User:Greg240170/sandbox

4-State Bar Code

The symbology for the 4-state bar code was created by the Royal Mail Research and Development Imaging and Interpretation group led by Greg Hopkins. The code is used by Royal Mail to encode unique letter identification and Postal delivery information. The symbology was also adopted for use by customers of Royal Mail as the RM4SCC. Paul Kimmens was responsible for the development of the RM4SCC using the 4 state code symbology.

The symbology allows for a single upright bar to encode one of four possible states or may be considered as two separate binary digits.

It is important to distinguish between the code printed by Royal Mail during the processing of letter mail (which is currently (2018) an orange fluorescent code and was previously a blue phosphorescent code) and the Customer bar code RM4SCC printed in black ink by users of the mail system who wish to code and pre-sort their mail for access downstream of the public entry points.

The code is self clocking in that each and every bar or element is always present in some form and hence synchronisation can be maintained irrespective of letter speed or minor distortions in the code density.

Earlier codes used by Royal mail relied on the presence or absence of a phosphor dot spaced approximately ¼ inch apart. Due to wear on printer blocks and runout on later ink jet printed codes, it was sometimes difficult to identify exactly where each dot or bit was supposed to be. The fact that a clock element is always present in the 4 state symbology allows for synchronisation at each and every bar position thus preventing errors from occurring due to badly placed marks.

The code was created in response to the need for a high data density in order to implement automatic Optical Character Recognition (OCR) and the offline coding by human operators supporting the OCR system.

The encoding of address information for widespread automated sorting goes back to the early 1960s. Earlier prototype system were used in London some with physical coding and some directly sorted using London codes.

The postal code at the time comprised two parts, The ‘Outward’ part e.g. TQ5 for Brixham near Torquay and the ‘Inward’ part e.g. 8EU for the area around Brixham town Centre.

The Outward code is unique for each area whereas the Inward codes are used for localities within each area.

For example the following codes could exist sharing the same Inward code but with different Outward parts: TQ5 8EU (Brixham town centre); BN11 8EU (Worthing town centre).

The original coding system was applied to letter mail using two rows of phosphor dots in two rows. The lower row represented the Outward part of the postcode and the upper row represented the Inward part.

The Codes were read from right to left with the start dot on the right hand side being square and the rest 1/16” round. Each line of code comprised a start dot, twelve data dots (binary bits) and a parity bit. Even parity was employed and included the start dot. The data was encoded using ‘Code-Sort’ translators which allowed all the Outward and Inward alphanumeric information to be encoded in the 24 bits available on each letter.

Two requirements led to the need for a code which was significantly higher in data density than the 24 phosphor dots: 1) The addition of Delivery Point coding, 2) Manual support for the OCR system.

Delivery Point Coding

The Inward part of the UK postcode covers an average of about 180 addresses, some vastly more, and some significantly less. An addition to the Postcode called the delivery point was added as a two digit number and letter e.g. 1A – 9Z. In order for this code to be printed on the letters, more space was required.

Delivery point information was included to allow for local sorting at delivery level in local Delivery offices. At the time of writing, this is still the subject of experimental technology.

Where OCR technology was employed, there was a requirement to identify each letter uniquely so that delayed or off line coding could be carried out. ‘Coding’ in this sense is the entry of the written postcode information into the electronic system for translation into printed bar codes. Each letter which could not be read automatically would have a unique identifier printed and after the information was coded offline, the letter would be passed back through the system, identified and printed with the electronic postcode information.

After an extensive study of coding materials and systems for printing and reading the code, the 4 state code using ink jet printed phosphorescent ink was settled upon.

The upper part of a letter where the Inward used to be is now used for the unique identifier needed for the implementation of OCR systems and the lower row is used for full postcode information including delivery point. The data in both cases was encoded using an algorithm developed by ERA Technology in Leatherhead, England.

The codes are printed using ink jet printers and an ink developed by the Royal Mail Research and Development chemistry laboratories of Royal Mail in Swindon England. The original code reading technology was developed by Philips in Eindhoven Holland, with cameras developed by Wessex Technology in Portsmouth England.

The 4-state code was never patented by Royal Mail. Although the code is very robust, particularly when the data is encoded to allow for redundancy, the code was still liable to a percentage of error when printed on the variety of substrates present in the postal system. By allowing free access to the code by other Postal administrations and the development of the associated technology, it was felt that there would be general improvements in performance and more cost effective technology solutions in the future.