Packet error indication method

~ WO95/19080 2 1 8 0 3 4 2 PCT~S94/07861 PAC~ET ERROR INDICATION MET~OD BA~K~4~.J OF T~ I~Y~.~ l~N Field of the Invention :: The present invention is directed to packet-based communications systems, and more particularly, to a method for providing packet error indication in a multi-hop communications system.

Descri~tiQn of the Prior ~rt In a packet based multi-hop communications system, data is transmitted in packets from a transmission site to a reception site via a plurality of successive physical links.

For example, in future digital subscription television systems, packetized video and audio data may be transmitted from a programmer site to a cable head-end installation via a satellite tr~no~ ;on system, and after reception at the head-end, may be transmitted to individual subscriber locations via a cable distribution network. The satellite transmlssion system and cable distribution network each define a separate "physical link~.

Typically, prior to transmission over a given physical link, the packets of data to be transmitted are encapsulated in a "data link layer" (as defined in the International Standards Or~n-7~ion (ISO) Open System Interconnection (OSI) reference model (ISO 7498)) that provides, among other things, forward error correction capabilities. Essentially, a forward error correcting code is appended to each packet, or a series of packet~, prior to l~nn~hlng the packets over the physical link. At the terminal side of the physical link, the error correcting code WO95/19080 ~1 8 0 3 4 2 PCT~S94107861 is employed to correct, if possible, any errors that may have occurred during transmission over that physical link.

Popular error correcting techniques include Reed-Solomon block error coding and convolutional coding.

Although error correcting codes are capable of correcting certain data errors, some errors may be 80 extensive that correction is impossible.

However, the error correcting techniques are at least able to identify such uncorrectable errors.

In prior art multi-hop communications systems, when an uncorrectable packet error is detected by the receiving equipment of a particular physical link, the erroneous packet is simply discarded and the communications system must deal with the loss of that packet; most systems assume that the data in a corrupted packet is useless.

However, certain kinds of data, such as coded video data, may be useful despite an uncorrectable data error, and it would therefore be desirable to allow packets having uncorrectable data errors to continue through successive physical links to the ultimate reception site. However, merely allowing a corrupted packet to propagate through the system presents a problem, however, in that there is no way of alerting the ultimate reception site that an error occurred during transmission of that packet over a one or more of the physical links. A need arises therefore for a method of providing an indication to a reception site in a multi-hop communications system that an uncorrectable error occurred during tr~n~m;RR;on of a packet over one of the physical links in the communications system. The present invention satisfies this need.

~ ARY OF T~E lN Vh~ The present invention is directed to a method of indicating to a reception site in a packet-based multi-hop communications system whether an error occurred during transmission of a particular packet over one or more of the physical links in the system. According to the method of the present invention, prior to tr~nAm;AR;~n, a pre-defined group ~ WO95/19080 2 1 8 0 3 4 2 PCT~S94~78~1 -- 3 of bits in each packet to be transmitted is set to a first value which indicates that no errors are present in the data of that packet, i.e., that the packet data is valid. ~or each packet, the following steps are then performed at each successive physical link after tr~nF-iAA1On of the pac~et over that physical link: (a) detecting whether any uncorrectable errors occurred in the packet during trAnPml RA; on over that physical link; (b) if an uncorrectable error is detected, setting the pre-defined group of bits in the packet to a second value wherein the second value is indicative of an uncorrectable error in the data of the packet; and (c) relaying the packet to:the next successive physical link. A device at the lllt;r-te reception site may, upon receiving the packet, examine the value of the predefined group of bits to determine the~integrity of the Idatain the packet.

In a most preferred embodiment, the predefined group of bits in each packet comprises a single bit and is located in the header portion of the packet.

Additional features and advantages of the present invention will become evident hereinafter.

