WO2007060589A2

WO2007060589A2 - Method and apparatus for error-correction in a communication system

Info

Publication number: WO2007060589A2
Application number: PCT/IB2006/054327
Authority: WO
Inventors: Kun-Wah Yip
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 2005-11-28
Filing date: 2006-11-20
Publication date: 2007-05-31
Also published as: WO2007060589A3

Abstract

The present invention provides an error-correcting method and apparatus for communication systems. The method comprises receiving information data in a first signal burst and a first check data segment in the burst, and then estimating the received information data based on the first check data segment and determining whether the estimated information data is error-free or not. If the estimated information data is error-detected, a corresponding check data segment from a second signal burst is received and combined with the first check data segment to re-estimate the received information data, and this process continues until an error-free information data is obtained.

Description

METHOD AND APPARATUS FOR SIGNAL ERROR- CORRECTION IN COMMUNICATION SYSTEM

FIELD OF THE INVENTION The present invention relates to a communication technique, and more particularly, to a method and apparatus for signal error-correction in a wireless broadcasting system.

BACKGROUND OF THE INVENTION

With increasing popularization of mobile communication terminals, pure mobile voice communication can no longer satisfy people's demand to acquire information. More and more new applications exhibit great prospect due to their capabilities to provide business or entertainment contents are more convenient and have more information, among which people may receive and view TV programs on their mobile terminals. It's an important business opportunity for broadcasters to provide the service of viewing TV programs on mobile terminals. Such an application may further be extended to provide large file download to a plurality of end users. Such file download may not be limited to TV video stream, which also can be for other contents such as real-time gaming etc.

Thus, ETSI (European Telecommunications Standards Institute) has issued DVB-H (Digital Video Broadcasting-Handheld) specification, which is an extension to the DVB-T (Digital Video Broadcasting - Terrestrial) standard, with a focus on handheld device applications. While DVB-T has been proved to possess excellent performances in aspects of fixed, mobile, portable reception and so on, it still needs further improvement for handheld devices, in aspects such as power consumption, cell performance in presence of movement and network design. To this end, new technical modules are introduced into DVB-H, mainly including time slicing for decreasing power consumption at receivers, and

MPE-FEC (Multi-Protocol Encapsulation-Forward Error Correcting) for increasing Doppler tolerance and improving the capacity against pulse interference.

In the method of time slicing, data are transmitted in bursts. Accordingly, for a receiver, some components of a receiver, like RF (radio frequency) front end and DVB-H decoder, are enabled only during a burst time, and are disabled otherwise. As the burst time only occupies a small fraction of the total time, this operation mode may significantly reduce power consumption for the receiver. The MPE-FEC technique is to apply an additional FEC scheme in a data link layer besides existing FEC methods in the physical layer used for DVB-T systems. In DVB-H systems, Reed-Solomon (RS) error-correcting code is used for this additional FEC scheme. With the MPE method, transmission of IP datagrams is possible in DVB-H systems. These IP datagrams are RS encoded to generate the check data. The IP datagrams and the check data are conveyed in the MPE section and the FEC section, respectively.

Specific technical details of the DVB-H specification can be found in Annex F of ETSI document EN 300 744 Vl.5.1, entitled "Digital video broadcasting (DVB); framing structure, channel coding and modulation for digital terrestrial television," November 2004; and ETSI document EN 301 192 Vl.4.1, entitled "Digital video broadcasting (DVB); DVB specification for data broadcasting," November 2004. Specific implementation guidelines for DVB-H networks are given in ETSI document TR 102 377 V 1.1.1, entitled "Digital video broadcasting (DVB); DVB-H implementation guidelines," February 2005.

Furthermore, the techniques of time slicing and MPE-FEC have been disclosed in patent application WO 2005/032034A1 filed on 23rd September 2004 by J. Vesma and H.

Pekonen, entitled "Burst transmission". Both the patent application and the ETSI document TR 102 377 V 1.1.1 propose that the receiver may discard receiving data in the FEC sections when the received data in the MPE sections are decoded correctly, which leads to further power saving for the receiver. This is because multiple users in a broadcasting system are situated at different geographical locations to receive the broadcast message.

