CEN/TS 15844-2:2010

Postal services - ID-tagging of letter mail items - Part 2: BNB-78 Encoding Specification

This part of the Technical Specification defines the representation of ID-tags as a 78-position bar-no-bar code (BNB-78) printed in fluorescent ink in area R1 on the reverse side of items.
BNB-78 encoding is one of two encoding specifications supported by this Technical Specification  ) for the printing of ID-tags in area R1, the other being BNB-62, which is specified in CEN/TS 15844-3.
NOTE 1   Representation in the form of a 4-state code printed on the front of the item is covered in CEN/TS 15844-4 for flats and CEN/TS 15844-5 for small letters.
BNB-78 encoding supersedes the earlier specified BNB-62 encoding and shall be applied in all cases in which ID-tags are placed in area R1 on the reverse side of letter mail items of size up to and including C5, by issuers other than those explicitly authorised to continue use of BNB-62 encoding, namely An Post (Ireland), Canada Post and United States Postal Service.
NOTE 2   ID-tags encoded in area R1 are required by article RL 123 of the UPU Letter Post Regulations [1] to be compliant with UPU standard S18 – and by this with the related CEN/TS 15844. This supports only two encodings in area R1, namely BNB-78 as defined herein and BNB-62 as defined in CEN/TS 15844-3. The latter is authorised for continued use only by the three issuers mentioned above. Where ID-tags are used, and are applied in area R1 on the reverse side of letter mail items of size up to and including C5, the use of BNB-78 encoding is mandatory for all other issuers.
NOTE 3   BNB-78 encoding is not considered suitable for use on flats. CEN/TS 15844-4 defines a 4-state encoding which may be used for this purpose.

Postalische Dienstleistungen – ID-Kennzeichnung von Briefsendungen – Teil 2: Festlegung der BNB-78-Codierung

Traitement automatisé des envois postaux - Chronomarquage des envois postaux - Partie 2: Spécification de codage en BNB (Bar No Bar)- 78 caractères

Poštne storitve - Označevanje poštnih pošiljk z identifikacijsko številko - 2. del: Specifikacija kodiranja BNB-78

Ta del tehnične specifikacije določa prikaz identifikacijske številke v obliki 78-mestne kode BNB-78, ki je natisnjena s fluorescentnim črnilom na območju R1 na hrbtni strani pošiljk. Kodiranje BNB-78 je ena od dveh specifikacij kodiranja za tiskanje identifikacijskih številk na območju R1, ki ju podpira ta tehnična specifikacija (3); druga specifikacija je BNB-62, ki jo določa standard CEN/TS 15844-3. OPOMBA 1: Prikaz v obliki 4-statusne črtne kode, natisnjene na sprednji strani navadnih pisem, obravnavata standarda CEN/TS 15844-4 za večja pisma in CEN/TS 15844-5 za navadna pisma. Kodiranje BNB-78 nadomešča prej specificirano kodiranje BNB-62 in se uporablja v vseh primerih, kjer identifikacijske številke na območje R1 na hrbtno stran poštne pošiljke do vključno velikosti C5 namestijo izdajatelji, ki jim ni izrecno dovoljena nadaljnja uporaba kodiranja BNB-62, kamor sodijo An Post (Irska), Canada Post in United States Postal Service. OPOMBA 2: Identifikacijske številke, ki se nahajajo na območju R1, morajo biti po členu RL 123 Pravilnika o pisemski pošti UPU [1] v skladu s standardom UPU S18 – in s tem s standardom CEN/TS 15844. Ta podpira le dve kodiranji na območju R1, zlasti BNB-78, kot je določen v tej specifikaciji, in BNB-62, kot je določen v standardu CEN/TS 15844-3. Slednjega lahko še naprej uporabljajo le zgoraj navedeni trije izdajatelji. Če se identifikacijske številke uporabljajo na območju R1 na hrbtni strani poštne pošiljke do vključno velikosti C5, je uporaba kodiranja BNB-78 obvezna za vse ostale izdajatelje. OPOMBA 3: Kodiranje BNB-78 ni ustrezno za večja pisma. Standard CEN/TS 15844-4 določa 4-statusno črtno kodo, ki se lahko uporablja za ta namen.

General Information

• Status: Published
• Publication Date: 21-Dec-2010
• ICS: 03.240 - Postal services
• Technical Committee: CEN/TC 331 - Postal services
• Drafting Committee: CEN/TC 331/WG 3 - Automatic identification of items - Addresses
• Current Stage: 9093 - Decision to confirm - Review Enquiry
• Start Date: 05-Dec-2023
• Completion Date: 14-Apr-2025
• Ref Project: SIST-TS CEN/TS 15844-2:2013 - Postal services - ID-tagging of letter mail items - Part 2: BNB-78 Encoding Specification

Overview

CEN/TS 15844-2:2010 specifies the BNB-78 (78-position bar-no-bar) encoding for ID-tags on letter mail. It defines how ID-tags are represented as a 78-position bar-no-bar code printed in fluorescent ink in area R1 on the reverse side of letter items up to and including C5 size. The Technical Specification ensures global interoperability and compliance with UPU requirements (RL 123 / UPU standard S18).

