Title of Invention

METHOD , APPARATUS AND SYSTEM FOR ENHANCING ROBUSTNESS OF PREDICTIVE VIDEO CODECS USING A SIDE-CHANNEL BASED ON DISTRIBUTED SOURCE CODING TECHNIQUES

Abstract

A method apparatus and system for providing distributed source coding techniques that improve data coding performance, such as video data coding, when channel errors or losses occur. Errors in the reconstruction of the data is eliminated or reduces by sending extra information. Correlation between a predicted sequence and an original sequence can be used to design codebooks and find the so-sets required to represent the original image. This information may be sent over another channel, or a secondary channel.

Full Text

FORM 2 THE PATENTS ACT, 1970 (39 of 1970) & THE PATENTS RULES, 2003 COMPLETE SPECIFICATION (See section 10, rule 13) "METHOD, APPARATUS AND SYSTEM FOR ENHANCING ROBUSTNESS OF PREDICTIVE VIDEO CODECS USING A SIDE-CHANNEL BASED ON DISTRIBUTED SOURCE CODING TECHNIQUES" QUALCOMM INCORPORATED, an American company of 5775 Morehouse Drive , San Diego, California 92121-1714, United States of America The following specification particularly describes the invention and the manner in which it is to be performed. WO 211(15/12007') 2 PCT/US2005/019157 METHOD, APPARATUS, AND SYSTEM FOR ENHANCING ROBUSTNESS OF PREDICTIVE VIDEO CODECS USING A SIDE-CHANNEL BASED ON DISTRIBUTED SOURCE CODING TECHNIQUES Claim of Priority under 35 U.S.C. §119 [0001] The present Application for Patent claims priority to U.S Provisional Application No. 60/576,193, entitled "METHOD, APPARATUS, AND SYSTEM FOR ENHANCING ROBUSTNESS OF PREDICTIVE VIDEO CODECS USING A SIDE-CHANNEL BASED ON DISTRIBUTED SOURCE CODING TECHNIQUES", filed June 1, 2004, and assigned to the assignee hereof and hereby expressly incorporated by reference, in its entirety, herein. Reference to Co-Pending Applications for Patent [0002] The present Application for Patent is related to the following co-pending U.S. Patent Application Serial No. 10/396,982, entitled "Method, Apparatus and System for Encoding and Decoding Side Information for Multimedia Transmission" filed March 24, 2003, assigned to the assignee hereof, and expressly incorporated by reference herein. BACKGROUND I. Field [0003] The present invention relates generally to multimedia over error-prone networks, and more specifically to video coding and decoding techniques used in conjunction with multimedia over wireless networks. II. Background [0004] Current video compression techniques, such as, ITU-T H.263, H.263+, H.263++, ISO MPEG-4 and JVT H.264 have become increasingly efficient in compressing video. The ability to achieve good visual quality at relatively low bitrates has led to increased popularity of video applications in emerging multimedia applications over bandwidth limited channels, such as, wireless channels. However, the predictive nature of compression techniques employed by current video codecs makes WO 2(1(15/12(M»7«> PCT/US2005/019I57 3 the compressed video bit-streams very fragile and susceptible to degradation due to packet losses in the channel. [0005] Channel errors or packet loss can lead to loss of synchronization between a video encoder and a decoder. More specifically, a locally decoded copy of a reconstructed video frame at the encoder may not match the corresponding reconstructed video frame at the decoder. This loss of synchronization between a video encoder and a decoder is also sometimes termed "drift" since there will be a drift between the receiver's decoder and the sender's decoder. Drift can lead to significant losses in decoded video quality. [0006] The reduction in decoded quality due to a drift is a direct consequence of the prediction-based coding that underlies current video codecs. That is, to currently decode the current video frame, all previous video frames need to have been reconstructed correctly. This problem is exacerbated in a wireless channel where packet losses are far more frequent than in a wire-line network and tend to occur in bursts. [0007] Re-transmission schemes, such as Automatic Repeat Request (ARQ), or Forward Error Correction (FEC) schemes have been used to combat drift problems between the encoder and decoder and thereby alleviate the loss in quality due to packet losses. However, using ARQ or FEC schemes (or a combination of both) may not prove effective since the latency constraints of the video application may preclude the use of such schemes. Moreover, FEC-based schemes cannot guarantee that the data will be received and if it is not, the drift continues until the next infra-frame1 is received. [0008] The Wyner-Ziv Theorem is discussed in A. D. Wyner and J. Ziv, "The rate distortion function for source coding with side information at the decoder," IEEE Trans. Inf. Theoiy, vol. 22, pp. 1-10, Jan 1976. This Theorem addresses the problem of source coding with side-information. Consider two correlated sources X and Y. The encoder needs to compress the source A" when only the decoder has access to the source