FIELD OF THE INVENTION This invention relates to processes for encoding/decoding messages in a restricted communication network, particularly but not solely in a seven bit paging network where transmission of certain characters is prohibited.

BACKGROUND OF THE NTVENTIQN

A signalling protocol developed by the British Post Office Code Advisory Group (POCSAG) is widely used in paging systems. It provides a set of 128 seven bit characters specified in CCITT ASCII alphabet No. 5 and thereby enables standardised transmission of alphanumeric messages. Other well known protocols are the Golay Sequential Code

(GSC) and BRMES as defined by the European Telecommunications Standards Institute

(ETSI). However, paging networks generally prevent a user from sending all of the possible seven bit characters. Telecom New Zealand and Vodapage (UK) systems arbitrarily prohibit messages containing any of ASCII characters 0-31 or 127. This arises for example, through data entry programs which require characters 8, 10 and 13

(BACKSPACE, LINEFEED, CARRIAGE RETURN) to control user editing, and use of characters 17 and 19 (XON, XOFF) for flow control.

Preventing transmission of complete seven bit character sets as specified in the standards has in turn prevented implementation of encryption and compression services.

Information transmitted on a paging network therefore has not been protected from viewing by unauthorised third parties or staff at the network centres. It has also prevented remote control of paging receivers so that all receivers must be configured by physically interfacing with a programming device. For example, pay-to-use services cannot readily be implemented for selected groups of users, and lost or stolen receivers cannot be remotely deactivated.

SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide for use of full seven bit character sets in existing restricted communication networks, such as paging networks, or at least to provide the public with a useful choice.

Accordingly, in one aspect the invention may broadly be said to consist in a process for encoding a message to be transmitted over a paging network in which transmission of certain characters is prohibited, comprising: encrypting the message according to a

predeterrnined algorithm, and transforming the encrypted message according to a predeteimined algorithm so that prohibited characters become permitted characters.

Preferably the prohibited characters are transformed into a fixed permitted character which indicates that a transformation has taken place and a variable permitted character which reflects the original prohibited character.

Preferably characters are added after transformation to indicate length of a transmitted message and/or to indicate a message sender.

In another aspect the invention may also broadly be said to consist in a process for decoding a message received over a paging network in which transmission of certain characters is prohibited, comprising: transfoπning the message according to a predetermined algorithm so that certain permitted characters revert to certain prohibiteH characters, and decrypting the transformed message according to a predeterrnined algorithm.

Preferably characters are added before or after encryption to assist checking of a received message and/or to add a random component to repeated messages for hindering unauthorised decryption.

In another aspect the invention may broadly be said to consist in a paging receiver device which is programmed to carry out either or both of these two processes. BRIEF DESCRIPTION OF THE DRAWINGS A preferred embodiment will be described as an example of the invention with respect to the drawings of which:

Figure 1 is a flowchart of an encoding process,

Figure 2 is a flowchart for a corresponding decoding process, and

Figure 3 is a schematic diagram indicating a communication network and a paging receiver on which the invention might be implemented. DESCRIPTION OF A PREFERRED EMBODIMENT

Referring to these drawings it will be appreciated that processes according to the invention can be implemented in a wide range of restricted paging systems, using known equipment which need not be described in detail. Reference is made to WO 94/21051 for example.

Encoding and decoding routines may be installed on personal computers using MS-DOS or Windows, for example, to interact with a paging network by modem and telephone, or may be used on otherwise standard portable paging devices.

Permission or cooperation from the network operators is not required although each pair of sender/receiver devices must be aware of the prohibited characters in their particular signalling protocol and must be agreed on details of the encoding/decoding

processes to be used. They will need to run common encryption and transformatio algorithms for example.

A sender may optionally tag each message with an identifier which may be presen in both encrypted and unencrypted forms as a check on message origin and to facilitat decoding by the receiver. A simple trial and error decryption and identification ma otherwise be carried out. These and other features of the invention will be explained i more detail below.

Figure 1 indicates the processing steps which should be taken when encoding message according to the invention, and also gives an example using a message strin "ABCDEFGH". The string may incorporate sender identification at this stage.

In step 10 a sender identifies a recipient and types the message into a computer o dedicated paging device. Text may need to be compressed to seven bit bytes as necessar for existing paging networks. There will also be a maximum message length on eac network and the input message must normally be limited to substantially less to allow fo an increase during encoding.

A control byte may be added to indicate whether compression has occurred an supply infoπnation relating to the compression process. For example, bit 0 of the contro byte may be true for eight bit data, bits 1 to 3 of may contain remainder bits for Huffma compression, and bits 4 to 6 may indicate the kind of compression. In step 11 an agreed number of random characters may be added to the messag string so that a repeated message will not produce the same encoded output. In thi example two characters "zy" have been added as a prefix, although any characters in th complete set of the particular protocol may be used in a string of a chosen length an position on the message. In step 12 the message, extended as a result of step 11, is encrypted according to a gorithm known to the recipient and using a selected encryption key if appropriate. range of algorithms might be used, such as a public key system or perhaps a simpl routine based on a Feedback Shift Register algorithm.

