In this topic, several questions related to basic fax operation, fax calls on the PSTN, fax over IP (FoIP), and various modes of operation are listed. The questions are organized in the order of fax, fax over PSTN, fax over IP, and related connected points on bandwidth, messages, timing, and various modes of operation.
1. What is fax?
Fax is an acronym for a “facsimile” (facsimile means “a copy”) or “telefacsimile” system. A fax machine scans and sends written or graphical material to the destination fax machine for producing hard copy. A fax makes use of a standard telephone line (or VoIP operation) to transmit electrical signals to a remote fax machine.
2. What are the main functional components of a fax machine?
A standard facsimile terminal device consists of a paper input device (scanner); a paper output device (printer); accessories like a telephone keypad, telephone line control, display, headset, controller, and fax modem; and a limited amount of processing power.
3. How does a fax machine send pages?
The fax machine electrically scans and breaks up the document page into picture elements or pixels. These pixels are compressed, coded for error protection, passed through digital modulation, and delivered as a voice
band analog signal. At the destination, an analog signal is sampled, demodulated, decoded, and given to the printer. A fax makes use of a voice band analog signal. The intermediate transmission is not known to the fax machine. It can be PSTN or VoIP.
4. What are the main signal processing operations inside the fax machine? In the send direction, picture elements are compressed and coded with different schemes to protect from errors and bits are modulated with digital modulations. Several messages and acknowledgments based on the fax operation are sent as low-bit-rate frequency-shift keying (FSK) modulations. Tones, and caller ID are also used for call progress. The front-end hardware inside the fax machines delivers an analog signal. In the receive direction, a front end gives digital samples. These samples are demodulated for low-speed (messages and indications) and high-speed data. The demodulated bits are used for error correction and decompression before giving them to printer.
5. What are the different fax standards?
Fax devices are mainly classified as Group-1, Group-2, Group-3 (G3), and Group-4 (G4) depending on the image formatting and handling methods inside the fax machine. Group-1 and Group-2 are for analog fax and not relevant at present. G3 is a digital fax on analog lines, and G4 is a digital fax over digital telephones lines.
6. What is the role of the T.30 recommendation, and how does it relate with coding?
T.30 is the main controller of the fax call. The characteristics and the operation of G3 fax devices adhere to the requirements of the T.30 recommendation for a fax control signaling handshake. It is applicable to both the PSTN and fax over IP. Both fax machines will be aware of T.30 signaling. The T.30 fax control signaling messages are coded using a V.21 modem at
300 bps.
7. What are the different coding standards used in G3 and G4 fax machines?
A coding operation is similar to compression and error concealing of fax page pixel bits. A group-3 fax machine follows both ITU-T-T.4 and ITU-T T.6 recommendations for image compression and coding. It uses Modified Huffman (MH), modified read (MR) and modified MR (MMR) coding schemes. A group- 3 color fax machine uses Joint bi- l evel expert group (JBIG) and Joint Photographic experts group (JPEG) color-coding schemes. Group-4 fax follows the ITU-T T.6 standard for image compression and coding. It uses the MMR, Trellis, JPEG, and JBIG coding schemes.
8. What are the main differences between G3 and SG3?
The G3 fax machine supports the maximum data rates of V.17 14.4 kbps. The super-G3 (SG3) fax machines support maximum data rates of V.34 up to 33.6 kbps. All SG3 fax machines support error correction mode (ECM) mode by default. The G3 fax machine answers with an answer (ANS) tone,
whereas the SG3 fax machine answers with an ANS with amplitude modulation and phone reversal (/ANSam) tone during the call setup phase.
9. How many maximum lines can a standard A4 size paper contain?
The G3 fax machine scanner gives 1145 lines of information per page in the vertical direction (while using the lowest vertical resolution of 3.85 lines per mm) and 1728 bits of information per line in the horizontal direction on a standard A4 size piece of paper. It produces (1728)(1145) bits of approximately 2 million bits without any compression. Several compression methods are available. In compression, about 10 times compression is achieved that varies with page and compression scheme.
10. What are the basic principles of fax compression?
Considering black-and-white pixels, a fax page will have several consecutive pixels of white and black. Instead of sending every pixel as one bit, all the consecutive bits (grouped) are sent as one code word. This property is also exploited across multiple lines for two-dimensional compression. In the case of gray-level images, bit planes are derived and compressed to get better compression. Compression methods such as JBIG and JPEG are used with multilevel and multicolor fax compression.