~RIEF DEsr~Tp~T~N OF T~E ~ The foregoing summary, as well as the following detailed description of the preferred embodiment, is better understood when read in conjunction with the appended drawings. For the purpose of illustrating the invention, there is shown in the drawings an : ' '; t that is presently preferred, it being understood, however, that .he invention is not limited to the specific methods and instrlm~ntAlit;es disclosed. In the drawings:

Figure 1 is a block diagram of an exemplary pAcketbased multi-hop communications system within which the method of the present invention may be employed;

Figure 2 is a flow diagram illustrating a prior art method for ~An~l ;n~ uncorrectable packet errors in a multihop communications system;

WO95/19080 ~l 8 0 3 4 2 PCT~S94/0786l Figure 3 illustrates the content and arrangement of a packet header in accordance with one embodiment of the present invention; and Figure 4 is a flow diagram illustrating one embodiment of the method of the present invention.

DETAILED D~K1~ OF THE ~rir~ VIM~ Referring to the drawings wherein like numeral~ indicate like elements throughout, there is shown in Figure l an exemplary packet-based, multi-hop communications system 10 in which the method of the present invention may be employed.

As shown, data 12 to be transmitted is provided to an encoder 14 which segments and packetizes the data 12 to generate a stream of packets 16 for transmission to a reception site.

Each packet comprises a header portion 18 and a payload portion 20. Data 12 is carried in the payload portion 20 of each packet while the header portion 13 rrnt~;n~ systemrelated information necessary for operation of the communications system and for recovery of the data 12 from each packet 16 at the reception site.

As further shown, each successive packet 16 is transmitted to the reception site via a plurality of successive physical links, e.g. PHYSICAB ~INKS 1, 2 .. N. As ~plAin~ in the Background of the Invention, a given physical link may comprise a satellite transmission system, a cable distribution network, an optical fiber network etc.

Each physical link provides error correction of packets transmitted through that physical link ~typically by means of a data link layer as defined by the ISO OSI reference model).

For example, considering PHYSICAL LINK 1, packets 16 to be transmitted over that physical link are provided to an error coder 22a prior to transmission. Error coder 22a operates in accordance with any known error correction technique, such as a Reed-Solomon block error technique or convol-lt;~n~l coding technique, to generate an error correcting code 24a for each successive packet. As illustrated, the error correcting codes 24a are appended to and transmitted with their WO95119080 2 1 8 0 3 4 2 PCT~S9~/0~6l -- 5 -respective packets 16. Each successive packet 16 is then provided to a transmitter device 26a for transmission over a respective physical medium.

At the tGrm;n~l side of the physical link, a receiver 28a receives each successive packet 16 and provides each packet to an error correcting apparatus 30a. For each successive packet, the error correcting apparatus 30a employs the appended error correcting code 24a to detect and, if possible, correct any data errors that may have occurred during transmission of that packet over the physical medium.

Assuming that no errors occurred, or that any errors that did occur were corrected, PHYSICAB LI~R 1 will relay the packet to the next physical link ~e.g. PHYSICAL LINK 2) for transmission over that link. Transmission of each packet will ~nnt i nnr in this manner until the packet reaches the designated reception site. As indicated by the use of subscripts 'a', 'b' and 'c' in Figure I, each successive physical link may employ its own error correction and transmission techniques. At the reception site, a decoder 32 may be employed to recover the data 12 from the payload portions of each incoming packet 16.

As mentioned in the Background of the Invention, in prior art multi-hop communications systems, when an uncorrectable packet error is detected by the receiving equipment (e.g., error correction apparatus 30) of a particular physical link, the packet is simply discarded, and therefore, is not relayed to the next physical link for cnnt i nnP~ trangmissiOn through the system. Figure 2 illustrates the operation of the receiving equipment of a prior art multi-hop communications system in greater detail.

As shown, at step 40 a packet is received by a receiver (e.g., receiver 28) at the terminal end of the physical link, and at step 42, an error correction apparatus (e.g., apparatus 30) employs the appended error correcting code to determine whether an error occurred during trnnr~1~sion of the packet over that physical link. If no error is detected, the packet is simply relayed to the next physical link at W095/19080 2 1 8 0 3 4 2 PCT~S94107861 -- 6 step 48. If, however, an error is detected, control passes to step 44.

