Decode data using a Reed-Solomon decoder
Block sublibrary of Error Correction and Detection
Reed-Solomon encoding follows the same standards as any other cyclic redundancy code. The Integer-Output RS Decoder HDL Optimized block can be used to model many communication system Forward Error Correcting (FEC) codes.
For more about the Reed-Solomon decoder, see the Integer-Output RS Decoder block reference. For more information on representing data for Reed-Solomon codes, see Integer Format (Reed-Solomon Only).
The Integer-Output RS Decoder HDL Optimized block has four input ports and six output ports (5 required, 1 optional).
Port | Direction | Description | Data Type |
---|---|---|---|
dataIn | Input | Message data, one symbol at a time. | Integer or fixdt with any binary point
scaling. double is allowed for simulation but not
for HDL code generation. |
startIn | Input | Indicates the start of a frame of data. | Boolean
|
endIn | Input | Indicates the end of a frame of data. | Boolean
|
validIn | Input | Indicates that input data is valid. | Boolean
|
dataOut | Output | Message data with the checksum appended. The data width is the same as the input data port. | Same as dataIn |
startOut | Output | Indicates the start of a frame of data. | Boolean |
endOut | Output | Indicates the end of a frame of data, including checksum. | Boolean
|
validOut | Output | Indicates that output data is valid. | Boolean
|
errOut | Output | Indicates the corruption of the received data when error is high. | Boolean |
numErrors | Output (optional) | Count of detected errors. | uint8 |
Each input frame must contain more than (N-K)*2
symbols, and fewer than or equal to N
symbols. A
shortened code is inferred when the number of valid data samples between
startIn
and endIn
is less than
N
. A shortened code still requires
N
cycles to perform the Chien search. If the input is
less than N
symbols, leave a guard interval of at least
N-size
inactive cycles before starting the next
frame.
The decoder can operate on up to 4 messages at a time. If the block receives the start of a fifth message before completely decoding the first message, the block drops data samples from the first message. To avoid this issue, increase the number of inactive cycles between input messages.
The generator polynomial is not specified explicitly. However, it is
defined by the code word length, the message length, and the B value for the
starting exponent of the roots. To get the value of B from a generator
polynomial, use the genpoly2b
function.
The length of the code word in symbols, N
, must be
equal to 2M-1
, where
M
is the input word length. M
can
be between 3 and 16 bits.
The message length in symbols, K
. The number of parity
symbols, N-K
, must be a positive even integer, greater
than or equal to the input word length, M
.
Select Property
to enable the
Primitive polynomial parameter.
Binary row vector representing the primitive polynomial in descending order of powers. When you provide a primitive polynomial, the number of input bits must be an integer multiple of K times the order of the primitive polynomial instead.
This parameter applies when only when Property
is selected for Primitive polynomial.
Select Property
to enable the B
value parameter. When you select
Auto
, the block uses B =
1
.
The starting exponent of the roots.
This field is available only when you select
Property
for Source of B, the
starting power for roots of the primitive polynomial. The
default is 1.
Check this box to enable the numErrors
output port.
This port outputs the detected symbol error count.