.rs
.\" Troff code generated by TPS Convert from ITU Original Files
.\"                 Not Copyright ( c) 1991 
.\"
.\" Assumes tbl, eqn, MS macros, and lots of luck.
.TA 1c 2c 3c 4c 5c 6c 7c 8c
.ds CH
.ds CF
.EQ
delim @@
.EN
.nr LL 40.5P
.nr ll 40.5P
.nr HM 3P
.nr FM 6P
.nr PO 4P
.nr PD 9p
.po 4P

.rs
\v | 5i'
.sp 1P
.ce 1000
\v'12P'
\s12PART\ I
\v'4P'
.RT
.ce 0
.sp 1P
.ce 1000
\fBSeries H Recommendations\fR \v'2P'
.EF '%	 \ \ \ ^''
.OF ''' \ \ \ ^	%'
.ce 0
.sp 1P
.ce 1000
\fBLINE\ TRANSMISSION\ OF\fR 
.ce 0
.sp 1P
.ce 1000
\fBNON\(hyTELEPHONE\ SIGNALS\fR 
.ce 0
.sp 1P
.LP
.rs
.sp 28P
.ad r
Blanc
.EF '%	 \ \ \ ^''
.OF ''' \ \ \ ^	%'
.ad b
.RT
.LP
.bp
.LP
\fBMONTAGE:\ \fR PAGE 2 = PAGE BLANCHE
.sp 1P
.RT
.LP
.bp
.ce 1000
\v'6P'
\fBLINES\ USED\ FOR\ THE\ TRANSMISSION\ OF\ SIGNALS\fR 
.ce 0
.ce 1000
\fBOTHER\ THAN\ TELEPHONE\ SIGNALS,\ SUCH\ AS\ TELEGRAPH,\fR 
.ce 0
.sp 1P
.ce 1000
\fBFACSIMILE,\ DATA,\ ETC.,\ SIGNALS\fR 
.FS
Excluding the
transmission of sound\(hyprogramme and television signals, which is the
subject of the Series\ J Recommendations.
.FE
.ce 0
.sp 1P
.PP
Part\ I contains two classes of Recommendations: those which
define the characteristics of \fItransmission channels\fR (telephone\(hytype, 
group, supergroup,\ etc., circuits) used only to transmit signals other 
than telephone signals, and those which define the characteristics of the 
\fIsignals\fR used in 
such transmissions.
.EF '%	 \ \ \ ^''
.OF ''' \ \ \ ^	%'
.sp 1P
.RT
.PP
In this Part, \*Qwideband\*U is used to qualify the transmission
channels, and \*Qwide\(hyspectrum\*U the signals transmitted, so as to 
avoid any 
confusion between the transmission channels and the signals transmitted with
regard to the frequency bands involved in transmission over group links,
supergroup links,\ etc.
.PP
As far as possible, one should avoid specifying the characteristics of 
particular channels or signals in defining a new service and refer only 
to the characteristics of the channels mentioned in Section\ 1 of this 
Recommendation Series. 
.PP
Section 6 of this Series is reserved for Recommendations concerning
the characteristics of visual telephone systems.
.PP
Table 1 indicates the correspondence of Series H Recommendations to
Recommendations of other Series.
.RT
.sp 1P
.ce 1000
TABLE\ 1
.ce 0
.sp 1P
.LP
	\fISeries H Recommendations\fR 	\fIRecommendations of other Series\fR \v'4p'
.sp 9p
.RT
.LP
	H.12,\ \(sc\ 1
	M.1040 (Volume IV)
.LP
	H.12,\ \(sc\ 2
	M.1025 (Volume IV)
.LP
	H.12,\ \(sc\ 3
	M.1020 (Volume IV)
.LP
	H.13
	See Recommendation O.71 (Volume IV)
.LP
	H.14,\ \(sc\ 2
	M.910 (Volume IV)
.LP
	H.16
	O.72 (Volume IV)
.LP
	H.21
	See also the Recommendations M.800
(Volume IV) and R.77 (Volume VII)
.LP
	H.22
	See also the Recommendation M.810 (Volume IV)
.LP
	H.23
	Extract of Recommendations R.31
and R.35 (Volume VII)
.LP
	H.32
	R.43 (Volume VII)
.LP
	H.41
	T.11 (Volume VII)
.LP
	H.42
	T.12 (Volume VII)
.LP
	H.43
	T.10 (Volume VII)
.LP
	H.51
	V.2 (Volume VIII)
.LP
.sp 1
.bp
.LP
\fBMONTAGE: \ \fR PAGE 4 = BLANCHE
.sp 1P
.RT
.LP
.bp
.sp 1P
.ce 1000
\v'3P'
SECTION\ 1
.ce 0
.sp 1P
.ce 1000
\fBLINES\ USED\ FOR\ THE\ TRANSMISSION\fR 
.ce 0
.ce 1000
\fBOF\ SIGNALS\ OTHER\ THAN\ TELEPHONE\ SIGNALS,\fR 
.ce 0
.ce 1000
\fBSUCH\ AS\ TELEGRAPH,\ FACSIMILE,\fR 
.ce 0
.sp 1P
.ce 1000
\fBDATA,\ ETC.,\ SIGNALS\fR \v'1P'
.ce 0
.sp 1P
.IP
\fB1.1\ \fR \fBCharacteristics of transmission\fR \fBchannels used for 
other than telephone purposes\fR 
.sp 1P
.RT
.sp 2P
.LP
\fBRecommendation\ H.11\fR 
.RT
.sp 2P
.sp 1P
.ce 1000
\fBCHARACTERISTICS\ OF\ CIRCUITS\ IN\ THE\ SWITCHED\ TELEPHONE\ NETWORK\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.11''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.11	%'
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
\v'2P'
.ce 0
.sp 1P
.sp 2P
.LP
\fBRecommendation\ H.12\fR 
.RT
.sp 2P
.sp 1P
.ce 1000
\fBCHARACTERISTICS\ OF\ TELEPHONE\(hyTYPE\ LEASED\ CIRCUITS\fR \v'1P'
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.12''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.12	%'
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
\v'2P'
.ce 0
.sp 1P
.sp 2P
.LP
\fBRecommendation\ H.13\fR 
.RT
.sp 2P
.ce 1000
\fBCHARACTERISTICS\ OF\ AN\ IMPULSIVE\ NOISE\ MEASURING\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.13''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.13	%'
.ce 0
.sp 1P
.ce 1000
\fBINSTRUMENT\ FOR\ TELEPHONE\(hyTYPE\ CIRCUITS\fR 
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Recommendation O.71 in |
Fascicle IV.4 of Volume IV
.sp 1P
.RT
.ce 0
.LP
of the \fIRed Book\fR , ITU, Geneva, 1985)
.bp
.sp 1P
.ce 1000
.ce 0
.sp 1P
.sp 2P
.LP
\fBRecommendation\ H.14\fR 
.RT
.sp 2P
.ce 1000
\fBCHARACTERISTICS\ OF\ GROUP\ LINKS\ FOR\ THE\ TRANSMISSION\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.14''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.14	%'
.ce 0
.sp 1P
.ce 1000
\fBOF\ WIDE\(hySPECTRUM\ SIGNALS\fR 
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
\v'2P'
.ce 0
.sp 1P
.sp 2P
.LP
\fBRecommendation\ H.15\fR 
.RT
.sp 2P
.ce 1000
\fBCHARACTERISTICS\ OF\ SUPERGROUP\ LINKS\ FOR\ THE\ TRANSMISSION\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.15''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.15	%'
.ce 0
.sp 1P
.ce 1000
\fBOF\ WIDE\(hySPECTRUM\ SIGNALS\fR 
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
\v'2P'
.ce 0
.sp 1P
.sp 2P
.LP
\fBRecommendation\ H.16\fR 
.RT
.sp 2P
.ce 1000
\fBCHARACTERISTICS\ OF\ AN\ IMPULSIVE\(hyNOISE\ MEASURING\ INSTRUMENT\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.16''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.16	%'
.ce 0
.sp 1P
.ce 1000
\fBFOR\ WIDEBAND\ DATA\ TRANSMISSION\fR 
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
\v'2P'
.ce 0
.sp 1P
.IP
\fB1.2\ \fR 	\fBUse of telephone\(hytype circuits for voice\(hyfrequency\fR 
\fBtelegraphy\fR 
.sp 1P
.RT
.sp 2P
.LP
\fBRecommendation\ H.21\fR 
.RT
.sp 2P
.ce 1000
\fBCOMPOSITION\ AND\fR \ 
\fBTERMINOLOGY\ OF\ INTERNATIONAL\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.21''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.21	%'
.ce 0
.sp 1P
.ce 1000
\fBVOICE\(hyFREQUENCY\ TELEGRAPH\ SYSTEMS\fR 
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
\v'2P'
.ce 0
.sp 1P
.sp 2P
.LP
\fBRecommendation\ H.22\fR 
.RT
.sp 2P
.ce 1000