Y. When the mean squared error is the distortion measure and X— Y + N where N has a Gaussian distribution, the rate - distortion performance for coding X is the same whether or not the encoder has access to Y. [0009] Recently, based on the principles of source coding with side-information within a Wyner-Ziv framework joint source-channel coding techniques have been proposed to address the problem of drift. See, for example, R. Puri and K. Ramchandran, "PRISM: 1 An intra-frame is used by video encoders to break the prediction loop and hence stop drift. WO 2005/12007') 4 PCT/IIS2005/OI9157 A New Robust Video Coding Architecture based on Distributed Compression Principles," in AUerton Conference on Communication, Control and Computing, 2002; A. SehgaL A. Jagmohan, and N. Ahuja, "Wyner-ziv Coding of Video: An Error- Resilient Compression Framework," IEEE Trans, on Multimedia, vol. 6, pp. 249-258, 2004; A. Aaron, S. Rane, R. Zhang, and B. Girod, "Wyner-Ziv Coding of Video: Applications to compression and error resilience," in Proc. IEEE Data Compression Conf., 2003; and A. Aaron, S. Rane, D. Rebollo-Monedero, and B. Girod, "Systematic Lossy Forward Error Protection for Video Waveforms," in Proc. IEEE Int. Conf. Image Proc,2003. [0010] The codecs discussed in R. Puri and K. Ramchandran, "PRISM: A New Robust Video Coding Architecture based on Distributed Compression Principles," in AUerton Conference on Communication, Control and Computing, 2002; and A. Aaron, S. Rane, R. Zhang, and B. Girod, "Wyner-Ziv Coding of Video: Applications to compression and error resilience," in Proc. IEEE Data Compression Conf, 2003, are full-fledged video codecs that eschew the predictive coding framework. [0011] On the other hand, the codecs discussed in A. Sehgal, A. Jagmohan, and N. Ahuja, "Wyner-ziv Coding of Video: An Error-Resilient Compression Framework," IEEE Trans, on Multimedia, vol. 6, pp. 249-258, 2004; and A. Aaron, S. Rane, D. Rebollo-Monedero, and B. Girod, "Systematic Lossy Forward Error Protection for Video Waveforms," in Proc. IEEE Int. Conf Image Proc, 2003, retain the predictive coding framework but send some extra information to mitigate the effect of drift. [0012] Notwithstanding these advances, there is a need in the art for techniques that not only enables robust video delivery over wireless channels by helping alleviate the effects of packet losses on compressed video bit-streams transmitted over such wireless channels, but also satisfies the requirements of • Be able to work with existing compression techniques • Have low processing latency • Quickly eliminate/reduce the drift between the video encoder and decoder • Require low bitrate when compared to that used by the video codec WO 2005/120II7') 5 PCT/US2005/019157 SUMMARY OF EMBODIMENTS |0013] Embodiments disclosed herein address the above stated needs by providing distributed source coding techniques that improve video coding performance when channel errors or packet losses occur. [0014] The following terms are used • Main stream: video data compressed using predictive codecs • Main channel: logical channel on which Main stream is transmitted • WyZER stream: Wyner Ziv Error Resilience stream constructed using distributed coding principles • Secondary channel: logical channel on which WyZER stream is transmitted • Main channel encoder: encoder used to encode main stream • Main channel decoder: decoder used to decode main stream • WyZER encoder: encoder used to encode WyZER stream • WyZER decoder: dncoder used to decode WyZER stream • Side information: decoded main stream data which is used by WyZER decoder to reconstruct a frame at a receiver • DSC: Distributed Source Coding [0015] Correlation between a video frame at the encoder and a reconstructed video frame at the decoder in presence of channel errors is estimated. In one embodiment, the correlation is computed in the discrete cosine transform (DCT) domain. The correlation can be used in finding cosets as described in U.S. Patent Application Serial No. 10/396,982, filed March 24, 2003, entitled "Method, Apparatus and System for Encoding and Decoding Side Information for Multimedia Transmission." [0016] Embodiments of the invention can help alleviate the drift problem in conventional video codecs by WyZER stream as an error resilience stream. Embodiments can help reduce or eliminate drift by sending extra information. For example, in one embodiment, a secondary channel technique can be used in which WyZER stream can be used to stop drift and improve video quality. The correlation between a predicted sequence and an original sequence can be used to design codebooks and find the cosets required to represent the WO 2005/121(0796 6 PCT/US2005/019I57 original image. This information may be sent over another channel, or a secondary channel. In another embodiment, the number of cosets required can be communicated to a WyZER decoder over a secondary channel in the form of header information. [0017] In one embodiment, using DSC, another description of the