7 Facsimile adaptor
23.1463GPPRelease 17Technical realization of facsimile Group 3 non-transparentTS
7.1 Principles of the Facsimile Protocol Adaptation
The basic approach of the present document for facsimile group 3 is:
– to use the ITU-T Recommendation T.30 procedure at both ends of the connection between the FA and the associated facsimile terminal;
– to intervene within the FAs in order to avoid a T.30 timer timeout due to delays in the radio interface;
– to check the message speed; the R-FA selects the appropriate modem indicated by the DIS/DTC frames from the R-FAX. The T-FA switches the appropriate modem according to the TSI/DCS frames from the T-FAX after selecting the modem indicated by the DIS/DTC frames from the R-FA.
Following this strategy, an interchange model is defined concentrating on the facsimile relevant components. According to this model, three connection sections may be distinguished:
a) between the T-FA and the T-FAX;
b) between the FAs;
c) between the R-FA and the R-FAX.
The FAs shall cater for the correct establishment and control of these layers, including traffic channel synchronization and status information exchange, in particular, with respect to circuit 106 and circuit 109 (according to ITU-T Recommendation V.24). Once these circuits have been set to ON (traffic channel synchronization), they shall be kept in the ON condition during the entire facsimile phase of a connection.
7.2 FA functionality
Basically there are four problem areas:
– support of facsimile group 3 with a digital connection type;
– unpredictable delays on the radio interface due to actual RLP working conditions (ARQ);
– the need to change the transmission rate "locally" in the FAs both in the UE and in the MSC/IWF and to adapt it to the constant user rate of the radio interface;
– the inability to support some ITU-T Recommendation T.30 features.
To overcome these problems some particular functions within the FAs are necessary, such as:
– buffering of BCS frames and facsimile coded data prior to transfer;
– autonomous interventions such as BCS frame inhibiting within the FAs;
– provision of a FA protocol (IFP/UDPTL/UDP/IP/PPP) as an interchange protocol between the FAs.
The main features relevant to ITU-T Recommendation T.30 adaptation functions are detailed in the following.
The proper configuration is settled on both network sides by detecting the DIS/DTC frame just at the beginning of the phase B in ITU-T Recommendation T.30 protocol procedure.
The optional error correction mode, as defined in ITU-T Recommendation T.4‑Annex A and ITU-T Recommendation T.30‑Annex A, may be fully supported, provided some specific features are included in the FA procedure. ECM is optional for the FA.
In order to support ECM:
– additional BCS frames need to be detected;
– the handling of the message phase needs to be adjusted.
The overall framework as described in the present document applies, i.e. it shall be the same mode, ECM or non-ECM, at the transmitting side and the receiving side, because the FA stores FAX signals but does not convert the format of ECM and non-ECM.
The error correction mode is entered upon detection of the relevant bits in the DIS/DTC frame.
The working principle of the present document is based on the detection, control, deletion, and generation of key messages between the:
– the T-FAX and the T- FA;
– the T- FA and the R- FA;
– the R- FA and the R- FAX.
While in BCS phases, the following frames shall be detected:
– DIS/DTC, to monitor all operational parameters of the transmitting terminal;
– CSI, to monitor all operational parameters of the receiving terminal;
– DCS, to realize the actual operational parameters, e.g. message transmission speed accepted by the sender terminal and the relevant message transfer direction (see table 2/ITU-T Recommendation T.30);
– TSI, to indicate that the following FIF information is the identification of the transmitting terminal. It may be used to provide additional security to the facsimile procedures;
– TCF, to verify training and to give a first indication of the acceptability of the channel for this data rate;
– FTT, to reject the training signal and requesting a retrain;
– CTC/EOR, to fix the retransmission strategy by the facsimile transmitting terminal (error correction mode);
– CTR/ERR, to act as the confirmation message and so closing a BCS phase before a new message phase;
– EOM, to indicate the end of a complete page of facsimile information and to return to the beginning of the phase B;
– MPS, to indicate the end of a complete page of facsimile information and to return to the beginning of the phase C upon receipt of a confirmation;
– EOP, to indicate the end of a complete page of facsimile information and to further indicate that no further documents are forthcoming and to proceed to phase E, upon receipt of a confirmation;
– PRI-Q, to indicate the same as EOM/MPS/EOP command with the additional optional capability of requesting operator intervention, therefore PRI-Q is treated as EOM/MPS/EOP;
– DCN, to initiate the call release procedure.