\fBTRANSMISSION\ REQUIREMENTS\ OF\ INTERNATIONAL\ VOICE\(hyFREQUENCY\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.22''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.22	%'
.ce 0
.sp 1P
.ce 1000
\fBTELEGRAPH\ LINKS\ (AT\ 50,\ 100\ AND\ 200\ BAUDS)\fR 
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.LP
of the \fIRed Book\fR , ITU, Geneva, 1985)
.bp
.sp 1P
.ce 1000
.ce 0
.sp 1P
.sp 2P
.LP
\fBRecommendation\ H.23\fR 
.RT
.sp 2P
.ce 1000
\fBBASIC\ CHARACTERISTICS\ OF\ TELEGRAPH\ EQUIPMENTS\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.23''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.23	%'
.ce 0
.sp 1P
.ce 1000
\fBUSED\ IN\ INTERNATIONAL\ VOICE\(hyFREQUENCY\ TELEGRAPH | fR \fBSYSTEMS\fR 
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
\v'2P'
.ce 0
.sp 1P
.IP
\fB1.3\ \fR 	\fBTelephone circuits or cables used for various types of\fR 
\fBtelegraph transmission or for simultaneous transmissions\fR 
.sp 1P
.RT
.sp 2P
.LP
\fBRecommendation\ H.32\fR 
.RT
.sp 2P
.ce 1000
\fBSIMULTANEOUS\ COMMUNICATION\ BY\ TELEPHONY\ AND\ TELEGRAPHY\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.32''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.32	%'
.ce 0
.sp 1P
.ce 1000
\fBON\ A\ TELEPHONE\(hyTYPE\ CIRCUIT\fR 
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
\v'2P'
.ce 0
.sp 1P
.sp 2P
.LP
\fBRecommendation\ H.34\fR 
.RT
.sp 2P
.ce 1000
\fBSUBDIVISION\ OF\ THE\ FREQUENCY\ BAND\ OF\ A\ TELEPHONE\(hyTYPE\ CIRCUIT\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.34''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.34	%'
.ce 0
.sp 1P
.ce 1000
\fBBETWEEN\ TELEGRAPHY\ AND\ OTHER\ SERVICES\fR 
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
\v'2P'
.ce 0
.sp 1P
.IP
\fB1.4\ \fR 	\fBTelephone\(hytype circuits used for facsimile\fR 
\fBtelegraphy\fR 
.sp 1P
.RT
.sp 2P
.LP
\fBRecommendation\ H.41\fR 
.RT
.sp 2P
.sp 1P
.ce 1000
\fBPHOTOTELEGRAPH\ TRANSMISSIONS\ ON\ TELEPHONE\(hyTYPE\ CIRCUITS\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.41''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.41	%'
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.LP
of the \fIRed Book\fR , ITU, Geneva, 1985)
.bp
.sp 1P
.ce 1000
.ce 0
.sp 1P
.sp 2P
.LP
\fBRecommendation\ H.42\fR 
.RT
.sp 2P
.ce 1000
\fBRANGE\ OF\ PHOTOTELEGRAPH\ TRANSMISSIONS\fR 
.EF '%	Fascicle\ III.4\ \(em\ Rec.\ H.42''
.OF '''Fascicle\ III.4\ \(em\ Rec.\ H.42	%'
.ce 0
.sp 1P
.ce 1000
\fBON\ A\ TELEPHONE\(hyTYPE\ CIRCUIT\fR 
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
\v'2P'
.ce 0
.sp 1P
.sp 2P
.LP
\fBRecommendation\ H.43\fR 
.RT
.sp 2P
.ce 1000
\fBDOCUMENT\ FACSIMILE\ TRANSMISSIONS\ ON\ LEASED\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.43''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.43	%'
.ce 0
.sp 1P
.ce 1000
\fBTELEPHONE\(hyTYPE\ CIRCUITS\fR 
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
\v'2P'
.ce 0
.sp 1P
.IP
\fB1.5\ \fR 	\fBCharacteristics of data signals\fR 
.sp 1P
.RT
.sp 2P
.LP
\fBRecommendation\ H.51\fR 
.RT
.sp 2P
.sp 1P
.ce 1000
\fBPOWER\ LEVELS\ FOR\ DATA\ TRANSMISSION |
OVER\ TELEPHONE\ LINES\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.51''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.51	%'
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
\v'2P'
.ce 0
.sp 1P
.sp 2P
.LP
\fBRecommendation\ H.52\fR 
.RT
.sp 2P
.ce 1000
\fBTRANSMISSION\ OF\ WIDE\(hySPECTRUM\ SIGNALS\ (DATA,\ FACSIMILE,\ ETC.)\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.52''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.52	%'
.ce 0
.sp 1P
.ce 1000
\fBON\ WIDEBAND\ GROUP\ LINKS\fR 
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
\v'2P'
.ce 0
.sp 1P
.sp 2P
.LP
\fBRecommendation\ H.53\fR 
.RT
.sp 2P
.ce 1000
\fBTRANSMISSION\ OF\ WIDE\(hySPECTRUM\ SIGNALS\ (DATA,\ ETC.)\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.53''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.53	%'
.ce 0
.sp 1P
.ce 1000
\fBOVER\ WIDEBAND\ SUPERGROUP\ LINKS\fR 
.ce 0
.sp 1P
.ce 1000
(The text of this Recommendation can be found in Fascicle III.4 |
of Volume III
.sp 1P
.RT
.ce 0
.sp 1P
.ce 1000
of the \fIRed Book\fR , ITU, Geneva, 1985)
.ce 0
.sp 1P
.LP
.bp
.sp 1P
.ce 1000
\v'3P'
SECTION\ 2
.ce 0
.sp 1P
.ce 1000
\fBCHARACTERISTICS\ OF\ VISUAL\ TELEPHONE\ SYSTEMS\fR 
.ce 0
.sp 1P
.sp 2P
.LP
\fBRecommendation\ H.100\fR 
.RT
.sp 2P
.sp 1P
.ce 1000
\fBVISUAL\ TELEPHONE\ SYSTEMS\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.100''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.100	%'
.ce 0
.sp 1P
.ce 1000
\fI(former Recommendation H.61, Geneva, 1980;\fR 
.sp 9p
.RT
.ce 0
.ce 1000
\fIamended at Malaga\(hyTorremolinos, 1984\fR 
.ce 0
.sp 1P
.ce 1000
\fIand at Melbourne, 1988)\fR 
.ce 0
.sp 1P
.LP
\fB1\fR 	\fBDefinition\fR 
.sp 1P
.RT
.PP
The \fBvisual telephone service\fR is generally a two\(hyway
telecommunication service which uses a switched network of broadband analogue 
and/or digital circuits to establish connections among subscriber terminals, 
primarily for the purpose of transmitting live or static pictures.
.PP
Special application one\(hyway systems, e.g.\ surveillance and some
information retrieval systems, or a non\(hyswitched videoconference service, 
can be regarded as degenerate cases of the visual telephone service. 
.PP
The visual telephone service also includes the associated
speech.
.RT
.sp 2P
.LP
\fB2\fR 	\fBFacilities to be offered\fR 
.sp 1P
.RT
.PP
The design of the visual telephone service shall be such as to
offer at least the following basic facilities:
.RT
.LP
	a)
	Transmission of live pictures such as head and shoulders
of one person or a small group of persons, with moderate
definition.
.LP
	b)
	Transmission of the associated speech.
.LP
	c)
	Transmission of graphics information such as drawings and
documents with high definition (e.g.\ 625 lines or 525\ lines).
.LP
	d)
	Video conference service, with or without the use
of split\(hyscreen techniques.
.PP
The above\(hymentioned services shall, in general, be bi\(hydirectional, 
although uni\(hydirectional operation should be possible. Also, some of 
the 
facilities can be omitted, if not required, in order to minimize costs.
.PP
\fINote\fR \ \(em\ At the subscriber terminal, the use of ancillary equipments, 
e.g.\ for document reproduction, video tape recordings,\ etc., shall be 
possible.