When a particular user is very close to the broadcasting station, the received signal strength is considerably strong. At this time, it is possible that the FEC scheme implemented in the physical layer is sufficient to guarantee that the recovered data are correct, without involving the MPE-FEC scheme, which is implemented in the link layer, so that the receiver of the particular user does not need to receive check data in the FEC sections. If a user is far away from the broadcasting station, it is possible that all FEC sections are required to ensure correct recovery of the data. In case that a user is neither far away nor close to the broadcasting station, it is possible that the receiver of the user can ensure correct recovery with only part of, not all of, the FEC sections. However, decoding of data in the FEC sections takes certain time. The FEC sections disclosed in prior arts are closely spaced, and a receiver does not have sufficient time to perform FEC decoding and hence does not know when to stop receiving the remaining FEC sections. No related solution is provided in the above-mentioned patent application and the ETSI documents.

It is, therefore, necessary to provide a better error-correcting scheme to realize selective reception of the FEC sections flexibly so as to further reduce power consumption while maintaining better data recovery performance.

OBJECT AND SUMMARY OF THE INVENTION

An object of the present invention is to provide an error-correcting method and apparatus for communication systems, which enables selective reception of check data so as to achieve a good balance between reducing power consumption and maintaining the error-correcting performance.

An error-correcting method for communication systems according to the present invention, comprising steps of:

(a) receiving information data and a first check data segment in a first signal burst;

(b) estimating the received information data based on the first check data segment; (c) determining whether the estimated information data is error- free or error-detected;

(d) receiving a corresponding check data segment from a second signal burst if the estimated information data is error-detected; and

(e) re-estimating the received information data by combining the first check data segment with the corresponding check data segment. An error-correcting apparatus for communication systems according to the present invention, comprising: a receiving means for receiving and outputting information data and check data segments in signal bursts; a decoding means for decoding the information data and the check data segments outputted from the receiving means to estimate the information data, and determining whether the estimated information data is error- free; and a storage means for storing the information data and check data segments currently obtained by the decoding means for future use when the estimated information data is error-detected; wherein the decoding means is further configured to obtain corresponding check data segments in subsequent signal bursts from the receiving means when the estimated information data is error-detected, and combine them with the check data segments stored in the storage means to estimate the information data stored in the storage means.

With the error-correcting method and apparatus proposed in the present invention, check data segments for error-correction on the same set of information data are arranged to be transmitted in different bursts such that the receiver has sufficient time for decoding, so as to realize selective reception of check data and decrease power consumption for the receiver.

Other objects and effects together with a fuller understanding of the invention will become apparent and appreciated by referring to the following descriptions and claims taken in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

Detailed descriptions will be made below to the invention with reference to accompanying drawings and specific embodiments, wherein:

Fig. 1 illustrates the arrangement of information data and parity-check data in each burst in conventional DVB-H signal;

Fig.2 illustrates the arrangement of information data and parity-check data in each burst in DVB-H signal according to an embodiment of the present invention;

Fig.3 is a flow chart showing the error-correcting method for receivers of the present invention according to the DVB-H signal structure of Fig.2; Fig.4 illustrates the arrangement of information data and parity-check data in each burst according to another embodiment of the present invention; and

Fig.5 is a block diagram showing a receiver structure of the present invention. Throughout all the above drawings, like reference numerals will be understood to refer to like, similar or corresponding features or functions.

DETAILED DESCRIPTION OF THE INVENTION

To implement selective reception of check data, the present invention proposes a novel arrangement of check data segments, with which the check data segments for error-correction on the same set of information data can be scheduled to be transmitted in different bursts, such that a receiver may have sufficient time to perform decoding and hence determine when to stop reception of the remaining check data segments to reduce power consumption for the receiver. In a conventional DVB-H system, a user's data stream is transferred in bursts. Each burst consists of MPE section, which contains information data, and FEC section, which contains parity-check data. The information data and the parity-check data in the n th burst can be denoted by d_n and p_n , respectively. Fig. 1 depicts the arrangement of d_n and p_n in a DVB-H signal. Although not indicated in Fig. 1, it should be understood that data d_n and p_n in a burst are further manipulated by physical layer functions, including CRC