Key topics and technical requirements

• BNB-78 format: 78-position bar-no-bar code printed in fluorescent ink in area R1 on the reverse of letter items.
• Scope and limitations:
  • Applies to letter mail items up to and including C5; not suitable for flats (4‑state code specified in CEN/TS 15844-4).
  • BNB-78 supersedes BNB-62 for most issuers; BNB-62 (CEN/TS 15844-3) remains authorised only for An Post (Ireland), Canada Post and the United States Postal Service.
  • Format identifier limited to 18A.
• Issuer and value constraints:
  • Issuers must use an issuer code in the range NNN–ZZZ. New issuers request codes from the UPU Standards Board (allow ~3 months lead time).
  • Domain and equipment identifier ranges restricted (domain codes {0–9, A–C}, second equipment character fixed at 0, third in {0–9, A–C}) to control coding density.
• Bar-code physical characteristics:
  • Minimum pitch ~1.3 mm (density limit); effectively ≤80 bar positions including margins.
  • No runs of more than four bar-positions without bars for reader synchronisation.
  • Fluorescent ink optical characteristics and printing parameters specified to support reliable reading.
• Error handling and reading:
  • Includes error detection and correction algorithms (informative annex) enabling partial reads and corrective decoding.
  • Procedures for discrimination from other barcodes, validation, and conversion to message/binary formats (linked with CEN/TS 15844-1).

Applications and users

• Postal operators and national posts implementing UPU S18-compliant ID-tagging.
• Mail processing equipment manufacturers (printers, sorters, scanners) implementing BNB-78 printing/reading.
• Mail integrators, fulfilment and bulk-mail service providers applying machine-readable ID-tags in area R1.
• Software developers and systems integrators converting BNB-78 into message/binary interchange formats.

Related standards

• CEN/TS 15844-1 - ID-tag structure, message and binary representation
• CEN/TS 15844-3 - BNB-62 encoding specification (limited continued use)
• CEN/TS 15844-4 / -5 - 4-state codes for flats and small letters
• UPU S18, UPU S31 and UPU Letter Post Regulations (RL 123)

Keywords: CEN/TS 15844-2:2010, BNB-78, ID-tagging, postal services, bar-no-bar, R1 area, fluorescent ink, UPU S18, issuer code, error correction.

Standards Content (Sample)

SLOVENSKI STANDARD
01-maj-2013
3RãWQHVWRULWYH2]QDþHYDQMHSRãWQLKSRãLOMN]LGHQWLILNDFLMVNRãWHYLONRGHO
6SHFLILNDFLMDNRGLUDQMD%1%
Postal services - ID-tagging of letter mail items - Part 2: BNB-78 Encoding Specification
Postalische Dienstleistungen – ID-Kennzeichnung von Briefsendungen – Teil 2:
Festlegung der BNB-78-Codierung
Traitement automatisé des envois postaux - Chronomarquage des envois postaux -
Partie 2: Spécification de codage en BNB (Bar No Bar)- 78 caractères
Ta slovenski standard je istoveten z: CEN/TS 15844-2:2010
ICS:
03.240 Poštne storitve Postal services
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

TECHNICAL SPECIFICATION
CEN/TS 15844-2
SPÉCIFICATION TECHNIQUE
TECHNISCHE SPEZIFIKATION
December 2010
ICS 03.240
English Version
Postal services - ID-tagging of letter mail items - Part 2: BNB-78
Encoding Specification
Traitement automatisé des envois postaux - Postalische Dienstleistungen - ID-Kennzeichnung von
Chronomarquage des envois postaux - Partie 2: Briefsendungen - Teil 2: Festlegung der BNB-78-Codierung
Spécification de codage en BNB (Bar No Bar)- 78
caractères
This Technical Specification (CEN/TS) was approved by CEN on 1 December 2008 for provisional application.

The period of validity of this CEN/TS is limited initially to three years. After two years the members of CEN will be requested to submit their
comments, particularly on the question whether the CEN/TS can be converted into a European Standard.

CEN members are required to announce the existence of this CEN/TS in the same way as for an EN and to make the CEN/TS available
promptly at national level in an appropriate form. It is permissible to keep conflicting national standards in force (in parallel to the CEN/TS)
until the final decision about the possible conversion of the CEN/TS into an EN is reached.

CEN members are the national standards bodies of Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia,
Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland,
Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland and United Kingdom.

EUROPEAN COMMITTEE FOR STANDARDIZATION
COMITÉ EUROPÉEN DE NORMALISATION

EUROPÄISCHES KOMITEE FÜR NORMUNG

Management Centre: Avenue Marnix 17, B-1000 Brussels
© 2010 CEN All rights of exploitation in any form and by any means reserved Ref. No. CEN/TS 15844-2:2010: E
worldwide for CEN national Members.