.RT
.sp 2P
.LP
\fB3\fR 	\fBSystem parameters\fR 
.sp 1P
.RT
.sp 1P
.LP
3.1
	\fIPicture standards\fR 
.sp 9p
.RT
.PP
3.1.1
The 
video standards
of the subscriber sets shall be
compatible with, readily convertible to, or identical to, the local broadcast 
television standards. 
.bp
.sp 9p
.RT
.PP
3.1.2
Two classes of picture standards are recommended for the visual telephone 
system. They are given in Table\ 1/H.100. 
.sp 9p
.RT
.ce
\fBH.T. [T1.100]\fR 
.ce
TABLE\ 1/H.100
.ce
\fBPicture standards\fR 
.ps 9
.vs 11
.nr VS 11
.nr PS 9
.TS
center box;
cw(30p) | cw(66p) | cw(66p) sw(66p) , ^  | ^  | c | c.
Class	Items	 {
The region to which the figures should be applied
 }
		 {
Regions where TV broadcasting uses 25 pictures per second
 }	 {
Regions where TV broadcasting uses 30 pictures per
second
 }
_
.T&
lw(30p) | lw(66p) | lw(66p) | lw(66p) .
			
.TE
.nr PS 9
.RT
.ad r
\fBTable 1/H.100 [T1.100], p.\fR 
.sp 1P
.RT
.ad b
.RT
.LP
.sp 1
.PP
\fR Class \fIa\fR standards are identical to the local broadcast
video standards and will, in most cases, give sufficient definition for
real\(hytime picture transmission of a group of people (e.g.\ for conferencing)
and of graphics material.
.PP
Class \fIb\fR standards give sufficient definition for real\(hytime
transmission of a head and shoulder picture of one person or a small group. 
For the transmission of graphics information or other still pictures with 
high 
definition, a slow\(hyscan technique has to be applied. For instance, a system
using 625 or 525\ horizontal scanning lines and\ 5, or less, pictures per 
second which gives a Class\ \fIa\fR definition in the 1\ MHz bandwidth. 
.PP
Further study is required to define slow scanning parameters.
.RT
.sp 2P
.LP
\fB4\fR 	\fBCharacteristics relating to\fR 
\fBsplit\(hyscreen techniques\fR 
\fBfor Class\ \fR \fIa\fR 
\fBtelevision conference systems\fR 
.FS
Split\(hyscreen
techniques for systems using Class\ \fIb\fR standards require future study.
.FE
.sp 1P
.RT
.PP
In television conference systems which use split\(hyscreen techniques to 
make more effective use of the picture area, the following features for 
the terminals and transmitted signals are recommended. Preferred seating 
arrangement for such systems are given in Annex\ A.
.RT
.sp 1P
.LP
4.1
	\fIPicture format\fR 
.sp 9p
.RT
.PP
The transmitted picture should be 4 |  |  
aspect ratio
, split into upper and lower halves corresponding to the groups of seats. 
Viewed from the camera system, the left\(hyhand group should be in the 
upper half and the 
right\(hyhand group in the lower half.
.PP
The split should occur at the end of lines 166 and 479 for 625\(hyline
television systems and at the end of line\ 142 in Field\ 1 and line\ 141 in
Field\ 2 for 525\(hyline television systems, as shown in Figure\ 1/H.100.
.bp
.PP
Before display, the receive equipment may discard half lines and
first and last lines which are liable to be averaged during standards
conversion or vertical aperture correction of mixed signals.
.RT
.sp 2P
.LP
4.2
	\fIIdentification signal for split\(hyscreen system\fR 
.sp 1P
.RT
.sp 1P
.LP
4.2.1
	\fIAnalogue video signals\fR 
.sp 9p
.RT
.PP
The identification signal for split\(hyscreen system should be
inserted in the vertical blanking period, because the control is required 
for each television frame or field. 
.PP
The line where the identification signal is inserted and its
signal format are under study.
.RT
.sp 1P
.LP
4.2.2
	\fIDigital video signals\fR 
.sp 9p
.RT
.PP
An identification signal for split\(hyscreen system should be
provided. In the case of codecs in Recommendations\ H.120 and\ H.130 the 
format shall be that specified in Recommendation\ H.130. 
.RT
.sp 1P
.LP
4.3
	\fICompatibility with non\(hysplit\(hyscreen systems\fR 
.sp 9p
.RT
.PP
The simplest kind of a video telephone terminal is composed of a
single camera and other equipments. These terminals may be interconnected 
with split\(hyscreen system terminals. In that case, mechanical masks (if 
used) for 
the two split\(hyscreen displays (aspect ratio\ =\ 4 |  | .5) need to be 
removed, or if a display with 4 |  |  aspect ratio needs to be installed 
additionally. 
.RT
.sp 1P
.LP
4.4
	\fICameras and displays arrangement\fR 
.sp 9p
.RT
.PP
The entrance pupils of the TV camera optical system should be as
near as possible to the centre of the TV display showing remote conferees, 
in order to minimize errors in eye contact angle. 
.PP
Unless means are employed to place these pupils in line with the
display, e.g.\ by use of half\(hysilvered mirrors, the camera system should be
sited above the display and central to it.
.PP
In order to keep the maximum horizontal errors as small as possible, the 
cameras used had better be in a 
cross\(hyfire system
, as for example in
Figure\ A\(hy1/H.100, and the camera/display assembly should be sited on the
central axis of the terminal. However, in some cases, adoption of
parallel\(hyfire system
as shown also in Figure\ A\(hy1/H.100 is necessary
due to a restriction in equipment arrangement.
.PP
Whether the two cameras are arranged in cross\(hyfire or parallel\(hyfire 
is left open to each Administration since the selection does not affect 
the 
interconnection of different systems.
.RT
.sp 1P
.LP
4.5
	\fIPicture processing methods at transmitting terminals\fR 
.sp 9p
.RT
.PP
In order to obtain the correct relationship between the signals
from the two cameras for split\(hyscreen working, the cameras should be
synchronized but the vertical drive pulses should be rephased. The drive 
to one should be advanced by one quarter of the vertical period while the 
drive to the other should be retarded by the same amount. This causes a 
central strip of the target of each camera tube to be used and so minimizes 
the effects of 
distortions in the corners of the targets. Figure\ B\(hy1\fIa)\fR /H.100 
illustrates the preferred method. 
.PP
Alternative methods which are not recommended although they do not
give rise to problems of end\(hyto\(hyend compatibility are compared in
Annex\ B.
.RT
.sp 1P
.LP
4.6
	\fIReceiving equipment\fR 
.sp 9p
.RT
.PP
The receiving equipment should be capable of working with
discontinuities in the received signal that may be caused by switching 
between non\(hysynchronous video sources. 
.PP
\fINote\fR \ \(em\ A split\(hyscreen device should be capable of working with a
codec with the input and output frequency tolerances as specified in
Recommendation\ H.120.
.bp
.RT
.LP
.rs
.sp 47P
.ad r
\fBFIGURE 1/H.100, p.\fR 
.sp 1P
.RT
.ad b
.RT
.LP
.bp
.ce 1000
ANNEX\ A
.ce 0
.ce 1000
(to Recommendation H.100)
.sp 9p
.RT
.ce 0
.ce 1000
\fBSeating arrangements when applying split\(hyscreen techniques\fR 
.sp 1P
.RT
.ce 0
.ce 1000
\fBfor class\fR \fIa\fR \fBsystem\fR 
.ce 0
.PP
Preferred arrangements for video conferences using
split\(hyscreen techniques are:
.sp 1P
.RT
.PP
A.1
The conference terminal accommodation should be for 6 primary seats in 
two adjacent groups of\ 3 as shown in Figure\ A.1/H.100. 
.sp 9p
.RT
.PP
Provision for additional seating behind may be made, so long as
allowance is made for the central gap between the two halves. For example,
4\ additional persons may be seated on a second row as in the Figure.
.PP
A.2
The chairman's position should be in the centre of the
left\(hyhand group of seats (viewed from the camera) with user controls 
accessible from both this position and the one of the chairman's left. 
.sp 9p
.RT
.PP
Consequently, when split\(hyscreen pictures are displayed, stacked as received 
(i.e.\ shown as\ 3 over\ 3), the chairman's position is standardized as 
top centre. 
.PP
The suite of 3 chairs containing the chairman's position should also be 
regarded as the primary position for occasions when only half of a studio 
is in use. Such standardization is necessary for connection of 3\ studios 
in 
conference using time\(hydivision multiplex of pairs of TV signals to share a
common trunk between two studios.
