave data frames, is proposed which enables ADSL (Asymmetric Digital Subscriber Line) transmission at bit rates higher than 8,16 Mbps. The proposed method is based on the insight that the specification in the draft American National Standard for Telecommunications on ADSL (Asymmetric Digital Subscriber Line) published by ANSI (American National Standards Institute) in April 1995, limiting the interleave frame size to the length of one codeword limits the transmission rate to a maximum of 8,16 Mbps. This transmission rate limitation is avoided by optionally including more than one codeword in one interleave data frame. Interleaving is discussed at paragraph 6.4.2, wherein it is stated that the Reed-Solomon codewords in the interleave buffer are convolutionally interleaved to a depth D that varies and that is defined by the rule that each of the n bytes B
0
, B
1
, . . . , B
m-1
in a Reed-Solomon codeword is delayed by an amount that varies linearly with the byte index. More precisely, byte B
i
(with index i) is delayed by (D−1)* i bytes where D is the interleave depth. In the known method, as described in the last paragraph on page 2 of the above mentioned contribution, an interleave data frame contains two codewords, and in a first step is split into these two codewords which may have different lengths. Each of the codewords in a further step is extended with an overhead extension, called FEC redundancy in the already cited contribution. The overhead extensions added to codewords with different lengths can also have different lengths. The so obtained extended codewords, called Reed-Solomon codewords in the above cited contribution, are then joined to constitute an extended interleave data frame which is written in an interleave buffer whose memory cells are fixed in a matrix-shaped structure in such a way that each extended interleave data frame occupies one column in the matrix-shaped structure. Since successive Reed-Solomon codewords in the known method can have different lengths, the required flexibility for the overhead adding means in a forward error correction device which is enabled to perform the known method, renders this overhead adding means more complex. Furthermore, in the known solution, the number of columns of the matrix-shaped structure in the interleave buffer remains unchanged when compared to interleaving techniques which are in accordance with the specifications in the above cited draft ADSL Standard and wherein each interleave data frame contains thus only one codeword. As a result, the interleave depth and correction capability for burst errors remains unaffected.
An object of the present invention is to provide a method and equipment for interleaving data frames in such a way that high bitrates, i.e. bitrates higher than 8,16 Mbps for ADSL applications, are achieved without significant complexity increase of the interleaving means but with optimized interleave depth, i.e. with optimized correction capability for burst errors.
According to the invention, this object is achieved by the method, forward error correcting device and modulator described in claim
1
, claim
6
and claim
7
respectively.
Indeed, in the present method, codewords and overhead extensions are not allowed to have different lengths. Consequently, the complexity of the overhead adding means included in a forward error correction device which is further equipped with an interleaving device according to the present invention, is not increased when compared to the situation wherein only one codeword constitutes a data frame. Furthermore, since each codeword occupies another column in the matrix-shaped structure of the interleave buffer, the total number of columns compared to the known solution, is multiplied by a factor equal to the number of codewords which constitute one data frame. Evidently, since the matrix-shaped structure is filled column by column in step c and read out row by row in step d, the interleave depth, i.e. the maximum length of a burst error which disturbs less than two data bytes belonging to one codeword, is multiplied by the same factor when compared to the known method described in the contribution of Orckit Communications Ltd.
It is noticed that compared to the known method for interleaving described in the above mentioned Orckit contribution, an increase of the interleave depth of the same amount as in the present invention is obtained in another method for interleaving data frames, proposed by AMATI in its contribution to assist the Standards Committee T1-Telecommunications with reference T1E1.4/95-065, entitled ‘High Rate (more than 8 Mbps) ADSL Frame Format with Multiple Reed-Solomon Codewords’. The frame structure proposed by Amati provides an additional level of interleaving (with depth
2
) in addition to the standardized convolutional interleaving. Therein, each data frame, also called interleave frame, is again allowed to contain multiple codewords. However, these codewords are generated from an interleave frame in a first step of the method by separating bytes with odd and even indexes in this interleave frame. Consequently, to be able to perform this first step, a forward error correction device which performs the method proposed by AMATI has to be provided with means which separate bytes with odd and even indexes. Moreover, additional memory means have to be included in the forward error correction device to temporarily store therein the codeword of odd data bytes whilst the overhead adding means is extending the codeword of even data bytes, or to temporarily store therein the codeword of even data bytes whilst the overhead adding means is extending the codeword of odd data bytes. If such an additional memory means is not provided, both codewords have to be processed simultaneously by two parallel coupled overhead adding means. The extended codewords are again called Reed-Solomon codewords in the AMATI contribution. Compared to the present invention, wherein codewords are built up from successive data bytes in a data frame and wherein successive codewords are thus extended successively by one overhead adding means without the necessity to temporarily store any codeword, the method proposed by AMATI in their contribution T1E1.4/95-065 requires the use of a more complex forward error correction device.
In a particular implementation of the present method, as described in claim
2
, the data bytes at the transmitter are written into the matrix column by column and are read out of this matrix row by row. It should be noted however that the present method is not restricted to a specific way of writing data bytes into the matrix or reading data bytes out of the matrix since it is clear to a person skilled in the art how to modify the later described embodiment of the present invention to obtain implementations with different write/read schemes for the matrix.
An additional characteristic feature of the present method is that dummy bytes are added to the data frames as described in claim
3
.
In this way, the length of the incoming data frames is adapted in an artificial way so that it can be divided in codewords of equal length in step a. Such dummy bytes furthermore may be added to data frames in implementations of the present method wherein the number of columns and number of rows in the matrix-shaped structure have to be coprime, i.e. implementation
ALCATEL N.V.
Baker Stephen M.
LandOfFree
Complementary methods for interleaving and deinterleaving... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Complementary methods for interleaving and deinterleaving..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Complementary methods for interleaving and deinterleaving... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2545930