11. What are the typical compression ratios for MH, MR, and MMR, JBIG and JPEG schemes in fax?
Fax compression widely varies based on the image data in a page. Typically, Modified Huffman 5:1; Modified read 7.5:1; and Modified modified read 10:1, JBIG and JPEG from 10:1 to 20:1. JBIG best suits black and white gray scale images and JPEG give more Compression for color images.
12. What is end of line (EOL)?
Each coded line ends with end of line (EOL) in MH and MR coding schemes. The EOL is a unique code word of 11 zeros followed by bit “1″ [format: 000000000001]. The EOL plus one tag bit is used as a synchronization code in MR coding. The end of the document page is indicated by sending the six consecutive EOLs in an MH coding scheme and the six consecutive EOLs plus tag bit in the MR coding scheme.
13. What is minimum scan line time (MSLT)?
MSLT defines how much time the receiving machine requires to print a single scan line. The minimum scan line time (MSLT) is set by the receiving fax machine and transmitted to the sending machine during the initial messages handshaking.
14. What is the minimum transmission time per total coded scan line?
The minimum transmission time of a total coded scan line is defined as the total time taken for transmission of the total coded scan line. The minimum transmission time of a total coded scan line is about 20 ms in G3 fax machines. The transmission time varies with image data and resolution.
15. Why are “fill” bits used in a coded line?
A pause may be placed in the message flow by inserting fill bits. A fill can be placed between a line of data and an EOL but never within a line of data. A fill is inserted to ensure a minimum transmission time of data. Format: A variable-length string of zeros.
16. What are the different fax modems used in a G3 facsimile device?
A modem used in a “fax modem” is modulation and demodulation. A fax modem is not a dial-up modem in fax processing. The G3 facsimile device uses V.27ter, V.29, and V.17 modems, and the SG3 fax uses a V.34 modem for image transmission. A V.21 modem is a default fax modem in G3 and SG3 facsimile devices, and it is used for fax handshaking and negotiation.
17. What is V.21?
V.21 is an ITU standard describing a low-speed data transmission rate of 300 bps with FSK modulation for exchange of T.30 fax control information and messages and V.8 signaling messages. Fax image data are not sent on V.21. The messages in V.21 are more robust and can work even if V.21 FSK signal is compressed like voice.
18. What is V.21 (L) and V.21 (H)?
Modems that use full-duplex communication (for example V.34 modem) use two- channel transmission. In V.21, channel- 1 is called the V.21 (L) low-band mode that uses 980 and 1180 Hz for mark and space frequencies, and channel-2 is called the V.21 (H) high-band mode that uses 1650 and 1850 Hz. In full-duplex mode, the originating V.21 modem transmits in low band and receives in high band, and the receiving V.21 modem transmits in high band and receives in low band. Group-3 fax machines use channel-2 in both directions.
19. What are V.27ter, V.29, and V.17?
V.27ter is an ITU recommendation for high-speed image data transmission that is used for medium transfer rates of 2400 and 4800 bps. It makes use of phase – shift keying (PSK). V.29 is an ITU recommendation used for medium transfer rates of 7200 and 9600 bps that makes use of Quadrature amplitude modulation (QAM). V.17 is an ITU standard that is used for high transfer rates of high-speed fax page data at 7200, 9600, 12,000, and 14,400 bps. It makes use of Trellis coding and QAM.
20. What are the different rates in an SG3 V.34 fax?
The advanced supergroup-3 fax supports G3 modems (V.27ter, V.29, and V.17) and the V.34 data modem for fax transmission. The V.34 modem operates in both half-duplex and full-duplex mode to support 14 primary data channel rates from 2400 to 33,600bps in steps of 2400bps and two control channel rates of 1200 and 2400 bps. A V.34-based fax session starts with 2100 Hz amplitude modulated and phase reversal answering tone. Binary procedural data are exchanged at 1200 or 2400bps using V.34 in full – duplex mode.
21. What is the role of the V.8 recommendation, and how does it relate with coding?
The V.8 protocol defines procedures for starting sessions of data transmission and their signaling format. V.8 signaling determines the best mode of operation before the initiation of a modem handshake. V.8 signaling messages are coded using the V.21 modem at a 300-bps rate.