.RT
.LP
.rs
.sp 31P
.ad r
\fBFigure A\(hy1/H.100, p.\fR 
.sp 1P
.RT
.ad b
.RT
.LP
.bp
.ce 1000
ANNEX\ B
.ce 0
.ce 1000
(to Recommendation H.100)
.sp 9p
.RT
.ce 0
.ce 1000
\fBPicture processing methods in transmitting terminals\fR 
.sp 1P
.RT
.ce 0
.PP
Alternative methods of obtaining the split\(hyscreen signal which are compatible 
with the recom\(hy 
mended\ method\ and which might be
useful for experiments and demonstrations are shown in\ \fIb)\fR and\ \fIc)\fR 
of 
Figure\ B\(hy1/H.100.\ In method\ \fIb)\fR , the two cameras are directed
upward and downward to pick up right
and left halves of the conferencing room, respectively. Since circumferences 
of target and scanning areas are used, geometric and brightness distortions 
tend to occur. In method\ \fIc)\fR , vertical deflection currents are biased 
by the 
quantity corresponding to \(+- | /4 of target height. Vertical deflection bias
adjustment is needed every time cameras are exchanged. In method\ \fIa)\fR 
, the 
vertical driving pulses are phase\(hyshifted by \(+- | /4\ V. The recommended 
method, \fIa)\fR , avoids the problems of methods\ \fIb)\fR and\ \fIc)\fR 
. 
.sp 1P
.RT
.LP
.bp
.LP
.rs
.sp 47P
.ad r
\fBFigure B\(hy1/H.100, p.\fR 
.sp 1P
.RT
.ad b
.RT
.LP
.bp
.sp 2P
.LP
\fBRecommendation\ H.110\fR 
.RT
.sp 2P
.ce 1000
\fBHYPOTHETICAL\ REFERENCE\ CONNECTIONS\ FOR\ VIDEOCONFERENCING\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.110''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.110	%'
.ce 0
.sp 1P
.ce 1000
\fBUSING\ PRIMARY\ DIGITAL\ GROUP\ TRANSMISSION\fR 
.ce 0
.sp 1P
.ce 1000
\fI(Malaga\(hyTorremolinos, 1984; amended at Melbourne, 1988)\fR 
.sp 9p
.RT
.ce 0
.sp 1P
.LP
	The\ CCITT,
.sp 1P
.RT
.sp 1P
.LP
\fIconsidering\fR 
.sp 9p
.RT
.PP
(a)
that there is growing evidence of a customer demand
for a videoconference service;
.PP
(b)
that circuits to meet this demand can, at present, be
provided effectively by digital transmission using the primary digital group;
.PP
(c)
that switched digital transmission networks known as the Integrated Digital 
Network (IDN) and Integrated Services Digital Network 
(ISDN) are under study, but the methods of exploiting
these networks for the transmission of primary digital groups will not 
become clear until the studies have progressed further; 
.PP
(d)
that the existence of different digital hierarchies and different television 
standards in different countries complicates the problems of defining hypothetical 
reference connections; 
.PP
(e)
that a hypothetical reference connection may be used as a guide to simplify 
the problems of connections between countries with 
different television standards and digital hierarchies,
.sp 1P
.LP
\fIappreciating\fR 
.sp 9p
.RT
.PP
that rapid advances are being made in research and development of video 
coding and bit\(hyrate reduction techniques which may lead to further 
Recommendations being proposed for hypothetical reference connections for
videoconferencing at bit rates which are multiples or sub\(hymultiples of
384\ kbit/s during subsequent study periods, so that this may be considered 
as the first of an evolving series of Recommendations, 
.sp 1P
.LP
\fInoting\fR 
.sp 9p
.RT
.PP
(a)
that a hypothetical reference connection is a model
in which studies relating to overall performance may be made, thereby allowing 
comparisons with standards and objectives; on this basis, limits for various 
impairments can be allocated to the elements of the connection;
.PP
(b)
that such a model may be used:
.LP
	\(em
	by an Administration to examine the effects on
transmission quality of possible changes of impairment
allocations in national networks,
.LP
	\(em
	by the CCITT for studying the allocation of
impairments to component parts of international networks,
.LP
	\(em
	 to test national rules for prima facie compliance with any impairment 
criteria which may be recommended by the 
CCITT for national systems;
.PP
(c)
that hypothetical reference connections are not to be
regarded as recommending particular values of impairments allocated to
constituent parts of the connection, and they are not intended to be used 
for the design of transmission systems, 
.sp 1P
.LP
\fIand recognizing\fR 
.sp 9p
.RT
.PP
that the planning of the necessary transmission networks for a
videoconference service will be facilitated if recommended hypothetical
reference connections are available, even if only in a preliminary form 
without details of all transmission and switching arrangements, 
.bp
.sp 1P
.LP
\fIrecommends\fR 
.sp 9p
.RT
.PP
The term \*Qintra\(hyregional\*U is used here to describe connections
within a group of countries which share a common television scanning standard 
and a common digital hierarchy, and may or may not be in geographical 
proximity. The term \*Qinter\(hyregional\*U is used here to describe connections 
between groups of countries which have different television scanning standards 
and/or different digital hierarchies. 
.FE
(1) 
that the hypothetical reference connection and means for
digital transmission illustrated in 
Figures\ 1/H.110 and 2/H.110
shall be used as the model for studies of the overall performance of
international videoconference connections, both intra\(hyregional
and
inter\(hyregional
, which are provided using minimum numbers of encoding and decoding equipments; 
.PP
(2)
that hypothetical reference connections of a more
complex type, as, for example, those illustrated in Figure\ 3/H.110, being
representative of many connections that may be employed in practice, should 
be studied further. 
.PP
\fINote\ 1\fR \ \(em\ The hypothetical reference connection shown in
Figure\ 1/H.110 contains the basic transmission elements, but is incomplete
because switching has been excluded and the local ends and parts of the
national network at each end of the connection have been left unspecified.
.PP
\fINote\ 2\fR \ \(em\ Because the arrangements of transmission systems
interconnecting regions using different digital hierarchies have not yet 
been standardized, and because videoconferencing is likely to be a minority 
service in such transmission systems, it seems prudent to consider videoconference 
connections both where the primary hierarchical level on the inter\(hyregional
link is 1.5\ Mbit/s and where it is 2\ Mbit/s. In Figure\ 2b/H.110, the change
between 2048\ kbit/s and 1544\ kbit/s transmission is placed at the 2048\ 
kbit/s end of the long international network. The long distance part of 
the connection is thus operated at the lower bit rate. Where the international 
network is 
provided on a system which uses the 2048\ kbit/s hierarchy, Figure\ 2c/H.110
maintains the efficiencies offered by the arrangement shown in Figure\ 
2b/H.110, by making available the six vacated time slots for other use. 
Figure\ 2d/H.110 offers the possibility of improved picture quality compared 
with 
Figures\ 2b/H.110 and 2c/H.110 by making full use of the available 2048\ 
kbit/s for the videoconferencing signal. This arrangement would require 
a 2048\ kbit/s codec compatible with 525\(hyline Video Standards, or the 
use of an external 
standards converter. This is for further study.
.PP
\fINote\ 3\fR \ \(em\ The lengths which have been assigned to the parts of the
connections have been arbitrarily chosen, but have some consistency with
existing CCITT and CCIR Recommendations. They are intended to be representative 
of long international connections, but not the longest possible. The lengths 
will likely require revision when studies on the error rates of digital 
paths have progressed to the stage when the error rates of the paths used 
in the 
connections can be predicted.
.PP
\fINote\ 4\fR \ \(em\ The propagation delay is one of the main factors to be
studied based upon the structures and lengths of the connections in
Figures\ 1/H.110, 2/H.110 and\ 3/H.110. However, in the absence of subjective
test results, the specification of requirements for videoconferencing
connections must await further study. This study and particularly operational 
experience are required to determine the extent to which Recommendation\ 
G.114, which applies to telephone connections, relates to videoconferencing 
connections.
.PP
\fINote\ 5\fR \ \(em\ In Figures 1/H.110 and 3/H.110, the codecs may be 
located anywhere within the international or national networks including 
the 
international gateway or the customer's premises.
.PP
\fINote\ 6\fR \ \(em\ The extensions beyond the codec shown as A or D in
Figures\ 1/H.110 and 3/H.110 may include wideband analogue or high\(hyspeed
digital transmission systems on terrestrial bearers. It is not expected that
these transmission systems will have any significant influence on the quality 
of the picture or sound, or, on the propagation delay, other than that 
due to their length. 
