GB2073935A

GB2073935A - Tape format to facilitate error concealment and apparatus for recording and/or replaying same

Info

Publication number: GB2073935A
Application number: GB8110507A
Authority: GB
Original assignee: RCA Corp
Current assignee: RCA Corp
Priority date: 1980-04-11
Filing date: 1981-04-03
Publication date: 1981-10-21
Also published as: DE3114451A1; JPS56159813A; GB2073935B; FR2480472A1; FR2480472B1

Abstract

Sequentially occurring packets of information (1, 2, 3...) are seperated in a tape format by the statistically significant dropout length (SSDL) of the tape to enhance the reliability of conventional dropout error concealment techniques. Due to the statistically significant dropout length separation between the packets of information, each packet of erroneous (B) information (e.g., 2, 6, 10) is replayed in sequence to adjacent packets of statistically reliable (G) information from which estimated packets of information can be derived to conceal the packets of erroneous information. <IMAGE>

Description

SPECIFICATION Tape format to facilitate error concealment and apparatus for recording and/or replaying same In magnetic tape recording, the quality of the tape generally determines the error magnitude, error rate and error occurrence of the information recorded. Two characteristics of the tape predominate in such error determination, the precision with which the particles of the magnetic oxide are aligned on the tape and the frequency at which stipples of magnetic oxide or other defects are found along the length of the tape. Misalignment of the magnetic oxide particles causes background noise in the recorded information, but standard tape quality in this regard has improved to the level where insignificant error is attributable thereto, especially when digital information is recorded.However, the defects along the length of the tape cause a separation between the tape and the recording head, so that dropouts or gaps occur in the recorded information. These dropouts result in significant error, such as when video information is recorded in digital form. Such significant error can be treated with correction techniques by adding coding data to the recorded information and decoding that data during replay, or with concealment techniques by deriving estimated values from other recorded information to substitute for the erroneous information during replay. Since the error concealment techniques are easier to implement, they would be preferred except that such techniques are not reliable because the other recorded information used to estimate the substituted values, may also be erroneous.

In accordance with the principles of the present invention, by formating sequentially occurring packets of information on the tape at locations separated by the statistically significant dropout length (hereinafter SSDL) of the tape, the reliability of error concealment techniques is increased to a point where failure rates are arbitrarily small. Due to the SSDL separation, each packet of erroneous information is replayed in sequence to adjacent packets of statistically reliable information from which estimated packets of information can be derived to conceal the packets of erroneous information. To provide the SSDL separation in a tape record and/or replay system which utilizes this tape format, transducers are specifically arranged on a rotating head in one embodiment of the system, while delay lines are incorporated in another embodiment of the system.

In the DRAWINGS: Figure 1 illustrates a multitrack format for one preferred embodiment of the invention; Figure 2 illustrates the replay sequence of information packets from the tape format of Fig. 1 to show that each packet of erroneous information is disposed between packets of statistically reliable information; Figure 3 illustrates a transducer scanning arrangement for implementing the tape format of Fig. 1 in one record and/or replay system; Figure 4 illustrates a delay line arrangement for implementing the tape format of Fig. 1 in a record system; and Figure 5 illustrates a delay line arrangement for implementing the tape format of Fig. 1 in a replay system.

