Internet Standards/CD2.mdf/ccitt/1992/t/t150_1.asc

Text File | 1993-08-16 | 72KB | 3,424 lines

C:\WINWORD\CCITTREC.DOT_______________

The drawings contained in Recommendation have been done in Aautocad.

Recommendation T.150

 TELEWRITING TERMINAL EQUIPMENT

TABLE OF CONTENTS

This Recommendation consists of four parts, combined in one document

SCOPE

PART 1 ù Fundamental characteristics

  1    Introduction

  2    Definitions

  3    References

  4    Presentation functionalities

  5    Principles of telewriting coding

PART 2 ù Telewriting together with telephony

   1    General

   2    Main characteristics of the basic terminal

   3    Presentation functionalities of the basic terminal

   4    Transmission for the basic terminal

   5    Transmission blocks

   6    Transmission procedure

   7    Coding identifier

   8    Communication control, general requirements

   9    Communication control commands

  10    Description of the communication process

PART 3 ù Zone coding

   1    General

   2    Presentation elements

   3    Zone coding description

   4    Specification of terms used in coding

   5    Specification of the coding

   6    A coding example

   7    Data structure

   8    Temporary penùstop

   9    Control commands

  10    Summary code table

  11    Summary transmission data format

  12    Zone coding basic terminal

PART 4 ù Differential chain coding

   1    General

   2    Presentation elements

   3    Description of the coding

   4    Incremental mode mechanism

   5    Change of coding parameters

   6    Coding formats

   7    Incremental mode coding format

   8    Displacement mode coding format

   9    Encoding of the primitives

  10    Example of differential chain coding

1    Scope

    This Recommendation specifies techniqueùoriented characteristics
of telewriting and the application of telewriting in combination with voice
communication. Serviceùoriented requirements are defined in Recommen-
dation F.730. In the development of this Recommendation, compatibility
with other telematic services is taken into account. This Recommendation is
structured in four parts:

Part 1 ù    Fundamental characteristics

Part 2 ù    Telewriting together with telephony

Part 3 ù    Zone coding

Part 4 ù    Differential chain coding.

Part 1 ù Fundamental characteristics

1    Introduction

1.1    Telewriting is a communication technique that enables the exchange
of handwritten information through telecommunication means. The hand-
written information may consist of text in handwriting, drawings, diagrams,
etc.

1.2    By means of telewriting terminal equipment, the TRACE of the writ-
ing instrument as produced at the sending side, is reproduced at the receiv-
ing side including the effect of movement.

1.3    In the sending part of the terminal the handwritten input information
is converted into a digital signal: the coded representation of the handwrit-
ten information. Next, this digital signal is converted into a signal suitable
for transmission.

1.4    In the receiving part of the terminal the received signal is converted
into a digital signal, corresponding with the coded representation as
described in above. From this digital signal, the handwritten information is
reproduced.

1.5    The reproduction of the handwritten information can take place on a
screen, on paper or both. In this Recommendation, the characteristics of
communication through telewriting are defined with respect to the image on
a screen (soft copy). Reproduction on paper (hard copy) is considered to be
an optional function under local control.

1.6    Storage may take place between the writing (the input process) and
the reproduction (the output process). When retrieved from a store, the mes-
sage will appear on the receiver's screen in the same way as in the case of a
direct connection.

1.7    A page of handwritten information (or part of it) could be reproduced
as a still picture. This application, however, is not covered in the present
text.

1.8    Telewriting can be used in various ways:

ù    as independent communication technique,

ù    in combination with voice communication through a telephone net-
work,

ù    in the context of teleconferencing,

ù    in the context of information retrieval.

2    Definitions

2.1    telewriting image

    A collection of telewriting presentation elements, to be displayed
together.

    Note ù The telewriting image can exist in visible form at the output
device, or in the form of a coded representation.

2.2    presentation element

    Basic graphic element used to construct an image.

    Examples of telewriting presentation elements are: trace, closed area,
background.

2.3    coding rectangle

    Rectangular area representing the coding space in horizontal and ver-
tical direction, available for coding of a telewriting image.

2.4    image area

    (previously: text area)

    Rectangular part of the display area, to be considered as the image of
the coding rectangle.

2.5    background

    Presentation element being a rectangular area with the same size as
the image area, acting as a reference area on which telewriting foreground
information can be presented.

2.6    trace

    Presentation element being a curve of an arbitrary shape, starting
from a defined position, being completed incrementally and ending at a
defined position.

2.7    closed area

    Presentation element being an area enclosed within one trace which
constitutes a closed line.

2.8    marker

    Marked representation of a single position in a telewriting image.

    Note ù A marker is not a permanent part of a telewriting image, but
exists only as long as it is activated.

2.9    attribute

    A particular property which applies to a presentation element or to a
group of presentation elements.

    Examples: line thickness, colour.

3    References

    In the text of this Recommendation the following Recommendations/
standards are referred to:

ù    Rec. F.730: Service oriented requirements for telewriting applica-
tions.

ù    Rec. T.101: International interworking for videotex services;
Annex C, data syntax II.

ù    Rec. V.21: 300 bits per second duplex modem standardized for use
in the general switched telephone network.

ù    ISO 9281: Information processing ù Identification of picture cod-
ing methods.

4    Presentation functionalities

4.1    This section describes a set of presentation functionalities. This set of
functionalities is intended as a repertoire of presentation functionalities for
telewriting in general. For a specific application a subset may be defined.

4.2    In the description of presentation functionalities, the concept of
TRACE is being used. A trace is a curve of an arbitrary shape, starting from
a defined position, being completed incrementally and ending at a defined
position. Handwritten information is considered to consist of traces.

4.3    Representation of the handwritten information is accomplished by the
sequential reconstruction of the individual traces. This implies that the
effect of movement is retained during each reproduction.

4.4    Telewriting information is to be displayed on the display area of some
output device. The display area is considered to be a twoùdimensional sur-
face.

4.5    The display area is subdivided into an image area and a border area;
see Figure 1ù1/T.150.





Image area





Image
area



 FIGURE 1ù1/T.150

Subdivision of display area

4.6    The border area surrounds the image area. External form and dimensions
of the border area are not specified. The presence of a border area is not
mandatory. It is however inevitable in certain implementations.

4.7    The image area is rectangular. The two shorter edges of the image area
have a vertical orientation, the two longer edges have a horizontal orienta-
tion. The length ratio of shorter and longer edges is 3:4.

4.8    The position of telewriting information on the display area is defined
with respect to the edges of the image area.

4.9    Information on the display area is composed of presentation elements of
three categories:

ù    foreground,

ù    background,

ù    border area.

4.10    Foreground and background presentation elements are defined in the
image area only.

    Border area presentation elements are defined in the border area only.
The use of the border area is not defined for telewriting.

4.11    Foreground presentation elements include trace, marker and closed
area.

4.12    The presentation elements have the following characteristics:

ù    Trace : This is the curve as defined in º 2.6 of this part; the essence
of the handwritten information is represented by one trace or by
any combination of traces; the image area can contain an unde-
fined number of traces at a time.

ù    Marker : This is a marked representation of a single position; it
behaves as if it is overlaid on the foreground; a moving marker
does not create a trace; a marker can be switched on and off; one
user can generate only one marker at a time. The image area can
contain one locally generated marker and one remotely generated
marker.

ù    Closed area : This is the area that is enclosed within a closed trace;
this closed trace is the perimeter. A trace is a closed trace if it inter-
sects itself; a trace that is nearly closed can be converted into a
closed trace, by the addition of the lacking part of the trace.

ù    Background : The background is a defined reference area on which
foreground information is to be imaged; if the full image area is
filled with foreground information, the background is not visible.

ù    Border area : The border area is independent of the information in
the image area.

In case of a CRT display the border area is the remaining part
between image area and edges of the display area.

In case of a cellùstructured display device, the image area may
coincide exactly with the display area. In that case no border area
remains.