(Cyclical Redundancy Check) bits incorporation, channel coding, interleaving, scrambling, OFDM (Orthogonal Frequency Division Multiplexing) modulation and so on, before transmission over the air. Furthermore, according to the provision of DVB-H standard, the burst should further include a header for conveying parameters related to the burst and the

IP datagrams therein, which is not shown in Fig.l for purpose of simplicity. In addition, d_n further contains some data for enabling the receiver to determine whether the received d_n is correct, such as CRC data. The corresponding parity-check data p_n is generated by the transmitting side (for example, a broadcasting station) based on d_n . Referring to Fig.l, p_n is adjacent to d_n in the same burst. For a receiver, decoding of p_n is continuous and takes a relatively long time, so it's very difficult to make a determination whether to stop receiving the remaining p_n or not during the decoding process.

In the present invention, however, p_n is partitioned into a number of parts and these parts are transmitted in different bursts. A simple example will be given below to describe the technical solution of the present invention in detail with reference to Fig.2. In this embodiment, p_n is partitioned into two parts, p_n and p_n , which may have different lengths. Symbolically, p_n may be represented as P_n = P_n I P_n , where "I" denotes concatenation. Accordingly, the bursts are arranged in the manner as shown in Fig.2. Both d_n and p° are in the n th burst, and p* is in the ( n + 1 )th burst. Such a data arrangement is applicable to either the data link layer or the physical layer.

The FEC-encoded output data p° can be obtained by applying a prior art FEC method, which is called Mother Code, to encode d_n and then puncturing on the obtained

P_n Fig.3 is a flow chart showing the error-correcting method for receivers of the present invention according to the DVB-H signal structure of Fig.2. With reference to Fig.3, first, a user receives d_n and p_n from the n\h burst (Step SlOl), wherein d_n and p_n denote the received d_n and p_n respectively. The received d_n and p_n have possibly been corrupted by noise, interference, and multi-path fading during transmission. d_n and p_n may have already been processed by the physical layer functions, such as channel decoding to correct some possible errors. The receiver then estimates d_n by decoding d_n and p°

(Step S 102). Here, the estimated d_n is denoted by d° , wherein the superscript "0" indicates that the estimated d_n is obtained from a knowledge of p° .

By means of CRC bits embedded in d_n , the receiver can determine whether d_n is error-free (step S 103). If errors are detected in d° , the receiver can receive p_n in the

( n + 1 )th burst (Step S 104), wherein p_n denotes the received pj, . The combination of p_n and p_π results in p° l p_π (Step S 105).

The receiver then estimates d_n again based on d_n and P_n I p_n (Step S 106). As more parity-check data are contained in p° I p_π than in p° alone, it is more likely that the estimated d_n (denoted as dj, herein) is error-free.

Finally, d[ may be outputted as the decoded d_n (Step S 107).

If at Step S 103 there is no error detected in d° , the receiver can skip receiving p_π by early shutting down some corresponding components of the receiver during the ( n + 1 )th burst and output d_n directly as the decoded d_n . The above description is made to an example, where p_n is partitioned into only two parts. However, the present invention is not limited to this case and the arrangement of p_n in bursts may be designed flexibly according to the system requirement in practical use.

The following description is made to an embodiment with reference to Fig.4, in which p_n is partitioned into two or more parts. In this embodiment, p_n is partitioned into M parts, wherein M is an integer larger than or equal to 2. It yields p_n = p° I pi I • • • I P^ ^X ' where p^ , i = 0,1,- • -,M - 1 , can be of different sizes. The data in the nth burst is denoted by S_n ands_n = d_n I p_n I p

¹ _^1 1 p_n_₂ 1 • • • I . The data arrangement in the bursts of the embodiment is shown in Fig.4.

Two requirements need to be satisfied for the burst structure of Fig.4. First, after the n th burst is processed by the physical layer, the physical (time) positions of its data constituents, namely, d_n,p°,p¹ _n_₁,- -

, in the physical signal remain the same positions. Secondly, no data symbol from one constituent contributes to the formation of another constituent after the physical layer functions are performed, that is, each constituent of each data is independent. These two requirements ensure that the receiver is able to discard receiving some of p_n , i = 0,1,- - -,M - I .