While in BCS phases, the following frames shall additionally be detected. However, these following signals shall not be transmitted to the radio network:
– NSF, to indicate the specific user requirements, which are not covered by ITU-T T-Series Recommendations;
– NSC, to indicate the response to NSF, however this signal is not transmitted to the radio network;
– CFR and MCF, to trigger the message phase;
– PPR, as above, but after the fourth consecutive PPR request, the BCS phase continues with either CTC or EOR (error correction mode);
– RR, to ask for the status of the receiver;
– RTP, to indicate that a complete message has been received and that an additional message may follow after retransmission of training and CFR;
– RTN, to indicate that the previous message has not been satisfactorily received. Several receptions may be possible, provided training is retransmitted;
– PIP/PIN, to indicate that further transmissions are not possible without operator intervention;
– RNR, to indicate that the receiver is not ready to receive more facsimile data;
– ERR, to indicate the response to EOR;
– CRP, to require the retransmission of the previous received signal in errors.
Furthermore, all BCS command messages shall be monitored to eliminate repeated command messages at the FA towards the other FA and to initiate a repetition of those command messages, if necessary, at the FA towards the associated FAX. Additionally, all BCS response messages shall be monitored to be able to clear the former condition.
7.2.1 Packet elements between FAs
To cater for the appropriate facsimile transmission some protocol elements and the related procedures are defined. These protocol elements shall be exchanged between both FAs. They are based on E-T.38 and described as follows. They are structured as outlined in annex A:
IFP packet (see clause A.1 in annex A):
The IFP packet is used to relay ITU-T Recommendation T.30 BCS frames. It includes the possibility also to transmit parts of an entire BCS frame (segmentation).
1) T.30_INDICATOR:
– CNG packet;
– CED packet;
– TCF error detection packet: The TCF error detection packet is used to notify the opposite FA of the TCF error; – FA_busy_packet: The FA_busy_packet is used to notify the T-FA that the R-FA has received the image signal from T-FA;
– FA_non_busy_packet: The FA_non_busy_packet is used to notify the T-FA that the R-FA has received the MCF, RTP or PIP frame from R-FAX.
2) T.30_DATA:
– T.30_DATA conforms to T.38;
Packets defined in T.38 but not used in the present document:
– Training packet;
– V.21 Preamble Flags packet.
7.2.2 Interactions between and interventions within the FAs
Interactions between the FAs and between FA and TAF are necessary:
– where the content of the protocol packets, except NSF and NSC, shall be aligned with the capabilities of the supporting the UMTS PLMN (e.g. figure D.5);
– where the hardware flow control of V.24 shall be adopted for the flow control between the FA and the TAF (refer to 3GPP TS27.002);
‑ read the facsimile coded data from the buffer, for transmission across the radio interface using the appropriate FA protocol packets (see annex A);
– during the transmission of facsimile coded data where the document content is transcoded to save transmission capacity at the radio interface (see clause7.2.4.1);
‑ write the facsimile coded data received across the radio interface into a buffer at the R-FA;
– where the T-FA distinguishes the phase at the R-FA by using FA_busy_packet and FA_non_busy_packet (see clause7.2.1).