4.13    The various presentation elements can have attributes assigned to
them as defined in Table 1ù1/T.150.

TABLE 1ù1/T.150

Attributes of telewriting presentation elements





Presentation element

 Attributes

Trace

Line thickness, line texture, colour

Marker

Shape, size, colour

Closed area

Area texture, colour (interaction or
area attributes with background
attributes to be defined)

Background

Area texture, colour

Border area

Not defined

             Note ù The concept of colour includes ôintensityö.

4.14    Once an image is displayed, subsequent modification of attributes is
restricted as follows:

ù    trace: attributes unchangeable;

ù    marker: attributes can be changed at any instant;

ù    closed area: attributes unchangeable;

ù    background: attributes can be changed at any instant.

4.15    In case of intersection of two traces, the image of the older trace is
interrupted as far as it coincides with the newer trace.

4.16    In case of intersection of a trace and a marker, the image of the trace is
interrupted as far as it coincides with the marker. After removal of the
marker, the image of the original trace is restored.

4.17    With respect to erasure of foreground information, a distinction is
made regarding the area in which erasure takes place:

ù    full image area;

ù    defined part of the image area;

ù    individual traces.

4.18    Erasure of the full image area

    All foreground information in the image area is removed; the back-
ground assumes a preùdefined appearance.

4.19    Erasure of a defined part of the image area

    An area is identified either by means of a closed trace or as a defined
square, within which all foreground information is to be removed including
the perimeter itself.

4.20    Erasure of individual traces

    An existing trace is covered by a thicker trace with the same attributes
as the background: this type of erasure is processed in the same way as a
trace.

4.21    Any modification of background information can take place for the
full image area only.

5        Principles of telewriting coding

5.1    Telewriting coding relates to coding of telewriting information in
foreground and background and to erasure functions.

5.2    This section contains principles of telewriting coding. In Parts 3 and
4, details of telewriting coding are defined for two methods, namely zone
coding and differential chain coding, respectively.

5.3    The coding is defined at the ôtelewriting coding interfaceö, TCI. This
interface is introduced for ease of reference, but need not exist physically.

5.4    In the sending part of the telewriting terminal, the signal at the TCI
contains all data originating from handwritten input, selection of attributes
and use of erasure functions.

5.5    The signals at the TCI, both in sending and receiving parts, do not
contain data pertaining to transmission or communication functions.

5.6    In the receiving part of the telewriting terminal, the signal at the TCI
contains all data required to image the information in accordance with the
intentions of the originator.

5.7    The concept of the TCI is illustrated in Figure 1ù2/T.150.

Fig. 1ù2/T.150/T0803750-89 = 7 cm



5.8    The signal at the TCI includes x and y coordinate information regarding
telewriting presentation elements.

5.9    The x and y coordinates are related to a unit area of 1 ╫ 1. This implies
that the respective values of x and y always lie between 0 and 1 (0 included,
1 not included).

5.10    The origin of the coordinate system is in the lower left corner. The xù
axis is horizontal, the yùaxis is vertical.

5.11    The horizontal size of the telewriting image area corresponds with x =
1, the vertical size of this image area corresponds with y = 0.75. See Figure
1ù3/T.150.





y = 1

 1, 1

 y = 0.75

 Image area

 0, 0

x = 1

FIGURE 1ù3/T.150

Position of image area within unit area

5.12    All coordinates of the telewriting information are quantized relative to a
measurement grid in the unit area. The resolution of this grid determines the
accuracy.

5.13    The default resolution is 512 ╫ 512 grid units. The telewriting coding
can optionally also accommodate grid resolutions of 1024 ╫ 1024 and 2048 ╫
2048 grid units.

Part 2 ù Telewriting together with telephony

1    General

1.1    This part of the Recommendation defines the use of telewriting in
combination with voice communication through a telephone network
(PSTN).

1.2    For this application, both sides of the connection must have a com-
bined telephone and telewriting terminal.

1.3    The combined telephone and telewriting terminal should, as long as
the telewriting transmission function is switched off, behave like a normal
telephone set, both for incoming and outgoing calls. In this situation, the full
bandwidth is available for transmission of speech signals.

1.4    During a telephone conversation, the telewriting transmission func-
tion at either side of the connection, may be switched on and off, manually
or automatically.

1.5    Remark that in this part of the Recommendation ôswitching on and
offö of the telewriting function refers to the telewriting transmission func-
tions. Regardless of this, the telewriting equipment may be used locally,
whether or not a telephone connection exists.

1.6    By means of the telewriting terminal, the user can generate informa-
tion. This includes: creation of traces, marker switching on and off, move-
ment of the marker, use of erasure functions.

1.7    In this part, distinction is made between ôbasic terminalö and
ôenhanced terminalö.

1.8    The enhanced terminal is not defined yet, but compared to the basic
terminal it is anticipated to have additional capabilities regarding unat-
tended operation, transmission facilities and presentation functionalities.

2    Main characteristics of the basic terminal

2.1    In this section, a basic terminal is defined.

    In the basic terminal a set of functions is implemented that is to be
considered as a minimum requirement; thus a basic level of compatibility is
defined.

2.2    A basic terminal includes a telephone apparatus, a writing device and
a display device. Circuitry to implement control functions may be accom-
modated in a separate unit or may be included in one of the devices men-
tioned.

2.3    Information, generated at either side of the connection will be repro-
duced on the display devices at both sides of the connection.

2.4    Both sides of the connection can contribute, one after another, to the
same image.

2.5    In the basic terminal, transmission of telewriting signals is accom-
plished through a subùchannel, segregated from the speech channel. Trans-
mission of speech signals and telewriting signals can take place
simultaneously.

2.6    Halfùduplex transmission is used for conveying the telewriting sig-
nals through the subùchannel, i.e. the transmitter is prevented from sending
as long as the associated receiver receives telewriting signals from the other
side.

2.7    The total power level of speech plus telewriting signals should con-
form to the limits normally applicable to speech transmission and data trans-
mission.

2.8    The basic terminal can assume three modes of operation. The charac-
teristics pertinent to each mode, are described in Table 2ù1/T.150.

TABLE 2ù1/T.150

Modes of operation of the basic terminal



Speech only

The telewriting function remains in the
OFF condition.

Speech plus telewriting

The telewriting function can be switched
ON after the establishment of a connec-
tion. Speech signals and telewriting sig-
nals can be sent simultaneously.

Telewriting only

This mode can be switched ON after the
establishment of a connection. The send-
ing of speech signals is blocked, the
power level of the telewriting signals is
increased correspondingly. Reception of
speech signals is still possible.

2.9    In this Recommendation, the expression ôtelewriting ONö is used as a
common indication for either ôspeech plus telewritingö or ôtelewriting
onlyö.

2.10    A basic terminal may be able to continue transmission and reception of
telewriting signals after termination of the human conversation. In this case,
the telewriting transmission function will be switched OFF automatically
after completion of the telewriting transmission. (Defined in more detail
later on.)

2.11    For the coding of telewriting information, two methods are recognized
for use at the sending side: tone coding (defined in Part 3) and differential
chain coding (defined in Part 4).

    At the receiving side, the basic terminal should be able to properly accept
telewriting signals coded according to either method.

3    Presentation functionalities of the     basic terminal

3.1    The general description of presentation functionalities, as given in
Part 1, º 4, applies.

    With respect to this general description certain restrictions apply, as
defined in the following points.

3.2    The presentation functionalities as described for the basic terminal are
to be regarded as default capabilities.

    If required, characteristics of terminals with a higher level of sophisti-
cation will be described in a section on enhanced terminal.

3.3    The basic terminal employs a monochrome display device. The writ-
ing device generates coded representations of monochrome images only.

3.4    The attributes applying to the basic terminal are given in Table 2ù2/
T.150.

TABLE 2ù2/T.150

Attributes applying to the basic terminal



Presentation elements

 Attributes

Image size

 Horizontal: 512 GU
Vertical: 0.75 ╫ 512 GU
Options, the receiver must be able to accept:
Horizontal: 1024 and 2048 GU
Vertical: 0.75 ╫ 1024 and
    0.75 ╫ 2048 GU.