The receiver processes the received signals according to the order of the bursts. Similar to the flow chart for reception according to the embodiment in Fig.3, the receiver first receives d_n and p° from the n\h burst, then estimates d_n by decoding d_n and p° , and determines whether d° is error-free. If d° is error-detected, reception of p* in the (n+l)th burst continues and p° is combined with p_n to yield p° l p_n so as to estimate d_n based on d_n and p° I p* . If an error- free d_n is estimated, it will be outputted as the decoded d_n . If the estimated d° is still error- detected, the receiver continues reception of p^ in the (n+2)th burst and yields p° I p* I p^ to estimate d_n . This process continues until an error-free d_n is estimated.

According to the burst structure shown in Fig.4, a corresponding receiver structure is also provided in the present invention. In the following description, S_n = d_n I p° I p_n__! I P

^₂ 1 • • • I represents the received data in the nth burst, wherein d_n and p^ respectively denote the received d_n and p'_n__t , i = 0,l,- - -,M - 1 .

Referring to Fig.5, the receiver comprises a receiving unit 10, a decoding unit 20, a storage unit 50, a register file unit 30 and a register file update unit 40.

The receiving unit 10 is configured to process the input physical received signals and output d_n,p°, P^₁, - -,P^₊₁ . The receiving unit 10 can be controlled by a first external signal, and generate a first BUSY signal to indicate whether the receiving unit 10 is "BUSY". Here, the first external signal can be generated by an external unit (not shown), which may be a timer and the timer can obtain the information about the start time of the nth burst. The decoding unit 20 is configured to process the signals outputted from the receiving unit 10. And the decoding unit 20 can be controlled by a second external signal, and generate a second BUSY signal to indicate whether the decoding unit 20 is "BUSY". Similarly, the second external signal can also be generated by the external unit.

The decoding unit 20 may consist of a FEC decoder and an error- detection unit (not shown). The FEC decoder can perform FEC decoding in a conventional decoding way such that A_n is estimated from d_n ,p° ,p* ,- - -,p^ , wherein i=0~M-l. The error- detection unit can determine whether the A_n estimated by the FEC decoder is error- free.

The storage unit 50 is configured to temporarily store the information data and parity-check data obtained from the received bursts, for future use of the decoding unit 20. The register file unit 30 can be accessed by both the receiving unit 10 and the decoding unit 20. The register file unit 30 stores a first set of state values I_₁,I_₂,- -

and a second set of state values /ό,/^,- - -,/^^ , wherein each state value may be

ERROR-FREE or ERROR-DETECTED. The first set of state values and the second set of state values are used to record information indicating whether the previously and currently estimated information data after performing FEC decoding on all received bursts is error-free or error-detected.

The register file update unit 40 is configured to update the first set of state values stored in the register file unit 30 with the second set of state values therein, under the control of the decoding unit 20. It is apparent to those skilled in the art that, depending on particular implementation, a buffer 60 may be added between the receiving unit 10 and the decoding unit 20, for temporarily storing output signals from the receiving unit 10 before the decoding unit 20 can process the output signals. In addition, the receiver can have sufficient time for decoding as long as the time interval between two adjacent bursts is set appropriately. To give a detailed description to the operation principle and procedure of the receiver of the present invention, it's assumed that the n th burst is about to arrive and the receiver has processed all the previous bursts. In the register file unit 30, the state value I ₁ , i e {1,2,- ",Af - 1} , contains information on whether the estimated A_n-1 , after FEC decoding on all the previous bursts, is error-free (indicated by /_, = ERROR-FREE) or still error-detected (indicated by /_, = ERROR-DETECTED). Concurrently, for i that satisfies the condition I ₁ = ERROR-DETECTED, the storage unit 50 stores the corresponding p^.p^,.- - -.^:¹, as well as d_n__r

At some time before the arrival of the n th burst, the receiving unit 10 and the decoding unit 20 are enabled (waken up) from the sleeping state through the first external signal and the second external signal, respectively. As required, waking up of the receiving unit 10 and the decoding unit 20 may be simultaneous or not.