Although the tape format of this invention could be utilized in longitudinal scan or quadruplex tape record and/or replay systems, a portion of tape is shown in Fig. 1 with a multitrack helical scan format for one preferred embodiment of the invention. In this format, sequentially occurring packets of information are disposed on the tape at locations separated by the SSDL of the tape to assure that when dropouts occurs during replay, each packet of erroneous information is disposed adjacent to packets of statistically reliable information from which estimated packets of information can be derived to conceal the packets of erroneous information. The number of tracks in the multitrack format is not limited but only four tracks A, B, C, and D are shown in Fig. 1 for purposes of illustration.A record and/or replay head 10 (shown in Fig. 3) rotates to simultaneously scan these tracks with individual transducers 12, 14, 16 and 18. Consecutive packets of information are recorded at the SSDL interval on different tracks with the first packet on track A, the second packet on track B, the third packet on track C, the fourth packet on track D, the fifth packet back on track A immediately adjacent to the first packet and so on. The packets of information may be any discrete quantity of the recorded information, such as, in the case of digital video information, individual bits of a picture sample (pixel) or a whole pixel or a number of pixels in digital video recording applications. Each packet of information will contain a conventional signal means, such as check bits, for detecting whether it is erroneous during replay.It may appear that much space on the tape is wasted by this particular format but the SSDL's, and the angle of the tracks relative to the direction of tape travel are shown out of proportion for illustrative purposes. When those skilled in the art consider that the tracks are 16 inches long and less than 0.01 inch wide, while the tape is 1 inch wide and a large SSDL is 0.1 inch in the typical helical scan application, they will understand without further explanation that essentially no tape is wasted by this format.

As was discussed previously, defects along the length of the tape cause the head 10 to be separated from the tape and this results in a dropout of information during the record and/or replay operations. Although the size and frequency of these defects generally depend on the quality of the tape, the nature of these defects on tape of any given quality will also vary. Because the size of each defect determines the length of the dropout that is enccuntered thereform, a size distribution curve can be plotted fcr the defects on any tape and the SSDL determined therefrom in terms of tape track distance. Due to the SSDL separation on the tape between sequentially occurring packets of information, adjacent packets of erroneous information are not highly probable when packets of information are repiayed from the tape in their recorded sequence.This improbability is illustrated in Fig. 2 where it is assumed that a defect exists to cause a dropout across the tape in Fig. 1 so that information packets 2, 8 and 110 are erroneous. Each packet of erroneous information is shown as a "B" while each packet of statistically reliable information is shown as a "G" in Fig. 2 and only G's are adjacent to rhe B's in the replayed information. Therefore, conventional techniques can be reliably applied to conceal the erroneous packets of the statistically significant dropout information.

A transducer scanning arrangement for recording and/or replaying the tape format of Fig. 1 is illustrated in Fig. 3 where the transducers 1 2, 14, 16 and 1 8 are equally spaced from each other about the periphery of the rotary head 10 by a distance equal to the SSDL of the tape. Each transducer 12, 14, 16 and 18 is also located uniquely in the direction of the head's cylindrical axis and posts 20 and 22 guide the tape 24 about a peripheral portion of the head 10. Information to be recorded on the tape in the format of the invention is applied to the transducers 12, 14, 16 and 18 through slip rings (not shown) from a means 26 for quantizing that information into packets of the type selected and for distributing those packets sequentially to the transducers 12. 14, 16 and 18 in the manner selected.The packet quantizing and distributing means 26 would include conventional counters and switches in arrangements that are obvious to those skilled in the art. The same transducer scanning arrangement would also be utilized for replaying the tape format of Fig. 1. Information to be replayed is picked up from the tape by the transducers 12, 14, 16 and 18 and applied through the slip rings to a means 28 for accumulating the packets in their recorded sequence. The packet accumulating means 28 would also include conventional counters and switches in arrangements that are obvious to those skilled in the art.

A delay line arrangement for recording the tape format of Fig. 1 is illustrated in Fig. 4 wherein delay lines 30, 32 and 34 are disposed to separate the packets of information by the SSDL. Information to be recorded is applied to the tape through a transducer scanning arrangement similar to the one shown in Fig. 3, except that the transducers 12, 14, 16 and 18 would be aligned along the direction of the head's cylindrical axis.

The delay lines 30, 32 and 34 are disposed between the transducer scanning arrangement and a means 36 for quantizing the information in packets of the type selected and for distributing the sequential packets to the transducer scanning arrangement in the selected manner. The packet quantizing and distributing means 36 would include conventional counters and switches in arrangements that are obvious to those skilled in the art.