To perform the necessary interactions, the FAs at the UE side and at the MSC/IWF side shall monitor the BCS frames (commands/responses) and the facsimile coded data exchanged between the two facsimile terminals and perform the actions detailed below:
‑ receive the BCS frames from the associated facsimile terminal discarding the BCS frames repetitively received due to time‑out in the facsimile terminal;
‑ store the BCS frames completely received from the associated facsimile terminal;
– change information elements of the BCS frames indicating capabilities which may not be supported by the PLMN (e.g. figure D.5);
– change BCS frames received from the associate FAX into FA protocol packets, and transmit it to the opposite FA (see annex A and refer to ITU-T Recommendations T.38);
– change FA protocol packets received from the opposite FA into BCS frames, and transmit it to the associate FAX (see annex A and refer to ITU-T Recommendations T.38);
‑ transcode the normal facsimile coded data received from the T-FAX and store the data into a buffer in the T-FA (see clause 7.2.4.1);
Intervention with in the associated facsimile are necessary:
‑ control the half duplex connections between the FA and the associated facsimile terminal (refer to ITU-T Recommendation T.30);
‑ perform phasing/training with the associated facsimile terminal (refer to ITU-T Recommendation T.30).
‑ transmit the BCS frames to the associated facsimile terminal autonomously repeating the BCS frames towards the accepting facsimile terminal if necessary after time‑out or on request by the CRP (e.g. figure D.4);
‑ where BCS frames shall be repeated autonomously by the FA after a time-out when no response has been received from the associated facsimile terminal (e.g. figure D.18);
‑ transmit/receive the ITU-T Recommendation T.30 training check frames (TCF) to/from the associated facsimile terminal (refer to ITU-T Recommendation T.30);
‑ read and reverse transcoding of the buffered normal facsimile coded data and transmitting them to the associated Facsimile terminal (see clause 7.2.4.1);
– select a appropriate dummy signal (i.e. PPR, FTT, RTC, RTP) : For example, the dummy signal shall be sent to the T-FAX in order to avoid T.30 timer time-out at the turn of phases (e.g. figure D.7);
– adjust the time interval between the transmitted BCS frames (e.g. figure D.5);
– adjust the time interval between receiving a BCS frame and transmitting a response to it (e.g. figure D.14);
The support of guard tones by the FA in the MSC/IWF is an implementation option.
7.2.3 BCS phase
7.2.3.1 BCS command/response procedures
The ITU-T Recommendation T.30 procedures are segmented in sections of associated BCS commands and responses. Any command sent by a facsimile terminal shall be answered by an appropriate BCS response (refer to Appendix III in ITU-T Recommendation T.30). To guarantee that the transmission and reception of responses have the correct timing, the response time after reception of a previous command shall be 75m seconds, when the opposite FA is not busy. When the opposite FA is busy the corresponding timer shall be extended to 2,4 seconds.
According to the command/response relationship, the FA receiving a BCS command from its associated facsimile terminal is called the commanding FA. Similarly, the FA receiving a BCS response is called the responding FA.
The FA shall not transmit the preamble from the associated FAX to the radio network. The opposite FA shall record the BCS frame received from the other FA and transmit the preamble and procedure signal to the associated facsimile within the proper timing. The procedures are as follows:
1) BCS command procedures; A BCS command which has been received correctly from the associated facsimile terminal, is stored within the commanding FA. Any forwarding of BCS commands repetitively received from the facsimile terminal is further on inhibited. If an error occurs during the reception of the BCS command from the associated facsimile terminal, the commanding FA shall re-transmit the previously transmitted procedure signal to the associated FAX and request the proper procedure signal. The commanding FA transmits the received BCS frame using "IFP packets" to the radio network. The commanding FA shall not transmit the incorrect procedure signal received from the associated FAX to the radio network.
Contiguously received parts of a BCS command packet received from the radio interface are stored in the responding FA. The stored complete BCS command is used for autonomous transmissions towards the associated facsimile terminal which may start at the earliest possible time according to ITU-T Recommendation T.30.
The responding FA receiving "IFP packets" checks their correct sequence and starts reassembling the BCS command and transmits it when it has been completely received and no sequence error has been detected. All IFP packets except IFP packets for the next procedure are ignored. The sequence has to be reconstituted at the responding FA before sending it to the associated facsimile terminal.
