A variable of the MATRIX
data type encapsulates a two
dimensional matrix of 32bit floatingpoint values (corresponding to
the IEC 611313 data type REAL
) in [row][column]
format.
Literal values
Literal values of MATRIX
variables can be expressed in
BNF syntax as follows:
The syntax of float_value
is any string accepted by the
java.lang.Float.valueOf(String)
method.
Each vector_literal
element represents a row of
the matrix; for instance, the literal [[1,2],[3,4]]
represents the matrix
Missing elements are assigned zero values; for instance, the literal
[[],[1,2],[3,4,5]]
represents the matrix
For examples of the use of MATRIX
literals, see the
MATRIX_DEMO
and M_OPS_DEMO
system configurations.
ASN.1 Encoding
MATRIX
data is transferred according to the following
modified ASN.1 syntax:
The encoding of MATRIX
elements is constructed
in the sense of ISO/IEC 8825, with the following extensions to
the COMPACT encoding described in Annex E of IEC
614991:

rows
andcols
, representing the number of rows and columns in the matrix respectively, are encoded as values of theUSINT
type without identifier or length octets, i.e., as two successive 8bit unsigned integers.
This restricts the maximum number of elements of a
MATRIX
to 65025 (255×255). 
If the matrix data is to be transferred via
PUBLISH/SUBSCRIBE
, the maximum number of elements is further restricted by the maximum number of octets (1472) that can be transferred in a single UDP packet without segmentation; otherwise, anOVERFLOW
error will be declared by the corresponding FB instance. 
Since the encoding of a
REAL
value requires 32 bits (4 octets), the maximum number of elements in aMATRIX
instance to be transmitted viaPUBLISH/SUBSCRIBE
would then be (maximum octets  tag octets  length octets)/(element length) = (147212)/4 = 367 elements. For instance, a square matrix with 19 rows and columns, containing 361 elements, would fit in a UDP packet. 
MATRIX
data larger than the allowed UDP transfer should be transferred using theCLIENT/SERVER
mechanism.

This restricts the maximum number of elements of a

The matrix contents are encoded using the COMPACT syntax for an
ARRAY of IEC 611313
REAL
values, without any tag or length subfields. 
Data is encoded in column order; for instance, the matrix
would be encoded in the order
For an example of the application of these encoding rules, see the CODEC_TEST2
system configuration.