Trace

ù thickness

Unit thickness, as used in the output device.
Options: 2 ╫ and 3 ╫ unit thickness.

ù texture

 Solid, no options.

ù colour

 Monochrome, as used in the output device. The
receiver must be able to accept the codes of
traces with colours: red, green, blue, yellow,
magenta, cyan, white, black. A black trace has
the same colour as the background (used for era-
sure).

Closed area

ù texture

 Solid.

ù colour

 Same as background colour (used only for par-
tial erasure). The receiver must be able to accept
the codes of closed areas with colours: red,
green, blue, yellow, magenta, cyan, white, black.

Background

ù texture/colour

 No information about the background is trans-
mitted. Background can only be imagined as
dark screen. This corresponds with colour black.

Border area

 Border area is not specified, no information
about the border area is transmitted.

Marker

ù shape

 PLUS sign; other shapes may be possible
depending on terminal implementation.

ù size

 Not specified.

ù colour

 Marker colour is not transmitted; on a mono-
chrome device the marker appears in foreground
colour; on a colour device the marker may
assume a colour under local control.

Full erasure

 Black background is restored.

Partial erasure

 1)   closed area;

 2)   overwriting with thicker black trace.

GU Grid units

4    Transmission for the basic terminal

4.1    Transmission of the modulated telewriting signal takes place in a
small frequency band, segregated from the speech channel. This band is
referred to as the subùchannel.

4.2    The centre of the subùchannel is located at 1750 Hz. Details of the
implementation are not given here, but the requirements of ºº 4.6 and 4.7
should be met.

4.3    The binary telewriting signal is converted into a signal suitable for
transmission, by means of frequency shift modulation. Details are the same
as those specified in Recommendation V.21 for channel 2 (the high chan-
nel).

4.4    The modulation rate is 300 Bd, the bit rate is 300 bit/s.

4.5    The V.21 requirements for channel 2 are summarized as follows: The
nominal mean frequency of the transmission signal is 1750 Hz. The fre-
quency deviation is + or ù100 Hz. Consequently, the nominal characteristic
frequencies are 1850 Hz and 1650 Hz respectively. The higher frequency
corresponds to a binary 0.

4.6    The amount of speech signal power that can reach the local and
remote telewriting receivers, should be sufficiently low to avoid errors in
the demodulated telewriting signal.

4.7    The amount of telewriting signal power that can reach the local and
remote telephone receivers (i.e. the loudspeaker part) should be sufficiently
low to avoid disturbance of the conversation.

4.8    In the mode of operation ôtelewriting onlyö, the output power of the
telewriting transmitter shall be in accordance with the requirements
described in Recommendation V.21.

4.9    In the mode of operation ôspeech plus telewritingö, the modulated
Telewriting signal should be attenuated by 4 dB with regard to the level
determined by º 4.8. If experience shows that also the power of the speech
signal should be adapted, relevant requirements will be included in the next
issue of this Recommendation.

4.10    In the case of longùdistance communication an echo suppressor may
be present in the link. This will hamper the ôspeech plus telewritingö mode.
Since, generally, disabling of the echo suppressor cannot be guaranteed to
solve the problem, it is recommended to use the ôtelewriting onlyö mode,
alternating with the ôspeech onlyö mode.

4.11    The telewriting data as well as communication control commands are
structured in 8ùbit bytes.

    For transmission, each byte is packed in an 11ùbit transmission word
as defined below.

4.12    The structure of each transmission word is as follows:

1startbit, binary value ZERO

8 bits representing telewriting or control data

1 parity bit

stopbit, binary value ONE.

    This structure is illustrated in Figure 2ù0/T.150.





Start

Data

Parity

 Stop

1 bit

8 bits

1 bit

1 bit

 FIGURE 2ù0/T.150

Structure of a transmission word

4.13    For the value of the parity bit, EVEN parity applies. This Recommenda-
tion does not specify any action for the basic terminal in case of reception of
an erroneous parity bit.

4.14    The transmission words are conveyed in startùstop mode, i.e. the
pause following a transmission word until the occurrence of the next trans-
mission word, may in principle have any duration. However, the bits consti-
tuting the transmission word should be transmitted as a contiguous sequence
at the appropriate bit rate.

4.15    In addition to its task of transporting bits, the data send signal may
assume one of three possible states:

ù    MARK signal: a binary ONE condition, with a duration signifi-
cantly longer than a bit period.

ù    SPACE signal: a binary ZERO condition; this condition is not used
in the framework of this Recommendation.

ù    Carrier OFF: no send signal present.

5        Transmission blocks

5.1    To define the transmission structure, the concept of transmission
block is introduced. In the general case, a transmission block contains trans-
mission words and MARK signals. However, also transmission blocks con-
taining MARK signals only may occur.

5.2    The beginning of a transmission block is identified by the occurrence
of one out of two defined combinations of MARK signal and carrier OFF
condition, referred to as start combination No. 1 and start combination No.
2.

5.3    The start combinations are defined as follows:

ù    start combination No. 1    carrier OFF during at least 130 ms, fol-
lowed by

                MARK signal of 100 ▒ 20 ms followed by

                carrier OFF during 100 ▒ 20 ms followed by

                MARK signal of 200 ▒ 20 ms.

ù    start combination No. 2    carrier OFF during at least 130 ms, fol-
lowed by

                MARK signal of 400 ▒ 20 ms.

    See illustration in Figure 2ù1/T.150.

    The use of these start combinations is defined later.

Figure 2ù1/T.150/T0803760-89 = 7 cm



5.4    Immediately following the start combination of a transmission block,
one of the following signals should be sent:

ù    a MARK signal

ù    a single transmission word

ù    a sequence of transmission words.

    Between any two subsequent transmission words, a MARK signal
may occur, representing a pause in the writing process.

5.5    Every transmission block is terminated by a MARK signal of 500 ▒
20 ms. The MARK signal is to be followed by a carrier OFF condition of at
least 130 ms.

5.6    The MARK signals representing pauses may have various durations,
determined as follows:

ù    during PEN DOWN and absence of other telewriting activity, the
MARK signal may continue without limitation;

ù    after PEN UP the terminal will apply a limit of 500 ▒ 20 ms. Within
this limit the telewriting activity may continue without procedural
steps. If the limit expires, the carrier will be switched OFF. Thus
the transmission block is automatically terminated by the terminal.
Sending of further data requires the start of a new transmission
block.

5.7    The periods between transmission blocks are indicated by carrier OFF
conditions.

5.8    The formats of transmission blocks are summarized in Figure 2ù2/
T.150.

Fig. 2ù2/T.150/T0803770-89 = 4 cm



6    Transmission procedure

6.1    Prior to actually sending telewriting data, the terminal is to decide
whether it functions in MASTER mode or in SLAVE mode.

    In case of a transmission collision, the master terminal has transmis-
sion privilege over slave terminals.

6.2    The terminal decides about the master/slave status by sending the start
combination No. 1 and observing the received signal.

6.3    If the terminal, engaged in sending start combination No. 1, detects a
received carrier signal at its receiver input (during a carrier OFF interval) it
decides to be a slave and it postpones further attempts to send data. See Fig-
ure 2ù3/T.150.

6.4    If the terminal does not detect a received carrier signal during the
sending of the start combination, it decides to be a master and continues
sending. See Figure 2ù3/T.150.

6.5    In the case that only one terminal generates telewriting data, this ter-
minal assumes the master status. The receiving terminal remains in the slave
status.

6.6    As a header for the subsequent transmission blocks, a master terminal
uses start combination No. 2, a slave terminal uses start combination No. 1.
See Figure 2ù4/T.150.

6.7    The master/slave status decision in a given terminal remains valid
until it is cancelled as follows:

ù    A master terminal becomes a slave if it is not engaged in sending at
the moment that another terminal sends start combination No. 1.

ù    A slave terminal becomes a master terminal at the moment that it
sends a start combination No. 1 and no receive carrier signals are
being detected.