Contents Page
Foreword .3
Introduction .4
1 Scope .5
2 Normative references .5
3 Terms and definitions .5
4 Symbols and abbreviations .5
5 Usage limitations – Requirement for a special issuer code .6
6 Value range limitations .6
7 Encoding specification .7
7.1 Data to be encoded .7
7.2 Data field encoding table .9
7.3 Calculation of the error detection and correction bits. 10
7.4 ECC field encoding table . 11
7.5 Construction of the bar code to be printed on the item . 12
8 Printing of the bar code . 13
8.1 Optical characteristics of the ink . 13
8.2 Bar code characteristics . 14
9 Reading and interpretation of BNB-78 bar codes . 16
9.1 Discrimination from other bar codes . 16
9.2 Fully read bar codes . 16
9.3 Partially read bar codes . 16
9.4 Checking data field values . 16
9.5 Manual interpretation of bar codes . 17
10 Conversion to the message and binary representations . 17
Annex A (informative) Error detection and correction algorithms . 19
A.1 Introduction . 19
A.2 Calculation of error control groups . 19
A.3 Checking the read value for errors . 19
A.4 Correcting an erroneous bar code . 20
A.5 Pre-processing . 21
Annex B (informative) S18 ID-tag 78-position BNB bar code template . 23
Bibliography . 25

Foreword
This document (CEN/TS 15844-2:2010) has been prepared by Technical Committee CEN/TC 331 “Postal
services”, the secretariat of which is held by NEN.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent rights.
CEN [and/or CENELEC] shall not be held responsible for identifying any or all such patent rights.
NOTE This document has been prepared by experts coming from CEN/TC 331 and UPU, under the framework of the
Memorandum of Understanding between the UPU and CEN.
1)
This document (CEN/TS 15844-2), is the CEN equivalent of UPU standard S18b-7. It may be amended only after
prior consultation, between CEN/TC 331 and the UPU Standards Board, in accordance with the Memorandum of
Understanding between CEN and the UPU.
2)
The UPU’s contribution to the document was made, by the UPU Standards Board and its sub-groups, in
accordance with the rules given in Part V of the "General information on UPU standards".
According to the CEN/CENELEC Internal Regulations, the national standards organizations of the following
countries are bound to announce this Technical Specification: Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech
Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania,
Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden,
Switzerland and the United Kingdom.

1)
The Universal Postal Union (UPU) is the specialized institution of the United Nations that regulates the universal postal service. The postal
services of its 189 member countries form the largest physical distribution network in the world. Some 5 million postal employees working in
over 660 000 post offices all over the world handle an annual total of 425 billion letters-post items in the domestic service and almost 6,7 billion
in the international service. Some 4,5 billion parcels are sent by post annually. Keeping pace with the changing communications market, posts
are increasingly using new communication and information technologies to move beyond what is traditionally regarded as their core postal

business. They are meeting higher customer expectations with an expanded range of products and value-added services.
2)
The UPU's Standards Board develops and maintains a growing number of standards to improve the exchange of postal-related information
between posts, and promotes the compatibility of UPU and international postal initiatives. It works closely with posts, customers, suppliers and
other partners, including various international organizations. The Standards Board ensures that coherent standards are developed in areas such
as electronic data interchange (EDI), mail encoding, postal forms and meters. UPU standards are published in accordance with the rules given

in Part VII of the General information on UPU standards, which may be freely downloaded from the UPU world-wide web site (www.upu.int).