2) BCS response procedures; A BCS response which has been received correctly from the associated facsimile terminal, is stored within the responding FA.
If an error occurs during the reception of the BCS response from the associated facsimile terminal, the responding FA shall re-transmit the previously transmitted procedure signal to the associated FAX and request the proper procedure signal. The responding FA shall not transmit the incorrect procedure signal received from the associated facsimile to the radio network.
Contiguously received parts of a BCS response received from the radio interface are stored in the commanding FA. The stored complete BCS response is used for autonomous answering towards the associated facsimile terminal which may start at the earliest possible time according to ITU-T Recommendation T.30.
After transmission of a response towards the FA/fax terminals by the FA function, after which the FA function waits for message data (CFR, CTR, MCF after MPS, PPS_MPS, PPS_NULL, ERR after EOR_MPS, EOR_NULL) a repeated BCS command shall be locally responded, without transferring a preamble element towards the radio link.
The FA shall take care of the control of the local modem. The condition is derived from the reception of certain FA protocol elements. Additionally, there shall be a control of the half duplex transmission path towards the associated facsimile terminal. For that purpose, the receiver signal is monitored and the transmitter is activated, only if no receive signal is active.
7.2.3.2 Compatibility checking
Some features shall not be supported in the UMTS PLMN environment. The FA is in charge of dealing with such compatibility checking which is carried out by monitoring certain BCS frames (DIS/DTC).
– Group 1 and group 2 equipments shall not be supported by the Teleservice as described in the present document.
– Error limiting mode shall not be supported.
– Only standard 300 bit/s Binary Coded Signalling shall be supported. For this purpose, the FA shall ignore the 2400 bit/s capability within phase B of the ITU-T Recommendation T.30 procedure by looking for DIS frames from ITU-T Recommendation V.21 modem only.
– The following facsimile message speeds shall be supported: 2400, 4800, 7200, 9600, 12000 and 14400bit/s. For this purpose the FAs are responsible to carry out appropriate actions, e.g. to set the applicable fields in the DIS frame accordingly.
– It is not possible to support non‑standard facilities since some of these contain proprietary methods of changing the modem speed that are un-interpretable to the IWF and hence impossible to track.
7.2.3.3 Training Check
The training check sequence (TCF) as described in ITU-T Recommendation T.30 is exchanged only locally between the FA and the associated facsimile terminal. The training check sequence sent by the FA shall be transmitted at the earliest possible time, and shall have a duration that complies with the minimum duration requirement of ITU-T Recommendation T.30. Because CFR is not transmitted end to end the T- FA generates CFR or FTT, and sends it to the T-FAX.
The T-FA shall send CFR to the T-FAX, 2.4s after it received TCF from the T-FAX.
However in the following cases, the T-FA shall send FTT to the T-FAX instead of CFR (see figure D.7):
– when it receives the TCF error detection packet from the R-FA;
– when it receives the FTT packet from the R-FA;
– when it recognizes the R-FA busy status by the FA_busy_packet.
(The T-FA shall transmit FTT to avoid the phase difference between the transmitting and the receiving side when the phase is returned to phase B by EOM.)
This is done by using the FA_busy_packet and the FA_non_busy_packet to monitor the state of the R-FA. When the T-FA receives the FA_busy_packet it can deduce that the training has been successful on the receiving side.
As a consequence of this local procedure, the T_FA shall check the received TCF to see whether the quality requirements are satisfied. Depending on the result of that check, the T-FA recognizing a corrupted line shall send FTT instead of CFR. If the T-FA receives corrupted TCF from the T-FAX, it shall transmit the TCF_error_detection_packet to the R-FA. On the other hand, if the R-FA receives the TCF_error_detection_packet from the radio network, it shall transmit a corrupted TCF to the R-FAX (see figure D.7).