In step 13 an agreed validation component may be added to the message so that th receiver may automatically check whether any interference has occurred and whether th decoding process has performed properly. Several bytes may be used to enable a appropriate cyclic redundancy check. In the example the component is an encoded uppe value remaining in a seed register. Alternatively a frailing null may be encoded.

In step 14 any prohibited characters in an encrypted message are transforme according to a further algorithm which is known to the receiver. At present the preferre algorithm requires a predetermined escape character selected from the range of permitte

characters, such as ASCII character 126 ("~") as shown in the example. The escape character should be in the range 0-31 or 96-127 to avoid those in common use. Other algorithms may also be envisaged.

The preferred algorithm takes the ASCII value of the prohibited character and adds or subtracts 32 depending on whether the value is less than or greater than/equal to 64.

This parameter has been chosen to rehably convert all currently prohibited characters into permitted characters under the existing network systems. The resulting permitted character is replaced in the message string proceeded by the escape character as an indication to the receiver that a transformation has taken place at that point. In step 15 the length of the encrypted transformed message should be determined and added to the string. In the example the length is 13 characters and this number corresponds literally to a prohibited character. The character is therefore transformed using the algorithm of step 14 into character 45 ("-") which is permitted. It is preferably added as prefix to the message string as shown. In step 16 an optional identifier is added to the message to facilitate decoding. The identifier characters would normally be chosen so that transformation is not required. As above, this string may be placed appropriately within the message, but preferably as a suffix.

If the overall length of the message then exceeds a predetermined limit, then the sender is warned and will be required to reduce the message content before transmission can take place in step 17.

Figure 2 indicates the processing stages which should be taken when decoding an alphanumeric message according to the invention. Tone only messages will sometimes be received but these are tagged by the paging network. In step 20 the message is copied first into a holding buffer and then repeatedly into a working buffer from which the message may be destructively decrypted by multiple keys if necessary.

In step 21 the length character added in step 15 of Figure 1 is evaluated and deleted from the message in the working buffer. In step 22 the actual length of the message is compared with the length character.

If the actual length is shorter than expected, the message is assumed to be unencoded and is displayed in step 32.

If the actual length is greater then a string of the expected length is retained in the working buffer and the remainder is stored elsewhere as an identifier variable.

In step 23 the message undergoes a transformation which inverts that applied in step 14 of Figure 1. Each occurrence of the escape character is deleted in the working buffer and the character immediately following is assumed to represent a prohibited character.

The preferred algorithm takes the ASCII value of the character as it appears in the message, and again, adds or subtracts 32 depending on whether the value is less than or greater than/equal to 64. The resulting prohibited character is substituted in the message string.

In step 24 the identifier variable is compared with items in a list of possible identifiers of senders who could have transmitted the message. In step 25 a matching identifier causes recall of a specification decryption key by which the sender is expected to have encrypted the message.

In step 26 no matching identifier has been found and the receiver begins a check through all listed keys for a decryption by trial and error.

In step 27 decryption takes place according to an algorithm and key expected to have been agreed between the sender and receiver.

In step 28 the decoding process checks that an agreed validation character has been received, such as if the upper byte in a decode seed register is equal to the decoded value of the character received.

In step 29 the decryption has been deemed invalid due to an unsatisfactory validation character. The process may return to step 26 for decryption according to any further untried keys or may proceed to display an undecoded message in step 32 if no further keys are available.

In step 30 the decryption has been validated and any random characters added in step 11 of Figure 1 are deleted from the message. In step 31 the finished decoded message is displayed on the receiver device.

Figure 3 indicates a communication network and receiver in which the invention may be implemented, by way of example. A network centre runs computer 30 and a radio transmitter 31 to provide a paging service or other personal communication service. An organisation which wishes to use the service may send messages to the network centre from the organisation's computer 32 and modem 33.

Individuals wishing to receive messages from the organisation will carry a receiver 35 which may take various forms. The receiver will typically incorporate an aerial 36, decoder unit 37 and processor/keypad/display unit 38. The decoder may incorporate a tunable RF stage 39, POCSAG decoder 40 and buffer or working memory 41.

Some of the decoder unit components may be provided in a replaceable module or on a magnetic card which is fitted to the unit 38. The various forms which a paging receiver may take will be well known to a skilled person.

Encryption according to the process of Figure 1 takes place in the organisation computer 32 before messages are sent to the network centre computer 30. Decryption of the received message takes place in the processor unit 38. The holding and working buffers mentioned in relation to step 20 of Figure 2 may be found in memory 41 for example.

The receiver 35 may also be capable of a return communication, such as an acknowledgement to the network centre 30 but in general the communication network will permit one way transmissions to the receiver only. The invention will be useful where tfie transmission of certain characters is prohibited by the network as mentioned above.