.PP
\fINote\ 7\fR \ \(em\ For inter\(hyregional operation, television standards
conversion between 525\(hyline and 625\(hyline video signals may be required. 
This 
conversion may be performed by the codecs themselves, or provided by external 
equipment. 
.PP
\fINote\ 8\fR \ \(em\ The arrangements shown in Figure 2/H.110 provide for the
simplest means of transmission. More complex means are possible and are not
precluded.
.PP
\fINote\ 9\fR \ \(em\ The hypothetical reference connection shown in
Figure\ 3/H.110 is of a more complex type than the connection shown in
Figure\ 1/H.110, in that it includes codecs in cascade, and, possibly an
external Television Standards Converter. The picture quality attainable with
these more complex connections may be degraded with respect to that attainable 
using the connection illustrated in Figure\ 1/H.110. This and other aspects 
of the more complex connection must be studied further. 
.bp
.RT
.LP
.rs
.sp 47P
.ad r
\fBFigure 1/H.110 and symbols for Figure 1/H.110, p.\fR 
.sp 1P
.RT
.ad b
.RT
.LP
.bp
.LP
.rs
.sp 47P
.ad r
\fBFigure 2/H.110 \ \ 
(\*`a l'italienne), p.\fR 
.sp 1P
.RT
.ad b
.RT
.LP
.bp
.LP
.rs
.sp 20P
.ad r
\fBSymbols for Figure 2/H.110, p.\fR 
.sp 1P
.RT
.ad b
.RT
.LP
.rs
.sp 25P
.ad r
\fBFIGURE 3/H.110, p. 8\fR 
.sp 1P
.RT
.ad b
.RT
.LP
.bp
.LP
.rs
.sp 32P
.ad r
\fBSYMBOLES DE LA FIGURE 3/H.110, p. 9\fR 
.sp 1P
.RT
.ad b
.RT
.sp 2P
.LP
\fBRecommendation\ H.120\fR 
.RT
.sp 2P
.ce 1000
\fBCODECS\ FOR\ VIDEOCONFERENCING\fR 
.EF '%	Fascicle\ III.6\ \(em\ Rec.\ H.120''
.OF '''Fascicle\ III.6\ \(em\ Rec.\ H.120	%'
.ce 0
.sp 1P
.ce 1000
\fBUSING\ PRIMARY\ DIGITAL\ GROUP\ TRANSMISSION\fR 
.ce 0
.sp 1P
.ce 1000
\fI(Malaga\(hyTorremolinos, 1984; amended at Melbourne, 1988)\fR 
.sp 9p
.RT
.ce 0
.sp 1P
.LP
	The\ CCITT,
.sp 1P
.RT
.sp 1P
.LP
\fIconsidering\fR 
.sp 9p
.RT
.PP
(a)
that there is growing evidence of a customer demand for a videoconference 
service; 
.PP
(b)
that circuits to meet this demand can, at present, be
provided effectively by digital transmission using the primary digital group;
.bp
.PP
(c)
that the existence of different digital hierarchies and different television 
standards in different parts of the world complicates 
the problems of specifying coding and transmission standards for international 
connections; 
.PP
(d)
that the eventual use of switched digital transmission networks should 
be taken into account, 
.sp 1P
.LP
\fIappreciating\fR 
.sp 9p
.RT
.PP
that rapid advances are being made in research and development of video 
coding and bit\(hyrate reduction techniques which may lead to further 
Recommendations being proposed for videoconferencing at bit rates which are
multiples or submultiples of 384\ kbit/s during subsequent study periods 
so that this Recommendation may be considered as the first of an evolving 
series of 
Recommendations,
.sp 1P
.LP
\fIand noting\fR 
.sp 9p
.RT
.PP
that it is a basic objective of the CCITT to recommend a unique
solution for international connections as far as possible,
.sp 1P
.LP
\fIrecommends\fR 
.sp 9p
.RT
.PP
that the codecs having signal processing and interface
characteristics described in \(sc\(sc\ 1, 2 and\ 3 below, should be used for
international videoconference connections.
.PP
\fINote\fR \ \(em\ Codecs of types other than those described in this
Recommendation are not precluded.
.RT
.sp 2P
.LP
	\fBIntroduction\fR 
.sp 1P
.RT
.PP
Section 1 of this Recommendation specifies the codec, developed for operation 
with the 625\(hyline, 50\ field/s television standard and the 2048\ kbit/s 
primary digital group. Its architecture has been chosen to permit variations 
in the detailed design of certain of the functional elements having the 
greatest influence on the picture quality. This enables future developments, 
aimed at 
improving the performance, to be incorporated without affecting the ability 
of different coders and decoders to interwork. For this reason, no details 
are 
given of such items as motion detectors or spatial and temporal filters. The
Recommendation confines itself to the details necessary to enable a decoder
correctly to interpret and decode the received signals.
.PP
The annexes to \(sc\ 1 which can be found at the end of this
Recommendation give details of some additional optional features which 
may be provided to supplement the basic design. 
.PP
Under the general heading of codecs not requiring separate television standards 
conversion when used on interregional connections, \(sc\ 2 describes a 
version of the codec for 525 line, 60 field/s and 1544\ kbit/s operation 
which also provides automatic television standards conversion when connected 
to 
the version of the codec described in \(sc\ 1 via a re\(hymultiplexing unit (to
convert between frame structures defined in \(sc\(sc\ 2.1 and\ 2.3 of
Recommendation\ G.704) at the junction of the 2048 and 1544\ kbit/s digital
paths. This codec is also suitable for use within regions using the 525\(hyline, 
60\ field/s television standard and 1544\ kbit/s transmission. 
.PP
Other implementations of \(sc\ 2 are to be studied, for
example:
.RT
.LP
	\(em
	a version of the codec for 625\(hyline, 50 field/s and
2048\ kbit/s operation capable of interworking with the codec
described in \(sc\ 3;
.LP
	\(em
	a version of the codec for 525\(hyline, 60 field/s and
2048\ kbit/s operation capable of interworking with the codec
described in \(sc\ 1.
.PP
Section 3 of the Recommendation describes a codec for
intra\(hyregional use in 525\(hyline, 60\ field/s and 1544\ kbit/s regions.
.PP
The frame structures associated with the codecs described in this
Recommendation are to be found in Recommendation\ H.130.
.PP
As the codecs are complex items using combined intraframe and
interframe picture\(hycoding techniques which tend to be known only to
specialists, Appendix\ I is provided giving a brief outline of the principles
involved in the codecs of \(sc\(sc\ 1 and \(sc\ 2.
.bp
.RT
.sp 2P
.LP
\fB1\fR \fBA codec for 625\(hylines, 50 fields/s and 2048 kbit/s transmission 
\fBfor intra\(hyregional\fR 
.FS
The term \*Qintra\(hyregional\*U is used here to
describe connections within a group of countries which share a common
television scanning standard and a common digital hierarchy, and may or 
may not be in geographical proximity. The term \*Qinter\(hyregional\*U 
is used here to 
describe connections between groups of countries which have different
television scanning standards and/or different digital hierarchies.
.FE
.PS 10
\fBuse\fR \fB and capable of interworking with the codec of \(sc\ 2\fR 
.sp 1P
.RT
.sp 1P
.LP
1.1\fR 	\fIScope\fR 
.sp 9p
.RT
.PP
Section 1 defines the essential features of a codec for the digital transmission, 
at 2048\ kbit/s, of signals for videoconference or visual 
telephone service in accordance with Recommendation\ H.100. The video input 
to the coder and output from the decoder in a 625\(hyline, 50\ field/s 
signal, 
according to the \*QClass\ \fIa\fR \*U standard of Recommendation\ H.100, or
alternatively, the 313\(hyline, 50\ field/s signal of the \*QClass\ \fIb\fR 
\*U standard. 
Provision is also made for a sound channel and optional data channels. 
A brief description of the operation of the codec is given in Appendix\ 
I. 
.PP
The Recommendation starts with a brief specification of the codec
(\(sc\ 1.2) and a description of the video interface. This is followed 
by details of the source coder (\(sc\ 1.4) which provides analogue\(hyto\(hydigital 
conversion 
followed by recoding with substantial redundancy reduction in the face\(hyto\(hyface 
mode. Paragraph\ 1.5 deals with the video multiplex coder which inserts 
instructions and addresses into the digitized video signal to control the
decoder so that it correctly interprets the signals received. Paragraph\ 1.6
describes
the transmission coder which arranges the various digital signals (video,
sound, data, signalling) into a form compatible with Recommendation\ G.732 
for transmission over 2048\ kbit/s digital paths. Paragraph\ 1.7 describes 
optional forward 
error correction facilities. Provision is made in the digital frame structure 
for the inclusion of other optional facilities such as a graphics mode, 
encryption and multipoint conferencing. Details of such facilities as
are at present available are given in the annexes to this Recommendation.