Introduction
A general introduction to all parts of the specification is provided in CEN/TS 15844-1. This part deals only with the
encoding of ID-tags in the form of a 78-position bar-no-bar code, BNB-78, printed on the reverse side of items, in
area R1, using fluorescent ink. It is arranged under six main headings:
Clause No. Description of content
5 Usage limitations – requirement for a special issuer code: explains the requirement for issuers wishing
to use the BNB-78 ID-tag representation to obtain a special issuer code for this purpose;
6 Value range limitations: defines limitations on the values of data elements used in ID-tags which are to
be represented on items in the form of a BNB-78 bar code;
7 Encoding specification: specifies the construction of a 78-position bar-no-bar code from ID-tag data
elements;
8 Printing of the bar code: to allow the association of computer data with a physical item, the ID-tag is
printed on the item itself. This clause defines required ink and printing parameters;
9 Reading and interpretation of BNB-78 bar codes: specifies the validation and error correction
requirements associated with the reading of ID-tags represented using BNB-78 bar codes;
10 Conversion to the message and binary representations: describes the correspondence between
BNB-78 representation and the binary and message interchange representations defined in
CEN/TS 15844-1.
The above definition is supported by two informative annexes:
 Error detection and correction algorithms: provides example implementations of the error detection and
correction algorithms used in encoding and reading the BNB-78 representation of the ID-tag. Other
implementations are possible and, depending on the environment, might well be more efficient. The examples
nevertheless serve as a possible baseline, and can be used for verifying the correctness of a different
implementation.
 S18 ID-tag 78-position BNB bar code template: provides a template which can be used for manual decoding of
the data elements in printed BNB-78 representations of an ID-tag. Such manual decoding should be used with
caution since, unless the complete bar code is read and processed through the appropriate error
detection/correction algorithm, there is no certainty that the value obtained has been read correctly.
1 Scope
This part of the Technical Specification defines the representation of ID-tags as a 78-position bar-no-bar code
(BNB-78) printed in fluorescent ink in area R1 on the reverse side of items.
3)
BNB-78 encoding is one of two encoding specifications supported by this Technical Specification for the
printing of ID-tags in area R1, the other being BNB-62, which is specified in CEN/TS 15844-3.
NOTE 1 Representation in the form of a 4-state code printed on the front of the item is covered in CEN/TS 15844-4 for
flats and CEN/TS 15844-5 for small letters.
BNB-78 encoding supersedes the earlier specified BNB-62 encoding and shall be applied in all cases in which
ID-tags are placed in area R1 on the reverse side of letter mail items of size up to and including C5, by issuers
other than those explicitly authorised to continue use of BNB-62 encoding, namely An Post (Ireland), Canada
Post and United States Postal Service.
NOTE 2 ID-tags encoded in area R1 are required by article RL 123 of the UPU Letter Post Regulations [1] to be
compliant with UPU standard S18 – and by this with the related CEN/TS 15844. This supports only two encodings in area
R1, namely BNB-78 as defined herein and BNB-62 as defined in CEN/TS 15844-3. The latter is authorised for continued
use only by the three issuers mentioned above. Where ID-tags are used, and are applied in area R1 on the reverse side of
letter mail items of size up to and including C5, the use of BNB-78 encoding is mandatory for all other issuers.
NOTE 3 BNB-78 encoding is not considered suitable for use on flats. CEN/TS 15844-4 defines a 4-state encoding
which may be used for this purpose.
2 Normative references
The following referenced documents are indispensable for the application of this document. For dated
references, only the edition cited applies. For undated references, the latest edition of the referenced
document (including any amendments) applies.
CEN/TS 15844-1:2010, Postal services — ID-tagging of letter mail items — Part 1: ID-tag structure, message
and binary
3 Terms and definitions
A number of common terms used in this document are defined in documents referred to in Normative
References and in the Bibliography. Definition of frequently used or particularly important terms as well as
other terms introduced in this document are given below.
See CEN/TS 15844-1:2010, Postal services — ID-tagging of letter mail items — Part 1: ID-tag structure,
message and binary
4 Symbols and abbreviations
See CEN/TS 15844-1:2010, Postal services — ID-tagging of letter mail items — Part 1: ID-tag structure,
message and binary
3)
References to "this Technical Specification" should be interpreted as references to CEN/TS 15844 as a whole, not
only to Part 2.
5 Usage limitations – Requirement for a special issuer code
The allocation of ID-tags and their application to items using BNB-78 representation is restricted to issuers to
whom an issuer code in the range {NNN, …, ZZZ} has been assigned. Issuers wishing to commence use of
the ID-tag should request the UPU Standards Board secretariat to issue them with such a code at least three
months prior to the planned implementation date.
NOTE 1 There are no constraints on the reading and use of ID-tags. Any mail handling organisation with appropriate
reading equipment may read ID-tags on items and use these for their intended purposes. However, the allocation of ID-
tags and their encoding on items is restricted to licensed issuers which have requested and obtained an appropriate issuer
code, together with their designated agents.
NOTE 2 Issuer codes are allocated in accordance with UPU standard S31. The Standards Board secretariat will need
information on the applicant’s mail processing infrastructure (number of main processing facilities, sorting offices, etc.) in
order to determine the number of issuer codes required and to select the most appropriate code value(s).
NOTE 3 The supported range allows for the identification of up to 2 197 (13*13*13) issuers, out of the range of 33 696
covered by UPU standard S31. Many high-volume generators of ID-tags, with multiple processing facilities, are expected
to require several issuer codes for ID-tag issuance. Thus, the number of issuers which may be licensed to use BNB-78
representation is limited.
6 Value range limitations
In addition to the component value limitations defined in CEN/TS 15844-1, the following limitations apply to ID-
tags which are to be represented in the form of BNB-78 bar codes:
NOTE 1 Given current technology, it is necessary to limit the coding density of bar-no-bar (BNB) codes to a pitch of
1,3 mm or more (less than 7,7 bars per centimetre). The BNB-78 representation has been defined so as to fit on a