22. What is G 3 C fax?
The G3 operation over digital networks like integrated services digital network (ISDN) is described in Annex- C of the T.30 recommendation. This option is known as Group-3C. G3C is mainly designed to be used on ISDN with procedures and signals based on T.30, T.4/T.6, and ECM mode.
23. What is G 4 fax?
Group-4 is mainly designed for an ISDN fax at 64 kbps based on ITU-T-T.6. G4 fax machines are digital and can interface directly with an ISDN line. Group-4 has three sets of classes (class-1, class-2, and class-3) based on pixel (pel) transmission densities. G4 can communicate with G3 with an optional dual-mode G4 class-1 fax that could use a standard G3C or G3 fax modem.
24. What is error correction mode (ECM)?
ECM is used with fax image data. ECM is an optional transmission mode built into fax machines or fax modems. ECM automatically detects and corrects errors by retransmission of the error data. It helps fax transmission errors caused by telephone line noise or any impairment. Fax machines have the capability to enable or disable this function.
25. Why is ECM mode the most common with SG3 fax machines? Generally ECM support is optional for G3-supported fax with MH and MR coding schemes. The use of ECM is mandatory for all facsimile messages using the V.34 half-duplex and full-duplex modulation system because of a higher bit rate. ECM is always used in G3 fax with MMR line encoding because an MMR coding scheme does not use EOL codes for synchronization.
26. Why is ECM used with the MMR coding scheme?
MMR is two- dimensional compression. The MMR coding does not use EOL codes between scan lines to limit the effect of line errors and for synchronization. Any scan line data corruption would also corrupt the interpretation of any data that followed it. The use of ECM with MMR reduces or eliminates the errors. ECM is also used with JPEG and JBIG image data in G3 color fax machines.
27. What are the block and frame sizes used in error correction mode (ECM)?
A page of coded data is divided into several small blocks. Each block contains multiple frames up to a maximum of 256 frames. The frame size can be either 256 or 64 octets (bytes) of coded fax data. These values of
frame size do not include facsimile control field (FCF) and frame number octet. Therefore, the total length of the high-level data link control (HDLC) information field, including both the FCF and the frame number octet, is 258 or 66 octets.
28. What is training in a fax call?
Fax machines while establishing capabilities will go through training. Capability and agreed rate do not guarantee the success of a fax at that rate. Training ensures that actual pages can be sent at the agreed rate. If training produces more errors, fax machines renegotiate and retrain at a lower rate before proceeding with actual fax transmission.
29. How does the receiving fax machine respond to line errors in a non-ECM
case?
ECM is retransmission of error frames before switching to the next block. When detecting line errors in non-ECM mode, the receiving fax machine has multiple options to handle errors, as follows:
• Respond to page reception with the ReTrain Positive (RTP) command. This response causes the transmitting fax to go through the training check (TCF) process before transmitting the next page.
• Respond to the page reception with the ReTrain Negative (RTN) command. This response causes the transmitting fax to go through the TCF with a lower modulation scheme.
• Continue with few errors or disconnect immediately when more error lines are received.
30. Describe different phases in a fax call.
The different phases in a fax call are A, B, C1, C2, D, and E. Phase A is for call setup, including dialing and transmission of fax tones CNG and CED. Phase B is the premessage handshake procedure. In Phase B, calling and answering fax machines identify themselves and exchange capabilities data for transmission parameters selection. In Phase C1, the calling unit sends a test pattern to determine the maximum data rate. The answering unit either accepts the data rate or requests a lower data rate. Phase C2 is the actual fax page (image) transfer at an accepted data rate. Phase D is the post-image handshake procedure (e.g., end-of-message confirmation and multipage document procedures). Phase E is for call release or for switching the call to another mode of operation.
31. What are the ANS family signals or fax and modem tones?
ANS is 2100 ±15 Hz, /ANS is 2100 ± 25 Hz with 450 ± 25-ms phase reversal modulated, and ANSam is 2100 ± 1-Hz tone with amplitude modulation at 15 Hz. The /ANSam is 2100 ± 1Hz, amplitude modulated with 15 Hz, and phase reversal modulated with 450 ± 25 ms. Low – speed fax machines (G3) send an ANS tone, whereas high-speed SG3 fax machines send the /ANSam tone during the call setup phase. The tones /ANS and ANSam are used by the dial-up data modem.