ù    A master status is cancelled by ôtelewriting OFFö.

Fig. 2ù3/T.150/T0803780-89 = 25 cm



Fig. 2ù4/T.150/T0803790-89 = 25 cm



7        Coding identifier

7.1    In the communication control procedures, the existing of two coding
methods is recognized, i.e. zone coding zone coding and differential chain
coding differential chain coding respectively.

    The method actually used is identified by the coding identifier PCE.
(PCE = picture control entity).

    A terminal receiving signals according to either method will be able
to activate the appropriate decoding function, by recognizing the coding
identifier.

7.2    The coding identifier is structured according to ISO 9281. In this stan-
dard, the coding identifier PCE is defined to comprise a picture coding
delimiter (PCD) and a coding method identifier (CMI). See Figure 2ù5/
T.150.





PCD


CM
I



PCE



 PCE    Picture control entity

PCD    Picture coding delimiter

CMI    Coding method identifier

FIGURE 2ù5/T.150

Structure of coding identifier

7.3    (Copy of ISO 9281, º 6.2.4 modified)

    The PCD shall announce or delimit the data for a particular picture coding
method. The PCD shall comprise the twoùbyte sequence 01/11, 07/00.

7.4    (Copy of ISO 9281, º 6.2.5)

    The CMI shall specify the particular coding method for the picture data that
follow it. The CMI may consist of one or more octets corresponding to the
bit combinations in the range 02/00 to 07/14 of an 8ùbit code table.

7.5    (Copy of ISO 9281, º 6.2.6)

    Each CMI identifying a particular picture coding method shall be registered
with the ISO Registration Authority for Picture Coding Methods (to be set
up).

7.6    The telewriting coding identifier, when included in a transmission block,
occupies the first three (or more if appropriate) transmission words follow-
ing the start combination. See Figure 2ù6/T.150.






Carrier
OFF


Start combi-
nation


Coding iden-
tifier

Telewriting
data and
MARK sig-
nals


MAR
K


Carrier
OFF

FIGURE 2ù6/T.150

Transmission format, including coding identifier

7.7    In a pointùtoùpoint configuration, the inclusion of the coding identi-
fier in the first transmision block only, would in principle be sufficient for
the whole session.

    However, for multipoint communication, the insertion of the coding identi-
fier in each transmission block is required.

    In view of this requirement, it is recommended that the coding identifier be
included in each transmission block containing telewriting data, irrespective
of the configuration.

7.8    The terminal should be designed such that transmission of the coding
identifier takes place automatically at the right moment.

7.9    For telewriting equipment according to this Recommendation T.150 the
following bit combinations should be used in the coding identifier. See
Table 2ù3/T.150.

TABLE 2ù3/T.150

Coding identifier bit combinations



Acronym

 Bit combination

PCD (2 byte sequence)

 01/11, 07/00

CMI Zone coding

 02/00, 04/00

CMI Diffùchain coding

 02/00, 04/01

    Note ù The above allocations are of a preliminary nature, pending
further development of ISO 9281.

8        Communication control, general requirements

8.1    This section defines requirements for the control of data exchange for
the basic telewriting terminal.

8.2    These requirements also apply to data exchange between any
enhanced terminal and a basic terminal.

8.3    The requirements permit the use of a twoùhop satellite circuit in the
connection between two terminals.

8.4    The requirements also permit multiùpoint communication via a
voice bridge.

8.5    Establishment and clearing of the telephone connection take place in
accordance with the requirements set by the telephone network.

8.6    For the basic terminal, automatic calling and answering are not
defined.

8.7    A basic terminal may, as an option, be equipped such that it can main-
tain the exchange of telewriting data after termination of the speech conver-
sation. This option is identified as ôautomatic call terminationö.

8.8    The automatic call termination implies that the telewriting function
(sending as well as receiving) is able to operate autonomously while the
telephone apparatus is in the ON HOOK condition.

8.9    To enable automatic call termination, the terminal must be able:

ù    to note that sending respectively reception of a telewriting trans-
mission block is going on, during the ON HOOK condition of the
telephone apparatus,

ù    to recognize the end of the final telewriting transmission block,

ù    to switch back to the speech only mode and to clear the telephone
connection.

8.10    Switching between the three modes ôspeech onlyö, ôspeech plus tele-
writingö and ôtelewriting onlyö can be done manually. In addition, switch-
ing the telewriting function OFF can take place automatically by means of
the communication control command SSO in the transmission signal. The
transitions between modes of operation are illustrated in Figure 2ù7/T.150.

Fig. 2ù7/T.150/T0803800-89 = 7 cm



9    Communication control commands

9.1    For control of the communication process, the commands SSO and
HLO are available.

    The coding of these commands is as follows:

SSO        1/7

HLO    0/5

    The meaning of these commands is described in Table 2ù4/T.150.

TABLE 2ù4/T.150

Communication control commands



Acronym

 Meaning

SSO

Set speech only

This command indicates that the terminals are
instructed to switch from telewriting ON to the
speech only mode

HLO

Hello

This command is to be sent by a terminal that
expects telewriting data, but does no receive such
data

9.2    A terminal will automatically send SSO upon the instruction by its
local user, to switch over from the telewriting ON mode to the speech only
mode.

    Transmission of SSO can take place in two ways:

ù    At the end of the current transmission block. SSO is attached to the
block, according the format defined below.

ù    By means of a separate transmission block. Such a block is sent
specifically for conveying SSO. Format: defined below.

9.3    A terminal receiving SSO will revert to the speech only mode and
does not recognize further telewriting signals.

9.4    The format for sending SSO is defined in Figures 2ù8/T.150 and 2ù
9/T.150.

Fig. 2ù8/T.150/T0803810-89 = 7 cm



Fig. 2ù9/T.150/T0803820-89 = 7 cm



9.5    The HLO command will only be sent in a block without telewriting data.
The format should be as defined in Figure 2ù10/T.150.

Fig. 2ù10/T.150/T0803830-89 = 7 cm



9.6    The HLO command is intended for use with automatic reception. This
command will be sent by a terminal if it has not received valid telewriting
data during a period of 35 seconds since:

ù    establishment of the telephone call;

ù    reception of the last valid transmission block.

9.7    The terminal receiving a HLO command responds with a MARK sig-
nal of 700 ▒ 20 ms.

9.8    A terminal in the telewriting ON condition, receiving signals other
than valid telewriting data (e.g. a tone from the telephone network) cannot
enter the send mode. In this case, the terminal returns to the speech only
mode without sending any command or other information (after a guardù
time of 35 seconds).

10    Description of the     communication process

10.1    In order to describe the full communication process, the concepts of
ôtelewriting activityö and ôteleù writing sessionö are introduced. These are
defined as follows:

ù            Telewriting activity ù Any action by the user that causes the
telewriting terminal (in the telewriting ON condition), to send data.
Examples of such actions are: pen down, marker ON, erasure.

ù            Telewriting session ù A period of time delimited by session
start and session end, during which two communicating terminals
have a relationship that enables them to exchange telewriting data.

10.2    The event determining session start is:

ù    the terminals are in the condition telewriting ON,

ù    at one of the terminals the first telewriting activity has occurred.

10.3    The event determining session end is:

ù    the terminals switch over to the telewriting OFF condition.

10.4    The session is established as soon as the coding identifier is received
and recognized by the receiving terminal.

10.5    At the beginning of the session, both terminals have the slave status.
During the session, only one terminal at a time can acquire the master status.

10.6    In the preceding text of this Part 2, all elements to be used in the com-
munication process are defined now.

    The process can be summarized as described in Table 2ù5/T.150.

10.7    The preceding description is given for a pointùtoùpoint configura-
tion. However, taking into account that only one terminal can have the mas-
ter status, this description is applicable to a multipoint configuration as well.
In this case it is indispensible that every transmission block contain a coding
identifier.

TABLE 2ù5/T.150

Communication process summarized



Step 1

Both parties agree by speech to switch to the telewriting
ON condition.