The delay duration of each delay line 30, 32, 34 is predicated on the time required to record a track distance equal to the SSDL and in digital recording systems this time equates to a number N of bits. Since each packet of information in track B is recorded sequentially after a packet of information in track A for the tape format of Fig. 1, packets of information are passed directly to track A and packets of information are passed through the delay line 30 of N bits to track B.Furthermore, each packet of information in track C is recorded sequentially after a packet of information in track B while each packet of information in track D is recorded sequentially after a packet of information in track C and therefore, packets of information are passed through the delay line 32 of two N bits to track C while packets of information are passed through the delay line 34 of three N bits to track D.

As shown in Fig. 5, a delay line arrangement similar to that shown in Fig. 4 would be utilized for replaying the tape format of Fig. 1.

However, the order of the delay lines 30, 32 and 34 is reversed with track D being passed without any delay, track C being passed through the delay line 30, track B being passed through the delay line 32 and track A being passed through the delay line 34 to regroup the packets of information in their recorded sequence. Information to be replayed is picked up from the tape by the transducer scanning arrangement and applied through the delay lines 30, 32, 34 to a means 38 for accumulating the packets in their recorded sequence. The packet accumulating means 38 would also include conventional counters and switches in arrangements that are obvious to those skilled in the art.

When delay line arrangements such as those described above are utilized in a system to record and replay the tape format of this invention, the SSDL of the tape can be adap tively derived. Those skilled in the art will realize without further explanation that circuits could be devised to derive the SSDL while recording and to adjust the length of the delay lines accordingly in the record and replay system. Furthermore, where separate record and replay systems are utilized, the adaptively derived SSDL values could be recorded on the tape to coordinate the length of the delay lines in the replay system with those of the record system for each tape unit on which information is recorded.

Claims

1. A tape format for a record and/or replay system, comprising: sequentially occurring packets of information being disposed on the tape at locations separated by the statistically significant dropout length of the tape to assure that when dropout occurs during replay, each packet of erroneous information can be replayed in sequence to adjacent packets of statistically reliable information from which estimated packets of information can be derived to conceal the packets of erroneous information.

2. The tape format of claim 1 wherein said sequentially occurring packets of information are disposed on separate tracks in a multitrack format.

3. The tape format of claim 2 wherein four individual tracks are included in said multitrack format.

4. In a tape recorded system of the type wherein video information is recorded on the tape: means for separating sequentially occurring packets of information on the tape by the statistically significant dropout length of the tape to assure that when dropout occurs during replay, each packet of erroneous information can be replayed in sequence to adjacent packets of statistically reliable information from which estimated packets of information can be derived to conceal the packets of erroneous information.

5. The tape record system of claim 4 wherein said information packet separating means includes a rotating head having transducers arranged thereon to provide the statistically significant dropout length separation between the information packets.

6. The tape record system of claim 4 wherein said information packet separating means includes an arrangement of delay lines through which the staiistically significant dropout length separation between information packets is provided.

7. In a tape replay system of the type wherein video information is replayed from the tape: means for gathering packets of information in their recorded sequence from the tape on which they are separated by the statistically significant dropout length of the tape and thereby assure that when dropout occurs during replay, each packet of erroneous information is replayed in sequence to adjacent packets of statistically reliable information from which estimated packets of information can be derived to conceal the packets of erroneous information.

8. The tape replay system of claim 7 wherein said information packet gathering means includes a rotating head having transducers arranged thereon to pickup sequentially recorded packets of information from tape locations separated by the statistically significant dropout length.

9. The replay system of Claim 7 wherein said information packet gathering means includes a delay line arrangement for aligning sequentially recorded packets of information from tape locations separated by the statistically significant dropout length.

10. Recorded tape having information recorded thereon in a format substantially as hereinbefore described with reference to Figs.

1 and 2 of the accompanying drawings.

11. A tape record or replay system substantially as hereinbefore described with reference to Fig. 3, 4 or 5 of the accompanying drawings.