32. What are the different answering tone anomalies?
The T.30 standard directs that the answering fax device send an answer tone of 2100 Hz for approximately 2.6 to 4 seconds before sending the first handshake message. Some fax machines send a 1650-Hz or 1850-Hz tone instead of a 2100-Hz tone, and some may omit the answer tone altogether and just begin with the first handshake message.
33. What is the calling indication (CI)? When will this tone be sent by a fax machine?
The calling indication is an alternative to call tone (CT) and carries information to permit the selection of call functions (e.g., facsimile or data). V.34-based modems send this tone during the call setup phase. A signal transmitted from the calling modem indicates the general communication function. CI is transmitted with an on/off cadence. The on periods consist of a repetitive sequence of bits at 300 bps, modulating on a V.21 low-band channel. The CI sequence consists of ten ones followed by synchronization bits and call control bits.
34. What are the different V. 8 signals used in V34 faxes?
The different V.8 signals used in V.34 faxes in the beginning of a fax call phase are listed as follows:
/ANSam tone: The SG3 fax machine answers with an /ANSam tone during a call setup phase.
Call menu signal (CM): A signal transmitted from the calling super G3FE primarily to indicate modulation modes available in the calling terminal when V.34 is enabled. CM consists of a repetitive sequence of bits at 300 bps, modulating the V.21 (L) low-band channel.
CM terminator (CJ): A signal that acknowledges the detection of a JM signal and indicates the end of a CM signal. CJ consists of three consecutive octets of all zeros with start and stop bits, modulating V.21 (L) at
300 bps.
Joint menu signal (JM): A signal transmitted from the answering super group-3 facsimile equipment (G3FE) primarily to indicate modulation modes available jointly in the calling-and-answering super G3FEs. JM consists of a repetitive sequence of bits at 300 bps, modulating the V.21 (H) the high-band channel defined in V.21.
35. What are nonstandard facilities and capabilities in a fax call?
The Group-3 facsimile standard permits a facsimile transmitter to request that the facsimile receiver switch to a nonstandard mode of operation if the receiver is equipped with the appropriate proprietary capability. The nonstandard facilities (NSFs)/nonstandard capabilities (NSCs) signal is used to invoke nonstandard features between two facsimile machines made by the same manufacturer. A digital command called the nonstan-dard facilities setup (NSS) is given in response to NSF/NSC. The features specified by NSF/NSC are not recognized by facsimile machines made by
other manufacturers. If a handshake is completed in the NSF/NSC mode, the format for facsimile transmission could be changed for a proprietary mode of operation.
36. What are the various main timing requirements in a fax call? The T.30 standard specifies various timing requirements:
• If a response to the message is not received within a specified time usually of within 3 seconds, then the messages are repeated up to three times or until an acknowledgment is received. Some unacknowledged messages like digital identification signal (DIS) are repeated every 3 seconds until the timeout of 35 ± 5 seconds.
• The fax machine responds with a message confirmation (MCF) command to the message end of message (EOM) and enters into the beginning of phase B. The timeout to enter into phase B after receiving the MCF command is 6 ± 1 second, which is defined as T2 in the T.30 standard.
• The timeout that a terminal will attempt to alert the local operator in response to a procedural interrupt is 10 ± 5 seconds. Failing to achieve operator intervention, the terminal will discontinue this attempt and shall issue other commands or responses.
• The receiving fax terminal can respond to a postmessage sequence with receiver not ready (RNR), and the calling terminal then queries the receiving fax terminal with receiver ready (RR). This sequence can be repeated up to 60 ± 5 seconds and then disconnects the call after timeout.
37. What is the duration of the silence period between signals using different modulation modes?
The ITU-T T.30 standard specifies a silence period of 75 ± 20 ms between signals using different modulation modes [e.g., the delay between digital command signal (DCS) and the V.27/V.29/V.17 training sequence]. The silence period between end of TCF and response to training [confirmation to receive (CFR) or failure to train (FTT)] is of 3 seconds.
38. How many times are the commands or messages repeated in a fax call? Unacknowledged command messages are automatically repeated after listening for a response for 3 seconds. Commands are sent up to three times or until acknowledgment whichever comes first.
39. What is the necessity of using fax over IP?
Voice transmission is converted to an IP network in many deployments. Maintaining separate PSTN telephone lines for fax is an extra cost. The fax over IP application enables standard fax machines to work with packet networks. VoIP adapters and gateways manage fax packet transmission. Fax signals are distorted through normal low-bit-rate compression of G.729A and G.723.1. Hence, separate packets through T.38 and an associated data pump of V.27ter, V.29, V.17, and V.34 manage the fax signal to create fax payload.