Step 2

Following telewriting ON, each terminal is in the receive
ready condition, i.e. the receiver is ON but it does not
receive telewriting signals.

Step 3

The first telewriting activity occurring at one of the ter-
minals causes that terminal to initiate the transmission of
the first transmission block.

Step 4

The terminal initiating the transmission of the first trans-
mission block assumes the master status.

Step 5

The session is established as soon as the receiving termi-
nal has received and recognized the coding identifier
contained in the first transmission block.

Step 6

Within the session, each terminal may alternatingly
assume send, receive and receive ready conditions, as
required by human actions and/or received signals.
When appropriate, the master status will be taken over
by an other terminal, as defined in the section on trans-
mission procedures.

Step 7

In case of a transmission collision, the terminal with
master status is permitted to continue sending; a terminal
with slave status has to await a new opportunity.

Step 8

The session is terminated when the terminals switch to
the telewriting OFF condition.

Part 3 ù Zone coding

1    General

1.1    This part of the Recommendation defines details of the zone coding
method.

1.2    For an application of zone coding together with telephony the com-
bined requirements from Parts 1, 2 and 3 apply.

1.3    This part also specifies how the coded signal is to be structured in 8
bit bytes, in order to fit in the transmission words defined in Part 2.

1.4    In the writing pad, the beginning of a stroke of handwriting is recog-
nized by the detection of the penùdown condition.

1.5    Each stroke generates a set of time serial coordinate pairs during
penùdown.

1.6    The coordinates of handwriting during penùdown are sampled at a
fixed rate of 40 samples/second.

1.7    The first sampling is initiated by penùdown, and continues, ending
when the pen is lifted.

1.8    The sequence of coordinate pairs is converted into a coded representa-
tion according to the zone coding rules. After this conversion the stroke is
represented by the presentation element TRACE.

1.9    The presentation elements are coded in the form of opcodes and oper-
ands.

1.10    The opcodes have a fixed 8ùbit length; the operands have a variable
length.

1.11    The telewriting coordinate information is contained in the operands.

2    Presentation elements

2.1    In tone coding, the following presentation elements are distinguished:

ù    trace

ù    marker

ù    partial erasure

ù    untrace

ù    set colour

ù    line thickness

ù    complete erasure.

    These elements and the format of the associated command streams are
defined in Table 3ù1/T.150.

2.2    The opcodes are defined in Table 3ù2/T.150 (notation x/y means col-
umn x, row y, in a 16 ╫ 16 code table).

3    Zone coding description

3.1    A trace is coded as a sequence of vectors (vector = D).

3.2    The beginning of a trace is the starting point of the first vector.

3.3    The end point of a vector constitutes the starting point for the next
vector in the trace.

3.4    The starting point position of the first vector of each trace is coded in
the form of a pair of absolute coordinates.

3.5    The position of each endpoint is determined by means of a measure-
ment system, the origin of which must coincide with the starting point of the
vector.

3.6    Within this measurement system, the endpoint position is found
through a three step approximation:

ù    step 1: the quadrant q, one value out of four; see Figure 3ù1/
T.150;

ù    step 2: the zone k within the quadrant; for division and numbering,
see Figure 3ù2/T.150;

ù    step 3: the relative address A within the zone.

3.7    In the coded representation, the quadrant and zone are indicated in a
differential way: dq and dk.

3.8    A set of 30 combinations of dq and dk are selected to be coded in a
compressed form, see Table 3ù3/T.150.

3.9    The relative address within the zone has a length depending upon the
size of the zone.

3.10    A vector end point position of which the combination dq and dk is not
defined in Table 3ù3/T.150 is coded by EFZ (escape from zone code) fol-
lowed by the absolute address.

3.11    The end of a trace is indicated by PLI (pen lift indicator) following
the last (relative or absolute) address.

3.12    The zone coding is defined more precisely in ºº 4 and 5. An example
of this coding is given in º 6.

4    Definitions of terms used in coding

4.1    The vector Di defined by:

Di = Pi ù Più1

= (dxi, dyi) = (xi ù xiù1, yi ù yiù1)

where Pi is the iùth coordinate pair during penùdown.

TABLE 3ù1/T.150

Presentation element commands



Trace TRn

    ù    The TRn command draws line segments that are
defined by a coùordinate information operand.

ù    The TRn command stream is:
ISP, TRn, . . . coùordinate information . . . ISP.

Marker MKn

 ù    The MKn command draws a marker pattern, the cen-
ter of which is specified by a coùordinate information
operand.

ù    The MKn command stream is:
ISP, MKn, . . . coùordinate information . . . ISP.

Partial Era-
sure PEn

 ù    The PEn command erases the closed area defined by a
coùordinate information operand.

ù    The PEn command stream is:
ISP, PEn, . . . coùordinate information . . . ISP.

Untrace UTn

 ù    The UTn command erases the square area (with its
sides parallel to the sides of the unit area) the centre of
which is specified by a coùordinate operand.

ù    The size of the square is defined as follows:
(32 ╫ 2nù9 ù 1) ╫ (32 ╫ 2nù9 ù 1) grid units.

ù    The UTn command stream is:
ISP, UTn . . . coùordinate information . . . ISP.

Set Colour
SC*

 ù    The SC* command sets a colour attribute to a particu-
lar trace. The colour attribute* can be set at the values:

    * = R: red    * = B: blue

    * = G: green    * = M: magenta

    * = Y: yellow    * = C: cyan

    * = W: white

ù    The effect of an SC* command remains valid until
the next SC* or CE command.

ù    The SC* command stream is:
    ISP, SC*, ISP, TRn, . . . coùordinate information .
. . ISP.

Line Thick-
ness LT*

ù    The LT* command sets a line thickness that is
defined by *, as follows:

    * = 1: one grid unit width,

    * = 2: two grid units width,

    * = 3: three grid units width.

ù    The effect of a LT* command remains valid until
the next LT* or CE command.

ù    The LT* command stream is:
    ISP, LT*, ISP, TRn, . . . coùordinate information .
. . ISP.

Complete Era-
sure CE

ù    The displayed image is erased completely.

ù    The CE command is:
    ISP, CE, ISP.

n    determines the grid resolution,

n = 9    means: grid resolution = 512 ╫ 512 (default value),

n = 10    means: grid resolution = 1024 ╫ 1024,

n = 11    means: grid resolution = 2048 ╫ 2048,

ISP    Information Separator.

TABLE 3ù2/T.150

Zone coding presentation opcodes



Element

 Com-
mand

 Code

Trace

 TR 9
TR 10
 TR 11

 12/9
12/10
12/11

Marker

 MK 9
MK 10
MK 11

13/9
13/10
13/11

Partial erasure

PE 9
PE 10
PE 11

14/9
14/10
14/11

Untrace

UT 9
UT 10
UT 11

15/9
15/10
15/11

Set colour

SC R
SC G
SC Y
SC B
SC M
SC C
SC W

11/0
11/1
11/2
11/3
11/4
11/5
11/6

Line thickness

 LT 1
LT 2
LT 3

 10/0
10/1
10/2

Complete erasure

 CE

 0/12

4.2    The quadrant number of the iùth vector, qi, is defined as (see Figure 3ù
1/T.150):

    qi    = 1 for dx  0, dy  0

    = 2 for dx < 0, dy  0

    = 3 for dx < 0, dy < 0

    = 4 for dx  0, dy < 0

Fig. 3ù1/T.150/T0803840-89 = 12 cm



4.3    Zone division and zone designation number

    The space of vectors without signs is divided into square zones. The zones
are numbered counterù clockwise, as shown in Figure 3ù2/T.150.