The receiving unit 10 then retrieves the values of Z ₁ , 1 ₂ , • • • , I__(M __1} from the register file unit 30. Based on these values, the receiving unit 10 determines the value of q such that I _q = ERROR-DETECTED and all of /_,__!, /_,_₂,- - -,/__(Λf_₁₎ are equal to

ERROR-FREE. In this case, the value of q indicates that the estimated information is error-detected from the (n-q)th burst. If /_, = ERROR-FREE for all i = 1,2,- - -,Af - I , assign q = 0, that is, the information data in all previous bursts have been estimated correctly. Further, the decoding unit 20 may also obtain the value of q.

When the n th burst arrives, the receiving unit 10 receives d_n and p° therefrom, and then sends these two data sections to the decoding unit 20. The decoding unit 20 decodes on (d_n , p° ) to estimate d_n , wherein the estimated d_n is denoted as d° . The decoding unit 20 then determines whether d_n is error-free. If so, the state of I_o' of the register file unit 30 is set as ERROR-FREE and d° is outputted as the decoded d_n . If d° is not error-free, the decoding unit 20 sets the state of I_o' of the register file unit 30 as

ERROR-DETECTED, and stores d_B and p° into the storage unit 50. If q = 0, it means that the information data in all previous bursts have been estimated correctly, then the receiver skips the following steps a) and b); otherwise these steps are repeated for i = 1,2, • • • , q . a). If /__; = ERROR-FREE, the receiving unit 10 no longer receives the subsequent data p^ and step b) will not be performed; if I ₁ = ERROR-DETECTED, the receiving unit 10 continues receiving p_n'__t and then sends them to the decoding unit 20. b). The decoding unit 20 obtains p_n' __t from the receiving unit 10 and retrieves

"'?!-¹ _! fr°^{m me} storage unit 50. Using these two data sections, FEC decoding is performed on (<!„_,, p^p^,- -,!)^) to estimate d_n__t . Here, the estimated d_n__t is denoted as d'_n__t . Then the decoding unit 20 determines whether d'_n__t is error-free. If so, it sets V₁ = ERROR-FREE while clearing the values of d^p^p^,- - -,]}^ in the storage unit 50, and outputs d_n' __t as the decoded d_n__l . If d_n' __t is still error- detected, the decoding unit 20 sets V₁ = ERROR-DETECTED and stores d'_n_, and p^ into the storage unit 50.

While the decoding unit 20 performs signal processing, the receiving unit 10 may receive other subsequent parity-check data segments.

After ji_n ^q__q is received, the receiving unit 10 indicates through the first BUSY signal that all the operations necessary for it to perform on the n th burst have been done. The first BUSY signal is used to indicate whether the receiving unit 10 is in an idle state or is processing data and thus unable to receive new bursts.

After d'_n__l is obtained, the decoding unit 20 instructs the register file update unit 40 to update the first set of states by copying the value of each V₁ onto the corresponding /__(!+1) for i = 0,1,2,- • •, M - 2. If q + l<M - 2 , the copying of the corresponding value may be skipped, for i = q + l,q + 2,- - -,M - 2 . Additionally, the decoding unit 20 indicates through the second BUSY signal that all the operations necessary for it to perform on the n th burst have been done. The second BUSY signal is used to indicate whether the decoding unit 20 is in an idle state or is processing data and thus unable to receive new bursts. The above-mentioned external units for generating the first and second external signals may respond to the first and second BUSY signals, for example, upon receipt of the first or second BUSY signal, the receiving unit 10 or the decoding unit 20 is controlled by the external units into sleeping state through the first or second external signal, respectively. According to the error-correcting method and the corresponding apparatus of the present invention, the check data is partitioned into two or more parts for transmission in several bursts. For a user near the broadcasting station, the signal strength at its location is relatively high, so reception of some, not all of, the check data, may satisfy the error-correction requirement for the receiver, which leads to reduction to the receiver's power consumption. For a user far away from the broadcasting station, error-correction performance may be improved by receiving more check data, without establishing an additional communication link to request the broadcasting station to retransmit data, which may reduce the extra cost caused by establishment of an additional communication link. Therefore, selective reception of check data is realized with the present invention, thus to achieve a good balance between saving power consumption and maintaining error-correction performance.

The error-correcting method and apparatus of the present invention is applicable to not only DVB-H systems, but also any broadcasting or communication systems using time slicing.