40. What is an IAF device?
An Internet-aware facsimile (IAF) is a facsimile machine that can access the Internet directly. It can work as a combined operation of fax machine and VoIP gateway. For voice, the IP phone serves a similar function.
41. What is special about an IAF device?
Usually IAF supports T.38 real-time and T.37 store-and-forward fax. Hence, it is more capable than the fax machine and fax over IP adapter. T.30 timer values may be extended by two or three times when both fax terminals are identified as IAF devices in negotiation. IAF also works as a fax server and hosts several fax functions. The bit number “123″ in the DIS/DCS message indicates an IAF device in negotiation.
42. What is the store-and-forward fax over IP, and when it is used? Store-and-forward fax is similar to e-mails, but at both end terminations, fax machines and computers are used to interface for fax pages. In the store-and-forward mode, the caller sends the fax messages stored on the one server to another server, and finally, it reaches the destination VoIP fax interface either as an e- mail message or as a fax to a standard fax machine.
43. What is T.37?
The ITU-T T.37 recommendation defines a standard method for store-and-forward delivery of fax through an IP network. Fax messages are sent as multipurpose Internet mail extensions (MIME) encoded e-mail messages using a simple mail transfer protocol (SMTP). T.37 can handle storage of fax – like e – mails.
44. What is real – time fax over IP ?
Real-time fax over IP works like a regular fax call. In real-time fax over IP, fax machines synchronize and send data over the IP link between the two connections. The two popular modes of sending real-time fax over IP link are T.38 fax relay and fax pass-through, which is similar to a G.711 VoIP voice call.
45. What is fax pass-through?
Fax pass-through sends the fax similar to a VoIP voice using G.711 |>law or A-law as compression. A-law and |>law are also used in the PSTN between digital loop carriers and telephone central offices. In VoIP, A-law and | -law are supported in all the systems. Fax pass-through mode makes use of the G.711 or G.726 at 32/40kbps to send fax. The performance matches closely to fax over PSTN under the best conditions. This method is also known as in-band fax transmission over VoIP.
46. What are the different voice modules controlled in a fax pass- through call?
Fax pass-through mode works similar to a VoIP voice call with the G.711 codec. On detecting fax tones (CED or /ANSam) by the gateway, a codec will be switched to G.711 (PCMU or PCMA), and some voice modules of voice activity detection (VAD), packet loss concealment (PLC), dual-tone
multi-frequency (DTMF) rejection, and echo cancellation are disabled for the duration of the fax session.
47. What are the benefits of fax pass-through?
When end – to – end packets delivery is good, fax pass – through creates less end-to-end delay and interoperability improves compared with T.38.
48. What are the main issues in fax pass-through?
Fax pass-through sends A-law or |>law compressed samples on the IP network. IP calls can have many impediments such as delay, jitter, reorder, loss of packets, and fragmentation of packets. For fax machines, continuity of signal is lost with IP impediments. Fax may not go through with end-to-end packet impediments. Fax pass-through requires much higher bandwidth than T.38-based fax.
49. What is the relevance of RFC2198 for fax pass-through?
RFC2198 defines the format to pack redundant real-time transport protocol (RTP) payload. To compensate for packet loss in the IP network, redundant payload will recover the lost packets. RFC2198 will have one or multiple extra previous payloads that will demand higher bandwidth on the IP network. In some deployments, forward error correction (FEC) as per the RFC2733 scheme is also used to minimize the influence of packet impediments.
50. What is T.38?
T.38 is an ITU standard that defines the procedures necessary to transfer facsimile data and signaling packets in real time over the Internet or on any other IP network. T.38 interfaces demodulation bits from V.17, V.27ter, V.29, and V.34 modules into IP packets and at the receiver gives back packets to bits for modulation.
51. What are the advantages of T.38 in VoIP?
T.38 eliminates the need for high-bit-rate codecs for transmission of fax over IP networks. It is more attractive because the bandwidth used is better compared with a in-band G.711-based fax in VoIP. Fax is half-duplex; hence, IP packets will use one side of the bandwidth. It works well even in poor IP network conditions when employing redundancy and FEC. It is also possible to control the properties of the fax session. T.38 can be sent with three times the redundancy compared with G.711.