    The zone width is taken as the power of two. Thus the width of the kùth
zone is defined as:

W(k)    = 2            for k = 1

        = 2 ╫ 2(kù2)/3    for k > 1

4.4    The kùth zone Zk is defined as:

1)    for k = 1

Zk = (| dx |, | dy |); 0  | dx |  W(k)ù1, 0  | dy |  W(k)ù1

2)    for k > 1

a)    for k = 0 (mod 3)

Zk = (| dx |, | dy |); W(k)  | dx |  2W(k)ù1, W(k)  | dy |  2W(k)ù
1

b)    for k = 1 (mod 3)

Zk = (| dx |, | dy |); 0  | dx |  W(k)ù1, W(k)  | dy |  2W(k)ù1

c)    for k = 2 (mod 3)

Zk = (| dx |, | dy |); W(k)  |dx |  2W(k)ù1, 0  | dy |  W(k)ù1

Fig. 3ù2/T.150/T.0803850-89 = 10 cm



4.5    The origin of the relative addresses in each zone is the lower left corner.
The relative address in the kùth zone, (Ax, Ay), is defined as:

1)    for k = 1

Ax = dx, Ay = dy

2)    for k > 1

a)    for k = 0 (mod 3)

Ax = | dx | ù W(k), Ay = | dy | ù W(k)

b)    for k = 1 (mod 3)

Ax = | dx |, Ay = | dy | ù W(k)

c)    for k = 2 (mod 3)

Ax = | dx | ù W(k), Ay = | dy |

4.6    Quadrant number difference dqi is defined as:

dqi = qi ù qiù1

where q0 = 1 for simplicity.

4.7    Zone number diffdrence dki is defined as:

dki = ki ù kiù1

where ki is the zone number obtained by the iùth vector, and k0 = 1 for
simplicity.

5    Specification of the coding

5.1    The first penùdown point is represented by the binary expression of
the absolute coordinate pair (x0, y0), as follows:





x0

 y0







MSB

LSB      MSB

L
S
B

    2 ╫ 9 bits

                    MSB    Most significant bit

                    LSP    Least significant bit

5.2    All successive penùdown points are represented by zone codes (ZC)
and relative addresses (Ax, Ay).

5.3    The zero vector (0, 0) is not coded and transmitted. It is also possible the
zone vector (| XiùXiù1 |  1, | YiùYiù1 |  1) will be rejected before being
coded.

5.4    The zone code is defined in Table 3ù3/T.150. The table specifies a zone
code number 1 to 30 and a bit combination for 30 combinations of dq and
dk.

5.5    The relative addresses (Ax, Ay) are represented by:

5.6    The bit length L is decided by:

    L = 2 log2 W(k).

5.7    For the combination of dq and dk, not defined in Table 3ù3/T.150, the
absolute addresses (xi, yi) follow EFZ, instead of ZC.

5.8    A stroke is terminated by the pen lift indicator (PLI) as soon as the pen is
lifted.

5.9    The full data format of a stroke is illustrated in Figure 3ù3/T.150.

6    A coding example

    The trace of handwritten information is shown in Figure 3ù4/T.150,
where Pi is the sampled point. An example of how to encode the coordinate
data is shown in Table 3ù4/T.150. The zone coded bit stream is shown in
Figure 3ù5/T.150.

7    Data structure

7.1    The zone coding opcodes and operands and the opcodes representing
control commands are transmitted in the form of data packets.

7.2    Each packet consists of a header octet ISP (information separator),
followed by an integral number of octets, and terminated by an ISP octet.

7.3    A packet may contain an undetermined number of opcodes. Bound-
aries of opcodes coincide with the boundaries of octets.

7.4    Data of variable length (the operand) is preceded by an opcode. After
each operand the packet is terminated by an ISP octet at the earliest regular
octet boundary.

7.5    If the end of the operand does not coincide with an octet boundary, the
remaining bit positions until the octet boundary shall be filled with bits of
the value ZERO.

    At the receiving end, these zeros are ignored.

TABLE 3ù3/T.150

Zone code table



Zone code No.

 dq

 dk

 Length of
the code
(bit)

 Code
(the left bit is
LSB)

  1

 0

  0

 2

     01

  2

 3

  0

 4

     00 01

  3

 1

  0

 4

     11 11

  4

 0

  3

 4

     00 10

  5

 0

  1

 4

     10 11

  6

 0

 ù
3

 4

     11 10

  7

 3

  3

 5

     10 01 1

  8

 0

 ù
1

 5

     00 11 1

  9

 3

 ù
1

 6

     10 01 01

 10

 3

 ù
3

 6

     10 00 01

 11

 2

  0

 6

     00 11 01

 12

 1

  3

 6

     10 10 01

 13

 1

  1

 6

     10 00 11

 14

 1

 ù
3

 6

     10 10 11

 15

 0

  4

 6

     10 00 10

 16

 0

  2

 6

     00 00 11

 17

 0

 ù
2

 6

     00 00 01

 18

 3

  2

 7

     10 00 00 1

 19

 3

  1

 7

     10 01 00 1

 20

 2

  3

 7

     10 10 10 0

 21

 1

  2

 7

     10 10 00 1

 22

 1

 ù
1

 7

     00 11 00 1

 23

 1

 ù
2

 7

     10 01 00 0

 24

 0

  6

 7

     00 00 00 1

 25

 0

 ù
4

 7

     00 11 00 0

 26

 0

 ù
6

 7

     10 10 00 0

 27

 3

  6

 8

     10 10 10 10

 28

 2

  1

 8

     10 00 00
01

 29

 2

 ù
1

 8

     10 10 10
11

 30

 2

 ù
3

 8

     00 00 00
01

 PLI

3

 EFZ

6

     11 0

 NULL

8

    00 00 10

     00 00 00 00

    PLI    Pen lift indicator

    EFZ    Escape from zone code





Ax

 Ay



MSB

LSB   MSB

L
S
B







EFZ

 xi

 yi



MSB

MS
B







x0

 y0

 ZC(1)

 A(1)

 ZC(2)

 A(2)

PLI

 x0, y0    Starting address

ZC(i)    Zone code of the iùth vector

A(i)    Relative address of the ith vector

PLI    Pen lift indicator

FIGURE 3ù3/T.150

Stroke data format

Fig. 3ù3/T.150/T0803860-89= 9 cm



TABLE 3ù4/T.150

Coding example



i


x,
y

 dx,
dy

 q

 k


dq


dk


Z
C

 Ax,
Ay


W(
k)

 L/
2

 ZCù
code

 0


1,
3


(1
)

(1
)

 1


1,
4

 0, 1

 1

 1

 0

 0

 1

 0, 1

 2

 1

 01

 2


2,
5

 1, 1

 1

 1

 0

 0

 1

 1, 1

 2

 1

 01

 3


5,
6

 3, 1

 1

 2

 0

 1

 5

 1, 1

 2

 1

 1011

 4


7,
6

 2, 0

 1

 2

 0

 0

 1

 0, 0

 2

 1

 01

 5


8,
5

 1, ù
1

 4

 1

 3


ù
1

 9

 1, 1

 2

 1


10010
1

 6


9,
5

 1, 0

 1

 1

 1

 0

 3

 1, 0

 2

 1

 1111

Fig. 3ù5/T.150/T0803870-89 = 12 cm



7.6    Successive packets may be sent contiguously, separated by a single ISP
octet. See Figure 3ù6/T.150.

Fig. 3ù6/T.150/T0803880-89 = 5 cm



7.7    If one of the octets containing variable length data accidentally imitates
an ISP octet, the transmitter inserts an extra ISP octet, so that the imitation is
duplicated. See Figure 3ù7/T.150.

    If the imitation results from a combination of bits in two adjacent octets, no
action is taken.

Fig. 3ù7/T.150/T0803890 = 7 cm



7.8    The receiver ignores the second ISP octet from each pair of ISP octets.

8    Temporary     penùstop

8.1    During the writing process, the pen may stop at an arbitrary instant,
remaining on the writing surface. As a consequence, the completion of the
current operand is suspended.

8.2    Generally, the instant of penùstop does not coincide with a byte
boundary. In order to provide the receiving party with upùtoùdate infor-
mation including the correct penùstop position, the content of the incom-
plete byte should be transmitted prior to the MARK signal representing the
writing pause.