.RT
.sp 2P
.LP
1.2\fR 	\fIBrief specification\fR 
.sp 1P
.RT
.sp 1P
.LP
1.2.1
	\fIVideo input/output\fR 
.sp 9p
.RT
.PP
The video input and output are standard 625\(hyline, 50 field/s colour 
or monochrome television signals. The colour signals are in, or converted 
to, component form. Colour and monochrome operation are fully compatible. 
.RT
.sp 1P
.LP
1.2.2
	\fIDigital output/input\fR 
.sp 9p
.RT
.PP
The digital output and input are at 2048 kbit/s, compatible with
the frame structure of Recommendation\ G.704.
.RT
.sp 1P
.LP
1.2.3
	\fISampling frequency\fR 
.sp 9p
.RT
.PP
The video sampling frequency and the 2048 kHz network clock are
asynchronous.
.RT
.sp 1P
.LP
1.2.4
	\fICoding techniques\fR 
.sp 9p
.RT
.PP
Conditional replenishment coding supplemented by adaptive digital filtering, 
differential PCM and variable\(hylength coding are used to achieve low 
bit\(hyrate transmission. 
.RT
.sp 1P
.LP
1.2.5
	\fIAudio channel\fR 
.sp 9p
.RT
.PP
An audio channel using 64 kbit/s is included. At present, coding
is A\(hylaw according to Recommendation\ G.711, but provision is made for 
future 
use of more efficient coding.
.RT
.sp 1P
.LP
1.2.6
	\fIMode of operation\fR 
.sp 9p
.RT
.PP
The normal mode of operation is full duplex.
.RT
.LP
.sp 1
.bp
.sp 1P
.LP
1.2.7
	\fICodec\(hyto\(hynetwork signalling\fR 
.sp 9p
.RT
.PP
An optional channel for codec\(hyto\(hynetwork signalling is included.
This conforms to emerging ideas in CCITT for switching 2\(hyMbit/s paths in the
ISDN.
.RT
.sp 1P
.LP
1.2.8
	\fIData channels\fR 
.sp 9p
.RT
.PP
Optional 2 \(mu 64 kbit/s and 1 \(mu 32 kbit/s data channels are
available. These are used for video if not required for data.
.RT
.sp 1P
.LP
1.2.9
	\fIForward error correction\fR 
.sp 9p
.RT
.PP
Optional forward error correction is available. This is required
only if the long\(hyterm error rate of the channel is worse than\ 1 in 10\u6\d.
.RT
.sp 1P
.LP
1.2.10
	\fIAdditional facilities\fR 
.sp 9p
.RT
.PP
Provision is made in the digital frame structure for the future
introduction of encryption, a graphic mode and multipoint facilities.
.RT
.sp 1P
.LP
1.2.11
	\fIPropagation delay\fR 
.sp 9p
.RT
.PP
When the coder buffer is empty and the decoder buffer full, the
coder delay is less than\ 5\ ms and the decoder delay is 130\ \(+-\ 30\ 
ms at 2\ Mbit/s or 160\ \(+-\ 36\ ms when only 1.5\ Mbit/s are in use 
.FS
These are typical figures. The delays depend on the detailed implementation 
used. 
.FE
.
.RT
.sp 1P
.LP
1.3\fR 	\fIVideo interface\fR 
.sp 9p
.RT
.PP
The normal video input is a 625\(hyline, 50 field/s signal in accordance 
with CCIR\ Recommendation\ 472. When colour is being transmitted, the input 
(and output) video signals presented to the analogue/digital convertors (and
from the digital/analogue convertors) are in colour\(hydifference component 
form. The luminance and colour\(hydifference components, E`\dY\u, 
(E`\dR\u\ \(em\ E`\dY\u) and (E`\dB\u\ \(em\ E`\dY\u) are as defined in 
CCIR Report\ 624. 
The analogue video input (and output) interface with the codec may be in the
form of colour\(hydifference components, colour components (R,\ G,\ B) or as a
composite colour signal. The video interface is as recommended in
CCIR\ Recommendation\ 656.
.PP
Optionally, any other video standard which can be converted to give
143 active lines per field may be used.
.RT
.sp 2P
.LP
1.4\fR 	\fISource coder\fR 
.sp 1P
.RT
.sp 1P
.LP
1.4.1
	\fILuminance component\fR \fIor\fR 
\fImonochrome\fR 
.sp 9p
.RT
.sp 1P
.LP
1.4.1.1
	\fIAnalogue\(hyto\(hydigital conversion\fR 
.sp 9p
.RT
.PP
The signal is sampled to produce 256 picture samples per active
line (320\ samples per complete line). The sampling pattern is orthogonal and
line, field and picture repetitive. For the 625\(hyline input, the sampling
frequency is 5.0\ MHz, locked to the video waveform.
.PP
Uniformly quantized PCM with 8 bits/sample is used.
.PP
Black level
corresponds to level 16 (00010000).
.PP
White level
corresponds to level 239 (11101111).
.PP
PCM code words outside this range are forbidden (the codes being used for 
other purposes). For the purposes of prediction and interpolation, the 
final picture element in each active line (i.e.\ picture element\ 255) 
is set to level\ 128 in both encoder and decoder. 
.PP
In all arithmetic operations, 8\(hybit arithmetic is used and the bits
below the binary point are truncated at each stage of division.
.RT
.LP
.sp 1
.bp
.sp 1P
.LP
1.4.1.2
	\fIPre\(hy and post\(hyfiltering\fR 
.sp 9p
.RT
.PP
In addition to conventional 
anti\(hyaliasing filtering
prior   to analogue\(hyto\(hydigital conversion, a 
digital transversal filtering
operation is carried out on the 625\(hyline signal to reduce the 
vertical
definition of the picture
prior to 
conditional replenishment
coding
. As a result of this process, 143\ active lines per field are used
instead of the 287\(12\ active lines of the 625\(hyline signal, although 
the effective vertical definition is greater than one\(hyhalf of that of 
a normal 625\(hyline 
display. An interpolation process in the decoder restores the 625\(hyline 
signal waveform. 
.RT
.sp 1P
.LP
1.4.1.3
	\fIConditional replenishment coding\fR 
.sp 9p
.RT
.PP
A 
movement detector
identifies 
clusters of picture
elements
which are deemed to be moving. The basic feature is a frame memory which 
stores 2\ fields of 143\ lines, each line containing 256\ addressable 
points. The memory is updated at the picture rate and differences between 
the incoming signal and the corresponding stored values are used to determine 
the moving area in the coder. A similar 
frame memory
must exist at the
decoder and be similarly updated under the control of addressing information
received from the coder. It is not necessary to specify the techniques 
used for movement detection because they do not affect interworking, although 
they do 
affect the resultant picture quality.
.PP
Detected 
moving areas
are transmitted by differential PCM with a maximum of 16\ quantization 
levels. The first picture element in each moving area is transmitted by 
PCM. Variable\(hylength coding is used on the DPCM code 
words.
.PP
The first picture element of each cluster and the complete PCM lines, when 
they are transmitted to provide systematic or forced updating, are coded 
in accordance with \(sc\ 1.4.1.1. 
.RT
.sp 1P
.LP
1.4.1.3.1\ \ 
\fIDPCM prediction algorithm\fR 
.sp 9p
.RT
.PP
The algorithm used for DPCM prediction is:
\v'6p'
.RT
.sp 1P
.ce 1000
X = 
[Formula Deleted]
, where X is the sample being predicted.
(See Figure 1/H.120.)
.ce 0
.sp 1P
.LP
.sp 1
.rs
.sp 10P
.ad r
\fBFigure\ 1/H.120, p.\fR 
.sp 1P
.RT
.ad b
.RT
.PP
.sp 2
For the purpose of prediction, line and field blanking are assumed to be 
at level\ 128 (out of\ 256). 
.sp 1P
.LP
1.4.1.3.2\ \ \fIQuantization law and\fR 
\fIvariable\(hylength coding\fR 
.sp 9p
.RT
.PP
511 input levels are quantized to a maximum of 16 output levels.
The quantizer does not assume the use of modulo\ 256 arithmetic.