minimum-sized envelope (127 mm in the USA). With allowance for leading and trailing space, this results in the
requirement for the BNB encoding of an ID-tag to be 80 bar positions or less. Furthermore, for reader synchronisation
reasons, there is a need to ensure that the bar code does not contain runs of more than four bar-positions without bars.
Lastly, to ensure a high degree of readability in practice, the bar code contains error detection and correction information,
enabling it to be read even if some of the bars have been obliterated or are otherwise unreadable. These requirements
result in the need to limit the value ranges of certain fields, to utilise a specific coding scheme based on groups of four
bars, and to place the data elements in a specific sequence. The value limitations are described below; the coding scheme
and data element sequence are described in Clause 7.
 the format identifier is limited to the value 18A;
 issuers are required to have, and use, an issuer code in the range NNN–ZZZ;
 domain codes are limited to the range {0, 1, …, 9, A, B, C};
 the second character of the equipment identifier (after the domain code) is limited to value 0; the third
character is limited to the range {0, 1, …, 9, A, B, C};
NOTE 2 This allows for up to 13 domains, each of which may have up to 13 items of equipment. Issuers for which
the allowed range of domain codes and equipment identifiers is insufficient may apply for additional issuer codes and
may then use the combination of issuer code and domain code to distinguish between an extended number of
domains.
 priority is limited to the values N and L;
NOTE 3 The first value, N, should be used for normal and high priority letters and as a default value where the
priority is considered not relevant or is unknown at the time of ID-tag allocation. Value L may, at the discretion of the
issuer, be used to indicate lower-than-normal priority.
 the item number is limited to the range 00000–15378;
 tracking indicator is limited to the values T and N.
The normal specification allows for 13 ID-tag allocation systems per domain, each with an item number range
of 15 379 values per 10-minute interval. It is permitted, but not recommended, to divide the coding range of
199 927 possible values offered by the combination of the last character of equipment identifier and the item
number differently. If it is known that the maximum processing capacity of the equipment is, for example,
below 10 000 items per 10-minute interval, this coding range may be split to allow support for 20 allocation
systems, 19 with an item number range of 10 000 items and one with a range of 9 927 items. However, for
representation purposes, the combined numbering range of 199 927 possibilities still has to be split into 13
ranges of 15 379 values, with the result that "equipment identifier" ceases to actually identify the unit of
equipment which allocated the ID-tag.
NOTE 4 That is, where a combined number range is used, the value to be used as third character of equipment
identifier is INT(extended item number/15 379) and not the real unit number or the equipment. Similarly, the value to be
used as ID-tag item number is |extended item number| and not the allocation system's counter value. For example, if
15 379
the extended range is split across 20 systems as in the example above, then the thirteenth system's item counter value
4327 would be represented as having equipment identifier third character value of INT(((13-1)  10 000 + 4 327)/15 379) =
INT(124 327/15 379) = 8, whilst the item number would be |12 4327| = 1 295.
15 379
NOTE 5 Other divisions are also possible, e.g. into 15 ranges of 12 500 and one of 12 427. Again, this is not
recommended because the "equipment identifier" ceases to actually identify the system responsible for allocating and
printing the ID-tag.
7 Encoding specification
7.1 Data to be encoded
The ID-tag components shall be converted into an array of 14 4-bit fields, numbered F to F , as follows:
0 13
NOTE 1 This clause defines the data to be included in the bar code. Because of limitations on bar code length and
pitch, it is not possible to encode the ID-tag information in the same format as is used in either the message or binary
representations of the ID-tag (as defined in CEN/TS 15844-1). The following specification is designed to make conversion
between the different representations as simple as possible.
NOTE 2 The UPU identifier is not encoded in the BNB representations of ID-tags: given the restrictions on the
encoding of area R1 which are defined in UPU standard S19 and in article RL 123 of the Letter Post Regulations [1], it is
presumed that any fluorescent BNB bar code, found in area R1, which correctly decodes in accordance with this Technical
Specification, is an S18 ID-tag.
F : a combination of the format identifier with the tracking indicator and the first (tens) digit of the day number
of the generation date, generated as follows:
Format Identifier Tracking Indicator First Digit of Day Value of F
18A T 0 1111
1 1110
2 1101
3 1100
N 0 1011
1 1010
2 1001
3 1000
NOTE 3 The BNB-78 representation supports only one value of format identifier, namely 18A, and two values of the
tracking indicator. The above effectively represents format identifier in the most significant bit of the field, with the
tracking indicator represented in the second bit and the first digit of the day in the least significant two bits. This
allows for future support of a second value of format identifier.
F : the second (units) digit of the day number of generation date (value 0 to 9), converted to a 4-bit value
using Table 2 (in 7.2);
EXAMPLE 1 In generation date value 000229, the units digit of the day number is 9. Look-up of this in the table
results in the 4-bit value 0101 (the table entry which corresponds to value 9).
F : the month number of the generation date (value 1 to 12), converted to a 4-bit value using Table 2;
NOTE 4 Note that, in order to limit the length of the bar code, the year is not encoded. It is assumed that items do
not remain in the postal system for more than 1 year and that, therefore, the year may be assumed to be:
a) the current year, if the current month and date are the same as, or later than, those encoded in the ID-tag;
b) the previous year, if the current month and date are earlier than those encoded in the ID-tag.
F : the generation time hour, based on a twelve-hour clock (value 0-11), converted to a 4-bit value using
Table 2;
F : the ten-minute interval number (first minute digit) of the generation time, plus either 0 for a.m. or 7 for
p.m., with the result (in the range 0 to 5 or 7 to 12) converted to a 4-bit value using Table 2;
NOTE 5 This effectively represents a.m./p.m. in the most significant bit of the field, with the ten-minute interval
number (i.e. the first digit of the time in minutes) being represented in the least significant three bits. For example, the
generation time value 17:40 results in an F4 value of 0011 (the table entry which corresponds to the value 11,
obtained by adding 4, the first minute digit, and 7, representing p.m.).