8.3    The above can be achieved by means of the insertion of 8 extra bits,
the NULL bits, in the bitstream. Each NULL bit has the binary value Zero.

8.4    The NULL bits are subdivided into two groups, one group preceding
the MARK signal, the other group following the MARK signal.

8.5    The number of NULL bits in the first group equals the number of
open bit positions in the current byte. This number is referred to as N.

8.6    By the inclusion of N NULL bits the current byte is complete and can
be transmitted. It is followed by the MARK signal.

8.7    As soon as the next writing activity occurs, the MARK signal is ter-
minated.

8.8    The remaining 8ùN NULL bits are to occupy the leading bitùposi-
tions of the first byte after the MARK signal.

8.9    The NULL bit mechanism is illustrated in Figure 3ù8/T.150.

Fig. 3ù8/T.150/T0803900 = 5 cm



9    Control commands

9.1    This section defines control commands, affecting the functioning of
the terminal at the presentation level.

    These commands are:

ù    complete erasure,

ù    escape,

ù    information separator.

9.2    Complete erasure CE

    This command is defined already in Table 3ù1/T.150. It is repeated
here because of the buffer control aspect.

    The displayed image is erased completely, both at the sending side
and the receiving side. Also the telewriting data in the transmission buffer at
the sending side, and in the reception buffer at the receiving side is erased.

9.3    Excape ESC

    This is a code extension command. ESC is to be followed by an 8ù
bit operand, defining an alternative code table. ESC + operand is to be sent
by an enhanced Telewriting terminal prior to each enhanced operation func-
tion. Details are to be defined in a section on enhanced terminal.

9.4    Information separator ISP

    ISP acts as a delimiter of command packets as defined in º 7. The ter-
minal should check received data streams for pairs of ISP octets and, where
required, should reject every second ISP octet.

9.5    The coding of the above commands is defined in Table 3ù5/T.150
(the notation x/y means column x, row y, in a 16 ╫ 16 code table).

TABLE 3ù5/T.150

Coding of control commands



Function

 Acronym

 Coding

Complete erasure

 CEC

 0/12

Escape

 ESC

 1/11

Information separator

 ISP

 1/15

    10    Summary code table

    A summary of the coding for the opcodes is given in Figure 3ù9/
T.150. All elements included have been defined in the previous sections.

Fig. 3ù9/T.150/T0803910-89 = 12 cm



11    Summary     transmission data format

    The transmission data format is illustrated in Figure 3ù10/T.150.

Fig. 3ù10/T.150/T0803920-89 = 14 cm



12        Zone coding basic terminal

12.1    The basic terminal must be able to receive and correctly process the
following presentation element commands:

    TR 9, MK 9, PE 9, CE, ISP.

12.2    The following presentation elements are optional:

TR 10, TR11

MK 10, MK 11

PE 10, PE 11

UT 9, UT 10, UT 11.

    I.e. the transmitter may or may not be equipped with these commands.

    The receiver must be able to receive and correctly process these com-
mands.

12.3    The following control commands are optional:

    ESC, LT*, SC*

    I.e. the receiver will accept these commands but does not undertake
any further action.

Part 4 ù Differential chain coding

1    General

1.1    This part of the Recommendation defines details of the differential
chain coding method.

1.2    For an application of differential chain coding together with tele-
phony, the combined requirements from Parts 1, 2 and 4 apply.

1.3    Differential chain coding is derived from the Videotex geometric cod-
ing as defined in Recommendation T.101, Annex C (CEPT Videotex).

1.4    The telewriting functionalities are nearly a subset of the Videotex
geometric functionalities, as defined in Recommendation T.101, Annex C.

1.5    Differential chain coding was developed for compression purposes. In
this coding method, the statistical properties of handwriting are employed.

1.6    This coding method uses spatial sampling of curves, as distinct from
sampling with a fixed frequency. The size of the sampling steps is deter-
mined by the size of the soùcalled coding ring.

1.7    The precision of this coding method is expressed in grid units, GU. In
the default situation, one GU corresponds to the binary fraction 2 ** ù9 of
the unit length.

1.8    Each stroke of handwriting is processed by the writing pad circuitry
and converted into a coded form.

    The coded representation of a stroke is called TRACE.

1.9    The coding of the presentation element trace, as well as the coding of
the remaining presentation elements is defined in terms of 7 bit coding.

1.10    Conversion into 8 bit structured coding as required for transmission,
is also specified in this Recommendation.

1.11    The word ôbyteö where used in this Recommendation, refers to a
combination of 7 or 8 bits, whatever is appropriate in the given context.

2    Presentation elements

    In differential chain coding, the following presentation elements are
distinguished:

ù    trace

ù    marker

ù    closed area

ù    partial erasure

ù    background

ù    complete erasure.

    The attributes are:

ù    colour

ù    trace thickness

ù    trace texture.

    These presentation elements together with the attributes are described
in Table 4ù1/T.150.

TABLE 4ù1/T.150

Differential chain coding presentation elements



Element

Description

Trace

The trace is coded as a trace opcode plus a set of coù
ordinate information defining a sequence of line seg-
ments. Trace corresponds with polyline in videotex.

Marker

The marker is coded as a marker opcode plus a single
coùordinate pair defining the position ot the marker's
center point.

Closed area

The closed area is coded by an opcode plus a set of coù
ordinate information defining a closed perimeter. The
closed area corresponds with fill area in videotex.

Partial erasure

Partial erasure is obtained by means of the closed area
concept. By giving the closed area the same attributes as
the background, erasure is achieved for the area enclosed
in the perimeter.

Background

At initialization and after complete erasure, the back-
ground shows default appearance. Modification of the
background is obtained by means of the closed area con-
cept. The closed area is chosen to have the size of the
image area. The area attributes are set to the desired
background appearance.

Complete era-
sure

Complete erasure is obtained by means of the clear
screen concept. The whole image area will be set to the
default background appearance.

Colour

Colour is an attribute, applicable to trace and closed are
(including background). The effect of a colour command
remains valid until the next colour command.

Trace thick-
ness

Trace thickness is an attribute. It is determinated by
means of a scale factor. The effect of a trace thickness
command remains valid for all subsequent traces, until
the next trace thickness command.

Trace texture

Trace texture is an attribute. It is determinated by means
of a parameter allowing a choice amoung defined tex-
tures. The effect of a trace texture command remains
valid for all subsequent traces, until the next trace texture
command.

Marker type

Marker type is an attribute. It is determinated by means
of a parameter allowing a choice among defined textures.
The default value of marker type is 1. If the specified
value is outside the range 0 . . . 4, the marker is not dis-
played.

    3    Description of the coding

3.1    The coded representation of a presentation element is called PRIMI-
TIVE.

3.2    A primitive is composed of one opcode and a number of operands as
required.

3.3    Certain opcodes are coded as a single byte, other opcodes are coded
as combinations of two bytes.

3.4    The operand part of a primitive may utilize either basic format encod-
ing or pointlist encoding.

3.5    In basic format encoding the operand part of the primitive contains
one or more operands, each consisting of one or more bytes.

3.6    In the pointlist encoding the operand part of the primitive contains
coordinate information regarding an individual point or regarding a
sequence of related points.

3.7    The position of an individual point, as well as the position of each
first point of a sequence, is coded in absolute coordinates, i.e. the xù and
yùcoordinate with respect to the origin of the coding space.

3.8    For the coding of the remaining points of a sequence, a choice is to be
made among two possibilities, namely displacement mode and incremental
mode.

3.9    In the displacement mode, each point (after the first) is coded by
means of two size value parameters. The first size value gives the xùcom-
ponent of the point's displacement from the preceding point in the sequence,
the second size value gives the yùcomponent of the displacement.

3.10    In the incremental mode, a mechanism is used in which a single
value, derived from a table, determines the position of a point with respect
to the preceding point. This mechanism is suitable for coding a sequence of
points containing a high amount of position information, such as a trace.

3.11    The mechanism, introduced in º 3.10, is based on the use of a coding
ring. At the beginning of trace, the starting point determines the centre point
of the first ring. The intersection of trace and ring is identified and deter-
mines the centre point of the second ring.