.PP
The quantization law and associated 
variable\(hylength codes
which are used for both luminance and colour\(hydifference picture elements in
moving areas which are not horizontally subsampled are given in
Table\ 1/H.120.
.bp
.RT
.ce
\fBH.T. [T1.120]\fR 
.ce
TABLE\ 1/H.120
.ce
\fBCode table for non\(hyhorizontally\(hysubsampled moving
.ce
\fBareas\fR 
.ps 9
.vs 11
.nr VS 11
.nr PS 9
.TS
center box;
cw(42p) | cw(42p) | cw(60p) | cw(36p) .
Input levels	Output levels	Variable\(hylength code	Code No.
_
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\(em255 to \(em125	\(em141	1 0 0 0 0 0 0 0 0 1	17
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\(em124 to \(em\ 95	\(em108	1 0 0 0 0 0 0 0 1	16
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\(em\ 94 to \(em\ 70	\(em\ 81	1 0 0 0 0 0 0 1	15
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\(em\ 69 to \(em\ 49	\(em\ 58	1 0 0 0 0 0 1	14
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\(em\ 48 to \(em\ 32	\(em\ 39	1 0 0 0 0 1	13
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\(em\ 31 to \(em\ 19	\(em\ 24	1 0 0 0 1	12
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\(em\ 18 to \(em\ \ 9	\(em\ 13	1 0 1	10
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\(em\ \ 8 to \(em\ \ 1	\(em\ \ 4	1 1	\ 9
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\ \ \ 0 to \ \ \ 7	+\ \ 3	0 1	\ 1
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\ \ \ 8 to \ \ 17	+\ 12	0 0 1	\ 2
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\ \ 18 to \ \ 30	+\ 23	0 0 0 1	\ 3
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\ \ 31 to \ \ 47	+\ 38	0 0 0 0 1	\ 4
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\ \ 48 to \ \ 68	+\ 57	0 0 0 0 0 1	\ 5
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\ \ 69 to \ \ 93	+\ 80	0 0 0 0 0 0 1	\ 6
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\ \ 94 to \ 123	+107	0 0 0 0 0 0 0 1	\ 7
.T&
cw(42p) | cw(42p) | lw(60p) | cw(36p) .
\ 124 to \ 255	+140	0 0 0 0 0 0 0 0 1	\ 8
_
.TE
.nr PS 9
.RT
.ad r
\fBTable 1/H.120 [T1.120], p.\fR 
.sp 1P
.RT
.ad b
.RT
.PP
The 
end\(hyof\(hycluster code
is 1\ 0\ 0\ 1 and is designated as
code number\ 11. The end\(hyof\(hycluster code is omitted at the end of 
the last 
cluster in a line irrespective of whether it is a 
luminance cluster
or  a 
colour\(hydifference cluster
.
.sp 1P
.LP
1.4.1.4
	\fISubsampling\fR 
.sp 9p
.RT
.PP
As the buffer fills, horizontal subsampling and field/field
subsampling are introduced.
.RT
.sp 1P
.LP
1.4.1.4.1\ \ 
\fIHorizontal subsampling\fR 
.sp 9p
.RT
.PP
Horizontal subsampling is carried out only in moving areas.
Normally, in this mode, only even elements are transmitted on even numbered
lines and odd elements on odd numbered lines. This gives rise to a line
quincunx pattern
in moving areas.
.PP
Omitted elements are interpolated in the decoder by averaging the two horizontally 
adjacent elements. 
.PP
Interpolated picture elements are placed in the frame stores. A
moving area cluster
will always start with a PCM value and finish with a transmitted DPCM picture 
element, even during subsampling. This means that in in some instances, 
the transmitted cluster needs to be extended by one element in comparison 
with the moving area declared by the movement detector. At the 
end of the active line, however, this cannot occur as clusters must not 
extend into blanking, so cluster shortening by one element can be necessary. 
.PP
Adaptive element subsampling
allows the transmission of
normally omitted elements, either to remove interpolation errors or, to 
provide a softer switch to subsampling and thus improve the picture quality. 
The 
signalling of the extra elements is achieved by using, on horizontally
subsampled lines only, 8\ quantizing levels for normally transmitted elements
and the remaining 8\ levels for the extra elements. Also, a cluster can 
finish either on a normally transmitted element or an \*Qextra\*U element. 
.bp
.PP
During horizontally subsampled lines, the quantization law and
variable\(hylength code shown in Table\ 2/H.120 will be used for both luminance 
and colour\(hydifference samples in moving areas. 
.RT
.ce
\fBH.T. [T2.120]\fR 
.ce
TABLE\ 2/H.120
.ce
\fBQuantization law and variable\(hylength code table\fR 
.ps 9
.vs 11
.nr VS 11
.nr PS 9
.TS
center box;
lw(72p) | lw(156p) .
	
.T&
cw(72p) | cw(156p) .
Quantization	Variable\(hylength codes
.TE
.TS
center box ;
cw(36p) | cw(36p) | cw(54p) | cw(24p) | cw(54p) | cw(24p) .
Input range	Output levels	Normal elements	Code No.	Extra elements	Code No.
_
.T&
cw(36p) | cw(36p) | lw(54p) | cw(24p) | lw(54p) | cw(24p) .
\(em255 to \(em41	\(em50	1 0 0 0 0 0 0 1	15	1 0 0 0 0 0 0 0 0 1	17
.T&
cw(36p) | cw(36p) | lw(54p) | cw(24p) | lw(54p) | cw(24p) .
\(em\ 40 to \(em24	\(em31	1 0 0 0 0 1	13	1 0 0 0 0 0 0 0 1	16
.T&
cw(36p) | cw(36p) | lw(54p) | cw(24p) | lw(54p) | cw(24p) .
\(em\ 23 to \(em11	\(em16	1 0 1	10	1 0 0 0 0 0 1	14
.T&
cw(36p) | cw(36p) | lw(54p) | cw(24p) | lw(54p) | cw(24p) .
\(em\ 10 to \(em\ 1	\(em\ 5	1 1	\ 9	1 0 0 0 1	12
.T&
cw(36p) | cw(36p) | lw(54p) | cw(24p) | lw(54p) | cw(24p) .
\ \ \ 0 to +\ 9	+\ 4	0 1	\ 1	0 0 0 1	\ 3
.T&
cw(36p) | cw(36p) | lw(54p) | cw(24p) | lw(54p) | cw(24p) .
\ \ 10 to \ 22	+15	0 0 1	\ 2	0 0 0 0 0 1	\ 5
.T&
cw(36p) | cw(36p) | lw(54p) | cw(24p) | lw(54p) | cw(24p) .
\ \ 23 to \ 39	+30	0 0 0 0 1	\ 4	0 0 0 0 0 0 0 1	\ 7
.T&
cw(36p) | cw(36p) | lw(54p) | cw(24p) | lw(54p) | cw(24p) .
\ \ 40 to \ 255	+49	0 0 0 0 0 0 1	\ 6	0 0 0 0 0 0 0 0 1	\ 8
_
.TE
.nr PS 9
.RT
.ad r
\fBTable 2/H.120 [T2.120], p.\fR 
.sp 1P
.RT
.ad b
.RT
.LP
.sp 2
.PP
With regard to prediction, if element A is a non\(hytransmitted
element in a moving area, it is replaced by A\ds\u(see Figure\ 1/H.120); if
element\ D is part of a subsampled moving area, and not transmitted in the
current frame, it is replaced by\ C.
.sp 1P
.LP
1.4.1.4.2\ \ 
\fIField/field subsampling\fR 
.sp 9p
.RT
.PP
Either field can be omitted. In the omitted field, interpolation
takes place only in those parts of the picture which are estimated to be
moving. \*QStationary\*U areas remain unchanged.
.PP
The estimated moving areas are formed from an OR function on the
moving areas in the past and future fields, as shown in Figure\ 2/H.120. 
In the figure, x is a moving element if a OR b OR c OR d are moving. 
.RT
.LP
.rs
.sp 15P
.ad r
\fBFigure 2/H.120, p.\fR 
.sp 1P
.RT
.ad b
.RT
.LP
.bp
.PP
For the purpose of field interpolation, PCM lines are considered as non\(hymoving 
and field blanking is assumed to be at a level of 128 out 
of\ 256.
.PP
In the interpolator for monochrome or luminance signals, the
operations 
[Formula Deleted]
carried
\-v'8p'
out before the combined average is taken. Thus
\v'6p'
.RT
.sp 1P
.ce 1000
\fIx\fR = 
[Formula Deleted]
.ce 0
.sp 1P
.PP
.sp 1
The interpolated values are placed in the frame store.