The message transfer phase in the R-FA shall be entered upon reception of CFR. The modem training at transmission speed shall start after the R-FA receives the facsimile coded data if it receives the data within 3 s after the receipt of CFR from the R-FAX (see figures D.5 and D.9).
The R-FA shall transmit the TSI, DCS to the R-FAX, if it does not receive the facsimile coded data within 3,0 s after the receipt of the CFR from the R-FAX (see figures D.5 and D.9).
7.2.3.4 Control of transmission rate
The controlling entity of the FA shall recognize a change of the transmission rate, and in conjunction with this a change of the modem function has to commence. The transmission rate shall be changed only locally, i.e. between the facsimile terminal and the FA at both the MSC/IWF and the UE ends.
The actual message speed and the modem function are derived from the content of the related BCS frames (DIS/DTC, DCS, CTC).
There are two methods to decide the transmission rate as follows:
– Use DIS DCS signal between FAXs (see figures D.4 and D.5);
– Trigger Fallback by the TCF error detection.
Fallback is triggered both on the transmitting and receiving sides.
In the case that fallback occurs on the transmitting side, the R-FA shall be notified of the fact by the TCF error detection packet (see figure D.9).
In the case that fallback occurs on the receiving side, the T-FA shall be notified of the fact by the FTT packet from the R-FA (see figure D.7).
7.2.4 Message phase
During the message phase (phase C of ITU-T Recommendation T.30) the EOL character shall be detected, (see clause 4.1.2 of ITU-T Recommendation T.4). The EOL character is a unique code word that may never be found within a valid line of facsimile coded data, and is used:
– to identify the start of the message phase;
– to control the transcoding procedure;
– to mark the end of the message phase (6 consecutive EOLs).
If the R-FA has stored more than 0,28 kbyte of facsimile coded data and detected two or more EOLs in the data, then the R-FA shall send the FA_busy_packet to the T-FA. After the R-FA receives MCF from the R-FAX, the R-FA shall send the FA_non_busy_packet to the T-FA. The R-FA shall notify the T-FA of the state of the reception side by sending the FA_busy_packet or the FA_non_busy_packet to avoid that the phase at the T-FA proceeds the next phase (as a result of EOP, MPS, EOM, CTC, PPS-Q, EOR-Q, RR) while the R-FA is still busy in the current phase.
7.2.4.1 Message Transcoding
To save transmission capacity at the radio interface the content of the document shall be transcoded. This applies only when using the normal facsimile data transfer. This does not apply when using the ITU-T Recommendation T.30 error correction mode.
The facsimile coded data received by the T-FA from the T-FAX is transcoded and transmitted to the corresponding R‑FA across the radio interface. Then it is transcoded and transmitted to the R-FAX by R-FA.
The transcoding is based on the minimum line length capability of the T.30 protocol for the normal facsimile data transfer. According to this the transmitting facsimile terminal has to fill up each coded scan line with FILL information to conform to this requirement (refer to ITU-T Recommendation T.4).
To do so, the FA associated with the T=FAX shall force this to use (at least) the standard value of 20 ms by replacing the applicable parameter value of the exchanged DIS/DTC messages. All FILL information of the facsimile coded data received from the FAX shall be deleted prior to forwarding the data across the radio interface. The FA associated with the R-FAX shall recognize and store the originally requested minimum line length to be able to regenerate to correct line length.
It should be noted that ITU-T Recommendation T.4 document coding may be 1‑dimensional or 2‑dimensional and, in addition, uncompressed. The FAs shall take care of this when transcoding the document content.
7.2.4.2 Generation of the normal data packet
This is specified in ITU-T Recommendations T.38 and X.691. X.691 provides details on octet alignment.
7.2.4.3 Generation of the error correction data packet
This is specified in ITU-T Recommendations T.38 and X.691. X.691 provides details on octet alignment.
7.2.4.4 Normal facsimile data
The T-FA starts the message phase by transmitting CFR or MCF to the T-FAX. The R-FA enters the message phase by receiving CFR or MCF from the R-FAX.