3.12    Each new intersection determines the centre point for the next ring.
Thus, the trace is represented by the starting point plus the series of intersec-
tion points. The end of a trace is indicated by means of the end of blockù
code.

3.13    The method for identifying the various points on a ring utilizes small
numbers for points with a high probability of being intersected and larger
numbers for points with lower probability.

3.14    The numbering system for the reference points on the ring is defined
in ºº 4.6 and 4.7.

4    Incremental mode mechanism

4.1    The coding data in the incremental mode does not reflect coordinate
size values, but represents a sequence of points identified by means of suc-
cessive coding rings. Each ring identifies one point.

4.2    A ring is a set of reference points, positioned on the perimeter of a
square. The position of the square is identified through the position of its
centre point. The sides of the square are parellel to the xù and yùaxes.

4.3    The characteristics of the ring are determined by its radius R, its angu-
lar resolution factor p and its direction D.

    The size of R is expressed in GU.

4.4    The number of reference points on a ring is N. The value of N is
determined by:

    N = , with p = 0, 1, 2, 3.

    It follows that the maximum number of reference points is N = 8R.

4.5    N must be even. If N is odd, the encoded operand (the point list) must
be discarded. If N is even for the first part of the operand, but N is odd for
the remaining part, the remaining part (with N being odd) is discarded.

4.6    To the reference points on the ring, point numbers are assigned as fol-
lows. The numbering starts with 0. The point with number 0 is called the
direction point.

4.7    The default position for the direction point is shown in Figure 4ù1/
T.150. Adjacent points are numbered 1 . . . N/2ù1 in anticlockwise direc-
tion, and ù1 . . . ùN/2 in clockwise direction. Figure 4ù1/T.150 shows
two rings with the numbered reference points.

4.8    In the figure the left ring is characterized by R = 3 and p = 0; the right
ring by R = 3 and p = 1.

Fig. 4ù1/T.150/T0804070-89 = 9 cm



4.9    The position of the reference points on each ring is fixed. However, the
allocation of the point numbers is adapted to the trace direction as follows.

4.10    For the first ring of a sequence (at the starting point), the direction point
is at default position, as shown in Figure 4ù1/T.150.

4.11    As soon as the growing trace intersects the first ring, the nearest refer-
ence point is determined. This point constitutes the centre point for the next
ring.

4.12    The direction point on the second ring is located at that position where
the next intersection would take place if the trace continued as a straight
line.

4.13    As the trace grows, the nearest reference point at each intersection is
determined. The respective point numbers of these points are converted into
variable length code words according to the Huffman code table, defined in
Table 4ù2/T.150.

4.14    The radius can have a value of R0, 2R0, 4R0 or 8R0, where R0 is the
basic radius.

    The angular resolution factor p can have a value of 0, 1, 2 or 3.

    To modify these parameters the code table contains the codes C1 . . . C6. For
their use, see further on.

    The basic radius R0 can be specified by the primitive ôset domain ringö.
The default basic radius follows from:

    default basic radius = 2 ** max (0, ù8ùgranularity code).

4.15    The length of the code table is fixed. The point numbering ranges from
ù20 to +19. For the encoding in cases of rings with a higher number of ref-
erence points, two escape codes are defined: IMùESC 1 and IMùESC 2.
For their use, see º 5.

4.16    At the end of the trace no further intersections occur. The variable
length coded string is terminated by end of block.

TABLE 4ù2/T.150

Huffman code table for differential chain coding



Code No.

Length

Codeùword

Point number

  1

  2

         00

   0

  2

  2

         10

   1

  3

  2

         01

ù1

  4

  4

       1100

   2

  5

  4

       1101

ù2

  6

  6

     111000

   3

  7

  6

     111001

ù3

  8

  6

     111010

   4

  9

  6

     111011

ù4

 10

  8

   11110000

   5

 11

  8

   11110001

ù5

 12

  8

   11110010

   6

 13

  8

   11110011

ù6

 14

  8

   11110100

   7

 15

  8

   11110101

ù7

 16

  8

   11110110

   8

 17

  8

   11110111

ù8

 18

 10

 1111100000

   9

 19

 10

 1111100001

ù9

 20

 10

 1111100010

  10

 21

 10

 1111100011

ù10

 22

 10

 1111100100

  11

 23

 10

 1111100101

ù11

 24

 10

 1111100110

  12

 25

 10

 1111100111

ù12

 26

 10

 1111101000

  13

 27

 10

 1111101001

ù13

 28

 10

 1111101010

  14

 29

 10

 1111101011

ù14

 30

 10

 1111101100

  15

 31

 10

 1111101101

ù15

 32

 10

 1111101110

  16

 33

 10

 1111101111

ù16

 34

 10

 1111110000

  17

 35

 10

 1111110001

ù17

 36

 10

 1111110010

  18

 37

 10

 1111110011

ù18

 38

 10

 1111110100

  19

 39

 10

 1111110101

ù19

 40

 10

 1111110110

  C1

 41

 10

 1111110111

ù20

 42

 10

 1111111000

  C2

 43

 10

 1111111001

  C3

 44

 10

 1111111010

  C4

 45

 10

 1111111011

  C5

 46

 10

 1111111100

  C6

 47

 10

 1111111101

 IMùESC 1

 48

 10

 1111111110

 IMùESC 2

 49

 10

 1111111111

 End of block

    5    Change of coding parameters

5.1    The escape codes IMùESC 1 and IMùESC 2 enable the extension
of the point numbering range on the ring. I.e. also points outside the range
ù20 to +19 can be addressed. By the code IMùESC 1, the absolute value
of the point number is increased by 20, the sign remains unchanged.

    By the code IMùESC 2, the absolute value of the point number is
increased by 40, the sign remains unchanged.

5.2    The two escape codes can be used in combination with each other in
any desired order. Some examples in Table 4ù3/T.150 illustrate their use.
The number between [ ] represents the point number.

TABLE 4ù3/T.150

Use of escape codes, examples



Description

 Intended point numbers

<IMùESC 1>                 [1]

  21

<IMùESC 1>                 [ù1]

ù21

<IMùESC 2>                 [14]

  54

<IMùESC 2>                 [ù12]

ù52

<IMùESC 1> <IMùESC 2>     [6]

   66

<IMùESC 2> <IMùESC 1>     [ù
18]

ù78

5.3    The codes C1 up to C6 are used to change the parameters R and p that
define the ring to be used. The use of these codes is defined in ºº 5.4 and
5.10.

    By the use of these codes the direction point is set at default position.

5.4    The range in which the parameters should remain is as follows:

R:    R0, 2R0, 4R0, 8R0 (with R0 being the basic radius);

p:    0, 1, 2, 3.

5.5    Code C1 means: change R and p to the next higher value. E.g. if
radius is R, the next higher is 2R; if p = 0 the next higher is 1.

    R cannot become greater tan 8R0 and p cannot become greater than 3.
E.g. if current radius is 8R0 or current p = 3, the code C1 has no effect.

5.6    Code C2 means: change R and p to the next lower value. The effect of
C2 is the inverse of code C1.

    R cannot become smaller than R0 and p cannot become smaller than
0. E.g. if the current radius is R0 or the current p = 0, the code C2 has no
effect.

5.7    Code C3 means: change R to the next higher value. The code C3 has
no effect if the current radius = 8R0.

5.8    Code C4 means: change p to the next higher value. The code C4 has
no effect if the current p = 3.

5.9    Code C5 means: change R to the next lower value. The code C5 has
no effect if the current radius = R0.

5.10    Code C6 means: change p to the next lower value. The code C6 has
no effect if the current p = 0.

6    Coding formats

6.1    The coding is specified in terms of 7ùbit coding. For use in the 8 bit
environment as specified for transmission, bit No. b8 of each octet shall be
set to ZERO.

6.2    For reference, an empty 7ùbit code table is shown in Figure 4ù2/
T.150.