.sp 2P
.LP
1.4.2
	\fIColour\(hydifference components\fR 
.sp 1P
.RT
.sp 1P
.LP
1.4.2.1
	\fIAnalogue\(hyto\(hydigital conversion\fR 
.sp 9p
.RT
.PP
The signal is sampled to produce 52 samples per active line
(64\ samples per complete line). The sampling pattern is orthogonal and 
line\(hy, field\(hy and picture\(hyrepetitive. For the 625\(hyline input, 
the sampling frequency is\ 1.0\ MHz, locked to the video waveform. 
.PP
The (E`\dR\u\ \(em\ E`\dY\u) and (E`\dB\u\ \(em\ E`\dY\u)
samples are sited so that the centre of the first
colour\(hydifference sample on any line is co\(hysited with the centre 
of the third luminance sample (addressed as number\ 2). The (E`\dR\u\ \(em\ 
E`\dY\u) 
and (E`\dB\u\ \(em\ E`\dY\u) signals are stored and transmitted
on alternate lines of the coded picture. The first active line of Field 
No.\ 1 contains (E`\dB\u\ \(em\ E`\dY\u) and the first active line of 
Field\ No.\ 2 contains (E`\dR\u\ \(em\ E`\dY\u). The colour difference
signal not being transmitted during any line is obtained at the decoder by
interpolation.
.PP
The vertical filtering (see \(sc 1.4.2.2) is arranged so that the
effective vertical positions of the colour\(hydifference samples in each of the
286\ active lines coincide with those of the corresponding luminance samples.
.PP
Uniformly quantized PCM with 8 bits/sample is used.
.PP
The (E`\dR\u\ \(em\ E`\dY\u) and (E`\dB\u\ \(em\ E`\dY\u)
signals are quantized using \(+- | 11 steps with zero signal corresponding to
level\ 128. The analogue video signals are amplitude\(hylimited
so that the digitized signals do not go outside that range (corresponding to
levels\ 16 to\ 239). The video levels are set so that a 100/0/75/0 colour bar
signal (see CCIR Recommendation\ 471 for explanation of nomenclature) will
occupy levels\ 17 to\ 239.
.PP
As for the luminance signal, forbidden PCM code words are available
for purposes other than transmitting video sample amplitudes.
.RT
.sp 1P
.LP
1.4.2.2
	\fIPre\(hy and post\(hyfiltering\fR 
.sp 9p
.RT
.PP
In addition to conventional anti\(hyaliasing filtering prior to
analogue\(hydigital conversion, a digital transversal filtering operation is
carried out on the 625\(hyline signal to reduce the vertical definition of the
picture prior to conditional\(hyreplenishment coding. As a result of this 
process, 72\ active lines of (E`\dR\u\ \(em\ E`\dY\u) and\ 71\ active lines 
of (E\dB\u\ \(em\ E`\dY\u) are used in Field No.\ 2 instead of 287\(12
active lines per field of a 625\(hyline signal. Similarly, Field No.\ 1 
contains 
72\ active lines of (E`\dB\u\ \(em\ E`\dY\u) and 71\ lines of
(E`\dR\u\ \(em\ E`\dY\u). An interpolation process in the decoder
restores the 625\(hyline signal waveforms.
.RT
.sp 1P
.LP
1.4.2.3
	\fIConditional replenishment coding\fR 
.sp 9p
.RT
.PP
Coloured moving areas
are detected, coded and addressed
separately from the 
luminance moving areas
, but the same principles are employed.
.PP
Detected moving areas are transmitted by differential PCM with a
maximum of 16 quantization levels. The first picture element in each moving
area is transmitted by PCM. Variable\(hylength coding is used on the DPCM code
words.
.PP
Complete PCM lines are transmitted to provide systematic and forced
updating coincident with luminance PCM lines.
.bp
.RT
.sp 1P
.LP
1.4.2.3.1\ \ 
\fIDPCM prediction algorithm\fR 
.sp 9p
.RT
.PP
The algorithm used for colour\(hydifference signals is:
.RT
.sp 1P
.ce 1000
x = A (see Figure 1/H.120)
.ce 0
.sp 1P
.LP
1.4.2.3.2\ \ \fIQuantization law and variable\(hylength coding\fR 
.sp 9p
.RT
.PP
As for luminance component (see \(sc\(sc 1.4.1.3.2 and 1.4.1.4.1).
.RT
.sp 1P
.LP
1.4.2.4
	\fISubsampling\fR 
.sp 9p
.RT
.PP
Horizontal subsampling is carried out in exactly the same way as
for the luminance signal, including adaptive element subsampling.
.PP
Field/field subsampling of the colour\(hydifference signals is also
similar to that of the luminance signal. Either field can be omitted and, in
the omitted field, interpolation takes place only in those parts of the 
picture which are estimated to be moving. Stationary areas remain unchanged. 
.PP
The estimated moving areas are formed by an OR function on moving
areas in past and future fields in the same manner as for luminance
(\(sc\ 1.4.1.4.2).
.PP
For colour\(hydifference signals, the interpolated value of x
is 
[Formula Deleted]
Field\ 1 or 
\-v'8p'
Field\ 2, respectively.
.PP
Both field and horizontal subsampling take place simultaneously with subsampling 
of the luminance signal and they are signalled to the decoder in 
the same way.
.RT
.sp 2P
.LP
1.5\fR 	\fIVideo multiplex coding\fR 
.sp 1P
.RT
.sp 1P
.LP
1.5.1
	\fIBuffer store\fR 
.sp 9p
.RT
.PP
The size of the buffer store is defined at the transmitting end
only and is 96\ kbit/s. Its delay is approximately equal to the duration of
one picture (40\ ms).
.PP
At the receiving end, the buffer must be of at least this length, but in 
some implementations of the decoder it may be longer. 
.RT
.sp 1P
.LP
1.5.2
	\fIVideo synchronization\fR 
.sp 9p
.RT
.PP
The method used for video synchronization permits the retention of the 
picture structure. The required information is transmitted in the form 
of line start and field start codes (LST and FST). 
.RT
.sp 1P
.LP
1.5.2.1
	\fILine start code\fR 
.sp 9p
.RT
.PP
The line start code includes a synchronization word, a line number code 
and a digit to signal the presence of element subsampling. 
.PP
It has the format:
.RT
.LP
	0 0 0 0 0 0 0 0 | 0 0 0 0 1 0 0 0 | \*QS\*U | 3\(hybit line No. code |
.PP
\*QS\*U is a 1 if horizontal subsampling occurs on the TV line
following the line start code. \*QS\*U is a \*Qdon't care\*U condition 
on empty or PCM lines. 
.PP
The line number code comprises the least three significant digits of the 
line number, where Line\ 0\ =\ first active line of Field\ 1 and 
Line\ 144\ =\ first active line of Field\ 2.
.PP
Lines numbered 143 and 287 are non\(hycoded lines, used for field
synchronization and line number continuity.
.RT
.sp 1P
.LP
1.5.2.2
	\fIField start code\fR 
.sp 9p
.RT
.PP
There are two field start codes, FST\(hy1 and FST\(hy2, where the first 
line of the field following FST\(hy2 is interlaced between the first two 
lines of the field following FST\(hy1. FST\(hy1 indicates the start of 
the first field, 
starting with line number\ 0. FST\(hy2 indicates the start of the second field,
starting with line number\ 144, as shown in Figure\ 3/H.120.
.bp
.RT
.LP
.rs
.sp 36P
.ad r
\fBFigure 3/H.120, p.\fR 
.sp 1P
.RT
.ad b
.RT
.PP
Each field start code comprises a line start code, followed by an 8\(hybit 
word, followed by the line start code of the first line of the next 
field.
.PP
The field start code is given in Figure 4/H.120.
.RT
.LP
.rs
.sp 6P
.ad r
\fBFigure 4/H.120 [T3.120] \ \ 
(\*`a traiter comme tableau MEP), p.\fR 
.sp 1P
.RT
.ad b
.RT
.LP
.bp
.PP
For FST\(hy1, F = 1 and for FST\(hy2, F = 0. A = 0 for normal
operation. If required, A = 1 is used to signal that the buffer state is 
less than 6\ kbits (used in switched multipoint applications). S is the 
subsampling digit as defined in \(sc\ 1.5.2.1. 
.PP
Field subsampling is signalled by two consecutive field start codes of 
the same number. For example: 
.RT
.sp 1P
[Formula Deleted]
.ce 1000