– If the R- FA does not receive the facsimile coded data from the radio network within 3s of receiving CFR from the R-FAX, it shall re-transmit a preamble of 1.0s and the previously transmitted TSI, DCS to the R-FAX after changing the modem to V.21.
– If the R- FA does receive the facsimile coded data from radio network within 3s of receiving CFR from the R‑FAX, it shall transmit the training to the R-FAX after changing the modem to V.17, V.27ter or V. 29.
– If the R-FA does not receive the facsimile coded data from the radio network within 3s of receiving MCF from the R-FAX, it shall re-transmit a preamble of 1.0s and the previously transmitted MPS to the R-FAX after changing the modem to V.21.
– If the R-FA does receive the facsimile coded data from the radio network within 3s of receiving MCF from the R-FAX, it shall transmit the training to the R-FAX after changing the modem to V.17, V.27ter or V. 29.
Following the training sequence, the R-FA shall transmit FILL towards the R-FAX (see clause 4.1.2 of ITU-T Recommendation T.4), disregarding all information received from the radio interface, until an EOL character is detected, which shall mark the beginning of the image data (see figure 1 of ITU-T Recommendation T.4).
If due to a preceding error the message phase may not be entered, the training sequence shall be aborted when a new IFP packet is received by the T-FA.
7.2.4.5 Error correction facsimile data
As the facsimile coded data between the FA and the facsimile terminal are structured in HDLC frames, the handling of this procedure segment shall exploit such formatting. The content of such an HDLC frame is called a block (refer to ITU-T Recommendation T.4 Annex A).
Each such block is included in the information field of an Error_Correction_Data_Packet of the FA protocol, which is processed for transmission across the radio interface as outlined in clause 7.2.4.3.
The message phase at both the PLMN ends is triggered by the transit of a confirmation frame (CFR, MCF, PPR, CTR or ERR) sent by the receiving terminal and marking the end of the BCS phase.
The PPR that the T-FA sends to the T-FAX has two purposes:
– PPR is used to request re-transmission of the facsimile coded data. When the T-FA receives corrupted facsimile coded data, it requests correct facsimile coded data to the T-FAX (refer to ITU-T Recommendation T.30).
– In addition to the T.30 purpose, PPR is used to avoid the timer timeout of the T.30 protocol. If the R-FA is busy, the T-FA shall transmit PPR to the T-FAX (see figure D.20).
If four consecutive PPR are counted within the same "partial page", the BCS phase continues.
The T-FA shall enter the message phase as described in ITU-T Recommendation T.30.
– If the R-FA does not receive the facsimile coded data from the radio network within 3s of receiving CFR from the R-FAX, it shall re-transmit a preamble of 1.0s and the previously transmitted TSI, DCS to the R-FAX after changing the modem to V.21.
– If the R-FA does receive the facsimile coded data from radio network within 3s of receiving MCF from the R-FAX, it shall transmit the training to the R-FAX after changing the modem to V.17, V.27ter or V.29.
– If the R-FA does not receive the facsimile coded data from the radio network within 3s of receiving MCF from the R-FAX, it shall re-transmit a preamble of 1.0s and the previously transmitted MPS to the R-FAX after changing the modem to V.21.
– If the R-FA does receive the facsimile coded data from the radio network within 3s of receiving MCF from the R-FAX, it shall transmit the training to the R-FAX after changing the modem to V.17, V.27ter or V.29.
Following the training sequence, HDLC flags shall be stuffed into the signal sent to the facsimile terminal until a FCD frame is detected, which shall mark the beginning of the image data.
If due to a preceding error the message phase may not be entered, the training sequence shall be aborted when a new IFP packet is received by the T-FA.
7.2.4.6 Controlling of facsimile coded data
The procedure for controlling the FAX data within the FA is described in the following two clauses.
It is recommended that the buffer size inside FA is more than 256 kbytes.