52. What is the bandwidth gain of T.38 over G.711?
Considering a fax at the 9600-bps rate, G.711 on the Ethernet interface requires 126.4 kbps in each direction. T.38 requires 24.8 kbps in one direction (send to receive direction) because the fax is half-duplex.
53. What are the disadvantages and concerns of T.38?
Many gateways may not support T.38. Interoperation is the critical issue with a T.38 fax. Many deviations exist across fax machines. Fax machine deviations, delays with T.38, and IP network impediments create several timing issues for T.38 and may force other modes of G.711 pass-through
to be used. These concerns can be eliminated by using redundancy and catering to tolerant design in T.38.
54. How will T.30 work with T.38?
T.30 is an ITU standard for fax signaling. It is a PSTN standard that is applicable to both PSTN and fax over IP. Both fax machines will be aware of T.30 signaling. T.38 is not known to fax machines, and it is internal to the VoIP adapters and the IP network. As part of a VoIP-based fax using T.38, once the call is connected and fax negotiation starts, each gateway takes part in the T.30 signaling with the local fax machines, but negotiation is end-to-end. T.30 signals are encoded into packets and relayed over the IP network using T.38 and other VoIP signaling.
55. What is a fax data pump?
This keyword is used by the engineers for representing total modulation and demodulation modules of a fax. The modules of V.21, V.27ter, V.29, V.17, and V.34 are referred to as a data pump. They take fax samples and demodulate to gives bits. They take bits and modulate to provide fax signals.
56. What are different T.38 version numbers?
The T.38 version numbers are 0, 1, 2, and 3. Internet facsimile protocol (IFP) over transmission control protocol (TCP) support is added in version
1. Version 3 supports a V.34 fax. The ASN.1 notation is modified in version
2. The modified Abstract syntax notation.1 (ASN.1) notation in version 2 and previous notations in version 0 or 1 cannot interoperate with each other. T.38 recommends use of version 2 for indicating the ASN.1 syntax. If no version number is provided in negotiation, the default version 0 is assumed.
57. What is the redundancy applicable to T. 38?
Redundancy is used in T.38 in order to recover the lost packets from IP network impediments. User datagram protocol (UDP) transport layer (UDPTL)-based FoIP makes use of the redundancy procedure given in T.38. IFP over the RTP- based fax makes use of a procedure given in RFC2198.
58. What should be the usual redundancy factors?
Redundancy is applicable to both low-speed V.21 HDLC IFP packets and high-speed T.4/T.6 image IFP packets. High-speed redundancy of two to three is very common in deployments. A redundancy factor of four to six is very common for low-speed IFP packets. Compared with pass-through modes, T.38 will consume very less bit rate even for a higher redundancy factor of three.
59. What is the applicability of FEC in T.38?
FEC is forward error correction. The main operation in FEC is an exclusive OR (XOR) operation of present and some ordered previous packets. Normal packets and XOR modified packets are sent on the network. Lost packets can be recovered through packet loss detection logic and XOR
operation. Some products use FEC in place of redundancy. FEC is optional in the implementations. Redundancy is used as a commonly supported operation.
60. How does pass-through compare with T.38?
Pass-through mode is simple for VoIP fax transmission. The inter-op issues will be minimal and work closely with PSTN fax quality. It is not preferred because it demands more bandwidth than T.38. Under packet impediments, fax pass- through fails easily and T.38 performs better than pass-through mode.
61. What is the relevance of FaxLab for T.38?
FaxLab is a tool that emulates about 166 facsimile device profiles. It can be used to perform compatibility testing among several facsimile devices. It can also perform extensive facsimile protocol analysis and identify protocol errors. It can be used to automate multiple fax calls. FaxLab will use only one analog front end on a TIP- RING interface. Hence, exact fax machine front-end characteristics and impedances are not emulated.
62. What are the fax quality measuring recommendations?
The ITU-T-E.453 and E.458 recommendations are used for classifying fax image quality, transmission-i nduced scan line errors, and call success. In general, the ITU-T-E.450 series addresses fax quality. Instruments like FaxLab support these recommendations.
63. Is T.38 applicable with H .323 and SIP ?
T.38 is used with several VoIP signaling protocols. Certain deviations exist with the negotiation and renegotiation phases based on the VoIP protocol used, but the major part of T.38 operations remains the same across all signaling.