F : the first character of the issuer code, converted to a numeric value using Table 1 below and then to a
4-bit value using Table 2;
Table 1 — Issuer code character look-up table
Character N O P Q R S T U V W X Y Z
Look-up value 12 11 10 9 8 7 6 5 4 3 2 1 0
NOTE 6 All three characters of the issuer code should be in the range N–Z. If this is not the case, the value range
limitations defined in Clause 9 have not been adhered to and the ID-tag value cannot be represented as a BNB-78
bar code.
F : the second character of the issuer code, converted to a numeric value using Table 1 and then to a 4-bit
value using Table 2;
F : the third character of the issuer code, converted to a numeric value using Table 1 and then to a 4-bit
value using Table 2;
F : the domain code (first character of the equipment identifier), converted to a 4-bit value using Table 2;
NOTE 7 This should be in the range 0–C. If this is not the case, the value range limitations defined in Clause 9 have
not been adhered to and the ID-tag value cannot be represented as a BNB-78 bar code.
F : the third character of equipment identifier, converted to a 4-bit value using Table 2;
NOTE 8 This should be in the range 0–C. If this is not the case, the value range limitations defined in Clause 9 have
not been adhered to and the ID-tag value cannot be represented as a BNB-78 bar code.
NOTE 9 The second character of the equipment identifier, which should always be 0, is not represented.
F : INT(item number/1 183), i.e. the result of integer division of the item number by 1 183, converted to a
4-bit value using Table 2;
F : INT(|item number| /91), i.e. the result of integer division of the modulus 1 183 value of item number by
11 1183
91, converted to a 4-bit value using Table 2;
F : INT(|item number| /7), i.e. the result of integer division of the modulus 91 value of item number by 7,
12 91
converted to a 4-bit value using Table 2;
F : a combination of |item number| , i.e. the modulus 7 value of item number and the item priority, generated
13 7
as follows:
Item Priority Modulus 7 value of Item Number Value of F
N 0 1111
(Normal 1 1101
or unknown 2 1011
Priority) 3 1001
4 0111
5 0101
6 0011
L 0 1110
(Low 1 1100
Priority) 2 1010
3 1000
4 0110
5 0100
6 0010
NOTE 10 Priority should be N or L; if it is not, then the limitations described in Clause 6 have not been applied
and the ID-tag cannot be represented in BNB-78 form.
NOTE 11 This effectively represents priority in the least significant bit of the field, with the last character of the
serial number (limited to the range 0 to 6) in the most significant three bits.
NOTE 12 The combination 1000 is, exceptionally, supported in this field and in the corresponding part of the bar
code. This is possible because the bar code representation of F is followed by the ECC fields, which always contain
a 1 (bar) in the first and/or second bar positions.
EXAMPLE 2 The ID-tag with message representation J18APZW601N110714505956N (as used in Part A) results in
the following data to be encoded:
F F F F F F F F F F F F F F
0 1 2 3 4 5 6 7 8 9 10 11 12 13
1011 0111 0011 1101 0010 0100 1111 1100 1001 1110 1010 1111 1010 1001
7.2 Data field encoding table
To ensure that there are no runs of more than four no-bars, data to be encoded are divided into 4-bit fields,
encoded according to the following table, in which bars are represented by 1, spaces by 0:
Table 2 — Data field encoding look-up table
Data value to be encoded Encoded representation in 4-bit field
0 1111
1 1110
2 1101
3 1100
4 1011
5 1010
6 1001
7 0111
8 0110
9 0101
A or 10 0100
B or 11 0011
C or 12 0010
D or 13 0001
E or 14 1000
NOTE 1 Since the pitch of the bar code could vary slightly, long runs of spaces could result in de-synchronisation: it
might then become impossible for the reader to tell how many spaces there were between two bars. To overcome this, the
coding system is designed to ensure that there are never more than four consecutive spaces.
NOTE 2 As can be seen, in the data section, only the hexadecimal values 0 to E can be encoded and other values or
characters are not permitted in the present version of the standard. Values D and E are only permitted in certain
fields/positions. As discussed in Clause 9, this imposes restrictions on the ranges of values for identifying issuers, their
domains and equipment.
NOTE 3 The codes are not a simple complement of the binary data value – the field encoding 1000 (bar followed by
three spaces) is placed at the end of the table because it can be used only in the first and last data fields of the ID-tag, in
which two or more data elements are combined. It is always followed by a field encoding containing a bar in the first and/or
second position.
NOTE 4 The field encoding 0000 (four spaces) cannot occur in the present version of the standard, but may occur in
the first field in a future version, using a different format identifier, associated with a 0 bit in the most significant position of
F . It would always be preceded by the start bar and followed by an encoded value beginning with a bar, so would not
breach the requirement to avoid runs of more than four consecutive spaces.
7.3 Calculation of the error detection and correction bits
The ID-tag data shall, for the purposes of representation on the item by means of a BNB-78 bar code, be
protected by the incorporation of error control codes, calculated in accordance with the following algorithm:
NOTE 1 It is not sufficient to simply record the ID-tag data content as a BNB pattern, as read errors would make the
result unusable. The algorithm defined uses Reed-Solomon encoding in GF23 to calculate 20 ECC bits which are inserted
at the end of the bar code. When the ultimate bar code is read, these bits are used to provide an error detection and
correction capability, allowing the ID-tag to be read even when it is imperfectly printed and/or partly obliterated.
NOTE 2 The specified algorithm results in four 5-bar check fields, providing a minimum Hamming distance of 5. This
means that at least five groups of four (in the data part) or five (in the ECC part) bars must be changed in order to convert
a valid ID-tag into a different valid ID-tag. As long as no more than two such groups are incorrectly read, and as long as
4)
only two groups are changed in the error correction process, the result must be the correct original bar code. Similarly,