It is to be understood by those skilled in the art that various improvement and modifications can be made to the method and apparatus for signal error-correction as disclosed in the present invention without departing from the basis of the present invention, the scope of which is to be defined by the appended claims herein.

Claims

CLAIMS:

1. A method for signal error-correction in communication systems, comprising steps of: (a) receiving information data and a first check data segment in a first signal burst;

(b) estimating the received information data based on the first check data segment;

(c) determining whether the estimated information data is error-free or error-detected;

(d) receiving a corresponding check data segment from a second signal burst if the estimated information data is error-detected; and (e) re-estimating the received information data by combining the first check data segment with the corresponding check data segment.

2. The method according to claim 1, further comprising steps of:

(f) determining whether the re-estimated information data is error-free or error-detected;

(g) continuing receiving a corresponding check data segment from next signal burst if the re-estimated information data is error-detected;

(h) re-estimating the received information data by combining the first check data segment with all the corresponding check data segments; and (i) repeating step (f) to step (i) if the information data estimated at step (h) is still error-detected.

3. The method according to claim 2, further comprising step of: stopping reception of subsequent corresponding check data segments and outputting the estimated information data when the estimated information data is error-free.

4. The method according to claim 3, wherein the first check data segment and all the corresponding check data segments are obtained by partitioning a check data into a plurality of segments, the check data generated based on the information data in the first signal burst according to a predefined criterion.

5. An apparatus for signal error-correction in communication systems, comprising: a receiving means for receiving and outputting information data and check data segments in signal bursts; a decoding means for decoding the information data and the check data segments outputted from the receiving means to estimate the information data, and determining whether the estimated information data is error- free; and a storage means for storing the information data and check data segments currently obtained by the decoding means for future use when the estimated information data is error-detected; wherein the decoding means is further configured to obtain corresponding check data segments in subsequent signal bursts from the receiving means when the estimated information data is error-detected, and combine them with the check data segments stored in the storage means to estimate the information data stored in the storage means.

6. The apparatus for signal error-correction according to claim 5, wherein the corresponding check data segments are obtained by partitioning check data into a plurality of segments, the check data generated based on the information data in signal bursts according to a predefined criterion.

7. The apparatus for signal error-correction according to claim 5, wherein the decoding means is further configured to output the estimated information data when the estimated information is error-free.

8. A communication device, comprising: an apparatus for signal error-correction, comprising: a receiving means for receiving and outputting information data and check data segments in signal bursts; a decoding means for decoding the information data and the check data segments outputted from the receiving means to estimate the information data, and determining whether the estimated information data is error-free; and a storage means for storing the information data and check data segments currently obtained by the decoding means for future use when the estimated information data is error-detected; wherein the decoding means is further configured to obtain corresponding check data segments in subsequent signal bursts from the receiving means when the estimated information data is error-detected, and combine them with the check data segments stored in the storage means to estimate the information data stored in the storage means; and a control means for controlling to enable the receiving means and the decoding means at start time of the signal bursts and disenable the receiving means and the decoding means after they have processed currently inputted data.

9. A method for transmitting signals in a communication system, comprising steps of:

(a) generating a check data based on a predefined information data;

(b) partitioning the check data into a plurality of check data segments;

(c) transmitting the predefined information data and one of the check data segments in a burst; and

(d) transmitting other check data segments in other bursts, respectively.

10. The method for transmitting signals according to claim 9, wherein the check data is partitioned into a plurality of serial check data segments according to the data arrangement order at step (b).

11. The method for transmitting signals according to claim 10, wherein the check data segment transmitted in step (c) is the first one of the plurality of serial check data segments.

12. The method for transmitting signals according to claim 11, wherein other consecutive check data segments are transmitted in subsequent consecutive bursts respectively in step (d).

13. A signal to be transmitted in a communication system, comprising an information data and a check data, wherein the check data comprises a plurality of serial check data segments, the information data and one of the check data segments are configured for transmission in a same burst and other check data segments are configured for transmission in other bursts respectively.

14. The signal according to claim 13, wherein the check data segment to be transmitted with the information data in the same burst is the first segment of the plurality of serial check data segments.

15. The signal according to claim 14, wherein the other serial check data segments are transmitted in subsequent consecutive bursts respectively.