At step 44, the error correcting apparatus determines whether the error is correctable. If the error is correctable, the error is corrected at step 46 and the packet is then relayed to the next physical link at step 48. If, however, the error is uncorrectable, the packet is discarded at step 50 and control passes back to step 40 for processing of the next packet. The corrupted packet is not relayed to the next physical link. The communications system must therefore deal with the lo88 of that packet.

Although uncorrectable errors in the data of a packet may indeed render the packet useless to a reception site, Applicants have recognized that certain types of data, such as coded video data, may be useful despite an uncorrectable data error. Accordingly, it would be desirable to allow corrupted packets to continue through the communications system to the reception site despite the existence of an uncorrectable data error. Applicants have rrror,n;7e~, however, that some m~rh~n;~- must be provided for alerting the reception site that an uncorrectable error occurred during tr~n~ ;on over one or more of the physical links in the system. A corrupted packet cannot simply be allowed to propagate through the system. Each physical link assumes that the packets provided to that link are valid, and any information concerning the integrity of a packet is lost when the packet is relayed to a subsequent physical link.

Thus, if a current physical link receives a corrupted packet from the previous link, and no errors occur during transmission over that current link, the current link will assume the packet is error free. A reception site w;ll therefore have no way of knowing whether a packet is valid.

As explained, prior art systems simply avoid this problem by discarding packets that contain uncorrectable data errors.

The present invention is directed to a method that allows packets rnn~;n;ng uncorrectable data errors to cnn~innP to propagate to a reception site by providing a CA 02180342 1998-07-30 W O 9S/19~80 ~ , PCTrUS94107~61 ~' -- 7 means for indicating to the reception site that an uncorrectable error occurred during transmission of that packet. Essentially, in accordance with the method of the present invention, a single bit in each packet, or alternatively, a group of bits, is pre-defined as a "packet error indicatorl'. At the transmission site, the pre-de~ined bit in each packet is set to a ~irst value, which indicates that no data errors are present in that packet, i.e., that the packet data is valid. At each successive physical link in the communications system, i~ an uncorrectable error is detected in a given packet, the packet error indicator bit in that packet is set to a second value which indicates that an uncorrectable error exists, i.e., that the packet data is invalid. Once set to the second value, the pre-defined bit(s~ may not be reset by any subse~uent physical links. A device at the reception site may, upon receiving the packet, examine the value of the pre-defined bit to determine the integrity of the data in the packet. As explained, the data in a packet may be useful to the reception site despite the existe}lce of an uncorrectable error.

Recently, Applicants and their Assignee proposed the method of the present invention, as de~ined by the appended claims, to the International Organization for Standardization (ISO) for inclusion in the recently adopted MPEG-2 Systems standard, and the method o~ the present invention, as de~ined by the appended claims, has been adopted as part of that standard. Accordingly, a pre~erred embodiment of the present invention is described herein in the context of a communications system wherein data is transmitted in accordance with the MPEG-2 System standard.

The MPEG-2 Systems standard is set forth in detail in the MPEG-2- Systems Committee Draft (ISO/IEC JTCl/SC29/WGll/No601, November, 1993) [hereinafter "MPEG-2 Systems Committee Draft"] The MPEG-2 Systems standard speci~ies a packet structure for the transmission of data from a transmission site to a reception site.

Each packet is re~erred to therein Woss/l9080 2 1 a o 3 4 2 PCT~594/07861 as a "Transport Packet" and a sequence of transport packets comprise a "Transport Packet Stream". An exemplary Transport Packet Stream 17 is illustrated in Figure 3. As shown, the Transport Packet Stream 17 comprises a number of successive Transport Packets 16. As explained in the Background of the Invention, and as is common in most packet-based systems, each Transport Packet 16 comprises a header portion 18 and a payload portion 20. Figure 3 further illustrates thç general content and arrangement of the header portion 18 of an MPEG-2 Transport Packet. As shown, the header portion 18 of each Transport Packet 16 comprises a number of fields, including a sync byte field 60, a payload unit start indicator field 64, a transport 5, 1 1 ;nr control field 66, a tLan~LL priority field 68, a packet identifier (PID) 70 field, an adaptation field control field 72 and a continuity counter field 74.