4)
This does not imply that an ID-tag value obtained by applying only one or two corrections to the captured BNB bar
code representation is necessarily correct. If the actual number of errors which occurred exceeds the capacity of the error
protection algorithm, the application of even a single correction may (but is unlikely to) result in an apparently valid, but
incorrect, value.
up to four groups can be corrected provided that it is known which groups were unreadable (erased), or read in error. This
results in the following characteristics:
a) correction of errors occurring in any two of fields F to F and E to E ;
0 13 14 17
b) correction of erasure of up to four fields.
NOTE 3 If the number of errors exceeds the error correction capacity of the algorithm, e.g. if there are read errors in
three groups in unknown locations, then the specified algorithm will not result in the original bar code, but might yield an
apparently valid result. This means that the algorithm is potentially sensitive to shifting of the bar code in case of erasure
of or failure to recognise the start bar. In theory, this could give rise to misreads. However, these are normally detected by
the validation mechanisms defined in Clause 9 and in the clause on Reading and validation of ID-tags on items in
CEN/TS 15844-1. Operational experience confirms that the ID-tag misread rate is low.
NOTE 4 It should be stressed that the algorithm specified below is expressed in a simple mathematical form.
Computer-coded implementations need not follow the same logic, provided that the bar pattern which results is identical to
that resulting from the algorithm given here. Details of a possible C-language implementation of the alogarith are provided
in A.2.
1 Treating each 4-bit element of F as a binary number (in the range 2 to 15, with 8 being possible only in F
and F ), construct the polynomial:
(17−n)
()
15−F ×
∑ n
n=0
2 Calculate the remainder, modulus 23, on dividing the above polynomial by the generator polynomial:
4 3 2
× +2× +10× +17×+9
3 Let this remainder be:
3 2
R × +R × +R ×+R
14 15 16 17
4 Calculate:
N =[23 -R ] for j = 14 to 17
j j mod23
5 Encode the values of N to N into an array of four 5-bit binary fields E to E using Table 3 (in 7.4).
14 17 14 17
EXAMPLE If R = 7, then N = 16 (23 – 7) and E = 01110 (table entry corresponding to data value 16).
14 14 14
7.4 ECC field encoding table
To ensure that there are no runs of more than four no-bars, ECC data to be encoded are divided into 5-bit
fields, encoded according to the following table, in which bars are represented by 1, spaces by 0:
Table 3 — ECC field encoding look-up table
Data value to be encoded Encoded representation in 5-bit ECC field
0 11111
1 11110
2 11101
3 11100
4 11011
5 11010
6 11001
7 10111
8 10110
9 10101
A = 10 10100
B = 11 10011
C = 12 10010
D = 13 10001
E = 14 11000
F = 15 01111
G = 16 01110
H = 17 01101
I = 18 01100
J = 19 01011
K = 20 01010
L = 21 01001
M = 22 01000
NOTE 1 Since the pitch of the bar code could vary slightly, long runs of spaces could result in de-synchronisation: it
might then become impossible for the reader to tell how many spaces there were between two bars. To overcome this, the
coding system is designed to ensure that there are never more than four consecutive spaces.
NOTE 2 Five bar fields are used only in the error correction section of the bar code, which, due to the use of modulus
23 in the calculation algorithm, may only contain values in the range 0 to 22.
NOTE 3 To maintain consistency with the data field encoding table, the codes are not a simple complement of the
binary data value. In particular, the encoding for values E (14) and M (22) fall outside the sequence used elsewhere in the
table.
7.5 Construction of the bar code to be printed on the item
A 78 bar position bar code shall be constructed by concatenating:
 a single start bar (l);
 the contents of Fields F to F (in that sequence), with the binary value of each being represented by four
0 13
bars using the convention one (1) = bar (l); zero (0) = space ( ) and with the most significant bit
represented first;
 the four Reed-Solomon error correction values E to E , encoded in accordance with the above
14 17
algorithm and in that sequence, with each being represented as a sequence of five bars using the
convention one (1) = bar (l); zero (0) = space ( ) and with the most significant bit represented first;
 a single stop bar (l).
In the ID-tag, bits and bars are numbered from left to right.
EXAMPLE (not to scale)
Figure 1 — Complete S18 ID-tag, showing the BNB-78 representation, including ECC bars, together
with the message representation
The resulting 78-position bar-no-bar code shall be encoded on the item, using fluorescent ink, in accordance
with the specifications in Clause 8, to be met by at least 99 % of production examples
8 Printing of the bar code
8.1 Optical characteristics of the ink
The ink shall exhibit fluorescence when excited by at least one and preferably each of the following light
sources:
NOTE 1 The choice for fluorescent ink is dictated by practical considerations, including the availability of existing
equipment and the fact that postcards (which would be damaged by encoding on the reverse side with highly visible inks)
can represent up to 15 % of cross-border and/or seasonal mail from certain locations.
 ultra-violet light with a wavelength of 350 nm to 400 nm;
NOTE 2 This case covers also halogen lamps equipped with filters designed to exclude visible wavelengths.
 blue light with a wavelength of 450 nm to 480 nm.
NOTE 3 These two types of light source are those commonly encountered on fluorescent ink bar code readers.