64. Why is there no RTP header for T.38 packets?
When the T.38 protocol was standardized, RTP was an emerging protocol. Initial solutions with fax over IP were perfected with UDPTL based and TCP. Recent revisions of T.38 support RTP. Several boxes in the deployment and market support only UDPTL-based implementation.
65. What is ISDN fax, and how does it work?
ISDN is four-wire digital telephony. The minimum configuration supports two channels of true 64-kbps voice or fax and one data channel. Voice and fax use 64- kbps channels, and signaling is managed in a data channel. A G4 fax is transmitted on an ISDN fax.
66. What are the different standards used for ISDN fax?
A list of a few ITU recommendations for ISDN voice and fax are listed here:
• Standards for G4 fax are ITU-T T.6, T.62, T.62bis, and T.85.
• Fax test charts: T.20 and T.21.
• F.161 International Group – 4 facsimile service.
• T.90 characters and protocols for the ISDN terminal.
• F.185, Internet facsimile: Guidelines for the support of the communication of facsimile documents.
• Q.850, usage of cause and location in the Digital Subscriber Signaling System No. 1 and the Signaling System No. 7 ISDN User Part.
• Q.921 operation of LAPD, ISDN User-Network Interface, Data Link Layer Specification.
• X.25 defines only the interface between a subscriber (DTE) and an X.25 network (DCE).
• Q.930: ISDN user-network interface layer 3—general aspects.
• Q.931: ISDN user-network interface layer 3 specification.
67. Why is the UDPTL protocol more popularly used than TCP and RTP in T.38?
UDPTL was the first approved approach for fax over IP. Many implementations in present deployments incorporated UDPTL protocol for fax over IP. T38 over UDPTL takes less bandwidth compared with TCP and RTP.
68. What are the main issues with fax interoperation?
The main issues with fax over IP start with increased end-to-end delay and packet impediments. Several problems in T.38 then multiply as a result of these issues.
69. How is the packet loss issue taken care of in T.38?
For higher packet losses, redundancy and FEC are used in T.38. IFP over UDPTL makes use of redundancy and FEC techniques as per ITU-T-T.38 for error correction. IFP over RTP uses redundancy as per RFC2198 and FEC as per RFC2733 techniques to compensate for a lost packet. When a fax machine is working in ECM-enabled mode, about a 2% packet drop is taken care of by ECM. TCP takes care of packet loss through retransmission.
70. What are the different data rate management methods in T.38?
Two methods of handling TCF are available for determining the highspeed data rate. Method-1 is the local generation of TCF, in which TCF is locally generated by the receiving gateway. This type of locally generating the training data is used in TCP implementations and is optionally used in UDP implementations. Method-2 is transferred TCF used with UDP (UDPTL or RTP) and is not recommended for use with TCP.
71. What is real-time fax with spoofing?
Spoofing techniques are used to extend the delay tolerance of fax machines. Spoofing and jitter compensation allow the fax machines to tolerate network delay without losing communication. These techniques add to the T.30 protocol used by fax machines to communicate, keeping them online beyond their normal T.30 timeout intervals. Fax spoofing inserts pseudo-packets into the data stream to make up for IP network delay that can cause fax device timeouts.
72. Does every fax system manufacturer comply with ITU-T standards, and are there any deviations?
ITU-T recommendations for fax specify the sequence of operations for fax call setup, handshake messages, and procedures for testing and timing. Most fax systems deviate from the ITU-T recommendations in one way or another. Some of these differences are also created by country-specific deviations. The common anomalies and deviations are listed below:
• Frame sequence deviations
• Preamble and flag sequence variations
• Answer tone anomalies
• Timing deviations
• Improper EOM usage
• Unusual data rate fallback sequences
• Common training pattern detection algorithms and incorrect duration of TCF
• Image transmission deviations
• Echo protect tone usage
• Image padding and short lines
• Retrain positive/retrain negative handshake message usage and inability to retransmit messages after receiving RTN
• Long duration lines
• Nonstandard disconnect sequences
• Disconnect (DCN) usage
73. Do all T.38 implementations follow the same packetization format (raw payload excluding the T.38 protocol header)?
Variations exist in different T.38 implementations. Some implementations send one T.30 signal frame in one packet, and others disassemble it and send it in multiple packets. Therefore, a T.38 implementation should handle both situations and assemble the multiple packets when necessary. This principle applies to image packets as well. Some implementations place an entire HDLC frame (between flags) into a single packet; others may ignore the frame boundaries when inserting the data into packets.