7.2.4.6.1 Transmitting Fax Adapter
In the T-FA the facsimile coded data being received from the T-FAX are transcoded stripping of FILL information and written into the buffer.
When data is read out from the buffer, an FA protocol element shall be generated and processed as described in Annex A or ITU-T Recommendation T.38. It is transferred to the R-FA using one of the standard FA protocol element. For that purpose the data shall be segmented into blocks.
Due to the ARQ techniques of the RLP, the throughput across the radio interface may be less than the message speed between the T-FAX and the T-FA, i.e. the content of the buffer may increase.
If the throughput at the radio interface is greater than the message speed between the T-FAX terminal and the T-FA (e.g. when the end to end speed is lower than 14400 bit/s), the buffer may be empty most of the time.
7.2.4.6.2 Receiving Fax Adapter
In the R-FA, FILL information is transmitted to the R-FAX at the beginning of each page, if necessary, to bridge the gap between the training sequence and the reception of the image data. In case of normal fax data, the duration of the FILL 0’s may be increased by only 4,5 s or less.
The reverse transcoding consists of the insertion of FILL information before the facsimile coded data is forwarded to the facsimile terminal to comply with the recognized minimum line length as defined in ITU-T Recommendation T.4.
At the beginning of each page, the facsimile coded data to be sent to the facsimile terminal is not read out from the buffer until either at least 2 instances of EOL have been received and the buffer size is at least 0,28 kbytes or, an RTC has been received, which does not depend on the end-to-end data transfer rate.
7.2.4.6.2.1 Fill insertion method for NON- ECM
The R-FA controls FILL insertion according to the accumulation value of image signals (see figure D.11).
The condition for starting FILL insertion:
– when the pix memory accumulation value is 0,28 kbytes or less.
The condition for stopping FILL insertion:
– when the pix memory value is 0,9 kbytes or more;
– when the pix memory value is not over 0,9 kbytes and the interval between EOLs (one line period) reaches 4,5 s.
When the interval between EOL reaches 4,5 s, the R-FA shall stop FILL insertion, and shall transmit EOL and shall start to transmit the next facsimile coded line. If however the pix memory value is below "the forced RTC transmission value"(=7byte or less), the R-FA shall send RTC to the R-FAX to force end of Phase C. Then it shall wait for a post-message (EOP, EOM, MPS) to be received from the air interface.
7.2.4.6.2.2 Flag insertion method for ECM
The R-FA controls Flag insertion according to the accumulation value of the image signals(see figure D.13).
The condition for starting Flag insertion:
– when pix memory accumulation value is 0,28 kbytes or less.
The condition for stopping Flag insertion:
– when the pix memory value is 0,9 kbytes or more.
7.2.5 Disconnect procedure
The T-FA sends the MCF frame to the T-FAX after receiving the FA_non_busy_packet from the R-FA. The T- FAX sends the DCN frame (see ITU-T Recommendation T.30) to the T-FA after receiving the MCF frame. The T-FA, upon detection of the DCN frame sent by the T-FAX to indicate the end of the facsimile transmission, shall initiate the disconnect procedure. The T-FA transmits the DCN packet to the R-FA after receiving the DCN frame from the T‑FAX, while the R-FA transmits the DCN frame to the R-FAX after receiving the DCN packet.
7.2.6 Timeouts
The T.30 timer between the T-FAX and the T-FA, and the T.30 timer between the R-FAX and the R-FA are independent of each other. Therefore, the overall fax adaptation function is in principle bound to the timing constraints associated with the associated FAX ITU-T Recommendation T.30 procedure. This means that, no matter of the reference configuration used at the mobile station, the progress of the call shall be mainly subject to the ITU-T Recommendation T.30 typical timing protections, settled externally.
However, due to the specific conditions caused by the UMTS PLMN system, there is the need for a special support with respect to BCS command repetitions as explained above. For that purpose, the FAs shall provide means for local time-out. The timer shall be started and stopped as described in annex D (see table D.2).