Practical experience indicates that most bar codes will be readable in most readers, even if the reader light source
does not optimally match the characteristics of the ink. However, to optimise readability by other mail handling
organisations, who might have readers with a light source which differs from that used by the ID-tag issuer, it is
recommended that the ink used has a broad excitation spectrum, or can be excitable by both UV-A light and blue
light. The use of an ink with only a single narrow excitation wavelength range might result in ID-tags printed on poor
or busy backgrounds being unreadable in readers with a light source which does not include the range concerned.
When excited, the ink shall fluoresce with peak emission wavelength in the range 580 nm to 605 nm.
The ink used shall comply with the health and safety regulations applicable to the organisation performing the
printing.
NOTE 4 Inks based on a solvent such as alcohol or MEK dry more rapidly and give a better signal than water-based
inks, particularly on glossy and highly absorbent surfaces. However, solvent-based inks are more toxic and carry fire risks.
The use of modern printing systems that limit environmental emissions and the taking of specific fire precautions can help
in this regard but it is essential that all relevant heath and safety regulations are followed.
The standard does not specify any requirement for human visibility of the ink in ambient lighting conditions.
However, experience shows that the addition of a light dye, enabling the presence or absence of an ID-tag to
be easily detected, is beneficial and this is therefore recommended. Such a dye should not be too
pronounced, to avoid defacing of postcards.
8.2 Bar code characteristics
a) Bar height: (4,0 ± 0,4) mm;
NOTE 1 The range allows for variation in the distance between the item and the ink jet.
b) Bar width (thickness): nominally 0,4 mm, with allowed variation between 0,3 mm and 0,6 mm;
NOTE 2 The range allows for different absorption rates/drying times on different materials.
NOTE 3 The combination of nominal bar-width with the specification of density (pitch) and pitch variation implies
that the space between bars must be greater than 0,617 9 mm – i.e., greater than the width of the bars themselves.
c) Encoding density (pitch): nominal pitch of 1,33 mm (i.e. approx. 7,5 bars per centimetre). The total length
of the bar code (78 bars including start and stop bar and error correction bars) shall lie within the range
10,0 cm to 10,7 cm;
NOTE 4 The lower limit has been selected as corresponding to the minimum pitch which can be expected to give
acceptable read results. It is based on tests which concluded that a pitch of down to 1,3 mm gave no appreciable
degradation of performance compared with the pitch of 1,67 mm used in the USPS domestic ID-tag. Pitches below
1,2 mm were found to give a performance degradation; in between (in the range of 1,2 mm to 1,3 mm), performance
could be maintained, but only if equipment was carefully adjusted.
NOTE 5 The upper limit corresponds to the space available (after allowing for leading and trailing spaces) on a
minimum size US envelope (i.e. 127 mm; standard letter mail has a minimum dimension of 140 mm).
d) Pitch variation: not greater than 5 %;
NOTE 6 The pitch is not permitted to vary by more than 5 % across the whole bar code.
e) Placement: on the reverse side of the item, in position R1 as defined by UPU Standard S19. No other
fluorescent material should appear in the first 128 mm of area R1;
NOTE 7 Position R1 is an area on the reverse side of the item, which has been reserved exclusively for the
placement of an S18 ID-tag in accordance with this Technical Specification.
NOTE 8 The placement of an ID-tag constructed in accordance with this Technical Specification at some other
location on the item is, subject to its not contravening other UPU standards, not precluded. However, ID-tags placed
in such other locations might well not be readable by parties other than the issuer.
f) Vertical position:
1) equipment shall be adjusted to print the tops of the bars at a nominal position of (13 ± 1) mm from
the bottom of the item;
2) on production samples, the complete bar code shall lie between parallel lines situated 4 mm and
16 mm from the bottom of the item;
NOTE 9 The permitted range accommodates possible variations in the height of the bars and a small degree of
item skew.
NOTE 10 The specified range applies across all items and does not mean that the code on a single item can be
spread over a vertical distance of 12 mm. On any single item, the skew limitations given below limit the vertical
spread much more (to approx. 7,7 mm if all bars are 4 mm in height).
g) Horizontal position:
1) first (start) bar situated (15 ± 1) mm from the leading edge;
2) last (stop) bar situated at least 5 mm from the trailing edge;
NOTE 11 At the maximum allowed pitch, the full 78-element code occupies 107 mm and can therefore be
accommodated, with necessary leading and trailing spaces, on envelopes of 127 mm (USPS minimum) or more in
length (provided that the start bar is not more than 15 mm from the leading edge on a minimum-sized envelope).
h) Skew:
1) the angle between the centre line of the bar code and the bottom of the mail piece shall not exceed
2° (horizontal skew);
2) the angle between each bar and the perpendicular to the centre line shall not exceed 5° (vertical
skew);
3) the sum of horizontal and vertical skew shall not exceed 5°.
NOTE 12 Skew limitations are illustrated in the following diagram:

Key
a positive skew
b negative skew
c horizontal skew
d vertical skew
Figure 2 — Illustration of skew limitations
9 Reading and interpretation of BNB-78
...