The small number in the bottom right hand corner of each field indicates the number of bits allocated for that field.

The function of each of these fields is described in detail in the af~L. ~;rnr~ MPEG-2 Systems Committee Draft.

In accordance with a preferred embodiment of the present invention, a one bit field 62 of each header 18 is pre-defined as a ~packet error indicator" ~note: the MPEG-2 Systems Committee Draft designates this field as a "Transport Error Indicator").

Whether a packet contains an uncorrectable bit error may be ~ In; r~ted through the value of the packet error indicator field 62. A first value may indicate that no error is present in the packet, and a second value may indicate that the data in the packet contains an uncorrectable error. In the preferred embodiment, the first and second values are '0' and '1', respectively. It is understood, however, that in other embodiments, different bit values may be used to indicate the presence or absence of packet errors. Additionally, in other 'a~; ~, the packet error indicator field 62 could comprise a group of bits.

According to the method of the present invention, prior to transmission of each packet 16 from the transmission WO 9~;/19080 2 1 8 0 3 4 2 PCT/US94J07861 g site, the packet error l~dicator field 62 of each packet 16 is aet to the first value, e.g. ~0', indicating that the packet data i8 free of errors. The packet error indicator field 6a of each packet may be set or "initialized~ by the encoder 14 of Figure 1 as it constructs each successive packet. Re~erring now to Figure 4, the following additional steps are performed at each successive physical link in the communications system. At step 80 a packet transmitted over the physical link is received at the t~rmln~l end thereof.

At step 82, an apparatus, e.g. the error correcting apparatus 30 of Figure 1, detects whether an error occurred during transmission of that packet over the physical link. If no error ls detected, control passes directly to step 88 where the packet is simply relayed to the next physical link. If, however, an error is detected, control passes to step 84.

At step 84, the error correcting apparatus determines whether the error is correctable. If the error is correctable, the error is corrected at step 86 and the packet is then relayed to the next physical link at step 88. If, however, the error is uncorrectable, control passes to step 90. At step 90, the packet error indicator field (i.e., bit) 62 in the header portion 18 of the packet is set to the second value, e.g. '1', to indicate that at least one uncorrectable bit error is present in the data of that packet. After setting the packet error indicator field 62, the packet 16 may then be relayed to the ne~t physical link.

According to the present invention, once the packet error indicator field 62 of a given packet 16 has been set to the second value, indicating that an uncorrectable error has occurred, the packet error indicator field 62 may not be reset to the first value unless the data in error is somehow corrected.

Each packet will eventually reach its destination at the reception site. A decoder at the reception site (e.g., decoder 32 of Figure 1) may, upon receiving a given packet, examine the value of the packet error indicator field WO95/19080 21 80342 PCT~Sg4/07861 ~ -- 10 62 in the header portion 68 of the packet to determined the integrity of the data 5~nt~in~ in that packet.

As the foregoing illustrates, the present invention is directed to a method of providing packet error indication in a multi-hop communications 6ystem. In particular, the present invention UV~L'_ -~ the deficiencies in prior-art systems by providing a means for indicating to a reception site whether an uncorrectable error occurred during trans~ission of a packet over one of the physical links in the system. By providing such an indication, the method of the present invention avoids the need to discard packets whenever an uncorrectable error occurs. As explained above, in some instances, the data in a corrupted packet may be useful at a reception site despite the presence of an uncorrectable bit error.

It is understood that changes may be made to the e~bodiments described above without departing from the broad inventive concepts thereof. Accordingly, this invention is not limited to the particular . '-'i nt~ disclosed, but is ;ntPn~d to cover all modification3 that are within the scope and spirit of the invention as defined by the appended claims.