video sequence can be sent at a lower rate over a secondary channel which can be used to correct errors, caused by channel losses, in a reconstructed video frame at the receiver. The bit-stream sent over the secondary channel can be decoded with the help of the video frame reconstructed by the predictive decoder. As a result, the effects of error propagation can be mitigated. [0018] In one embodiment, if the decoder can decode only main stream data, it will discard the additional information transmitted on the secondary channel. If the decoder can decode both streams, it will decode the main stream data first, and then use this information to decode the WyZER stream data and present a final reconstructed image to the user. BRIEF DESCRIPTION OF THE DRAWINGS [0019] Figure 1 is a block diagram of a wireless communication system. [0020] Figure 2(a) is a plot of the peak signal to noise ratio (PSNR) verses frame number for a video sequence encoded with an H.263+ encoder and with DSC techniques. [0021] Figure 2(b) is a plot of the PSNR verses frame number for the a video sequence encoded with an H.263+ encoder, with FEC, and with DSC techniques. [0022] Figure 2(c) is a plot of the PSNR verses frame number for another video sequence encoded with an H.263+ encoder, with FEC, and with DSC techniques. [0023] Figure 2(d) is a plot of PSNR verses error percentage for a video sequence entitled "Stefan" encoded with an H.263+ encoder, with FEC, and with DSC techniques. [0024] Figure 2(e) is a plot of PSNR verses error percentage for a video sequence entitled "Football" encoded with an H.263+ encoder, with FEC, and with DSC techniques. [0025] Figure 2(f) is a plot of PSNR verses mean burst length for the "Football" video sequence encoded with FEC and with DSC techniques. WO 2007/120079 7 PCT/US2005/019157 |0026] Figure 3 is a block diagram illustrating portions of a wireless communication system that can implement the DSC techniques. [0027J Figure 4 is a flow diagram illustrating an embodiment for determining intra- frame bit allocation to a side-channel encoder. [0028] Figure 5 is a flow chart illustrating an embodiment of determining a maximum number of partitions of a source codebook that will be used by a side-channel encoder. [0029] Figure 6 is a block diagram of the WyZER encoder [0030] Figure 7 is a block diagram of the WyZER decoder DETAILED DESCRIPTION [0031] Current video coding standards use a predictive framework to achieve good compression performance for video signals. Examples of such compression schemes include MPEG-4, H.263+ and H.264. However, when such compressed bitstreams are transmitted over error prone channels, the video quality is severely damaged, as demonstrated using objective PSNR metric in Figures 2(a)-(c). Examples of error prone channels include cdma2000®, WCDMA, GSM, and other emerging wireless networks. In this invention, video data compressed using predictive codecs is called "Main stream" and the logical channel on which main stream is transmitted is called "Main Channel". [0032] Examples to mitigate the adverse effects of channel errors include retransmissions of transmission units with errors (RIP retransmissions, hybrid ARQ, etc.), error resilient packetization (data partitioning where semantically more important bits in transmissions units are provisioned with greater error protection, data interleaving, etc.), bitstream syntax to limit the extent of lost source data (usage of ^synchronization markers, reversible variable length coding RVLC, etc.) [0033] In this invention, a separate bitstream is created to be used along with Main stream to improve error resiliency. This stream is constructed using Wyner-Ziv coding of the video data based on Distributed Source Coding principles. We call this stream Wyner Ziv Error Resilience stream (WyZER stream). The WyZER stream is transmitted on a "Secondary channel", a logical channel that is different from the main channel used for video codecs such as MPEG-4, H.263+ and H.264. The video frame reconstructed WO 2005/120079 8 PCT/US2005/019157 by the predictive video coder serves as the side-information to the WyZER decoder for the purpose of decoding the WyZER stream. [0034] According to another aspect of the invention, techniques are provided to estimate the correlation between the erroneously reconstructed frame and the frame itself in the discrete cosine transform (DCT) domain using some of the concepts detailed in R. Zhang, S. L. Regunathan, and K. Rose, "Optimal intra/inter mode switching for robust video communication over the Internet," in Proc. 33rd Ann. Asilomar Conf. on Sig. Syst. Comp., 1999. [0035] According to another aspect of the invention, the WyZER encoder operates in the DCT domain to take advantage of the energy compaction properties of the DCT. Depending upon the rate available for the side-channel and the correlation estimate, as many DCT coefficients as possible are encoded, starting from the DC and low-frequency coefficients. [0036] If the client receives both the Main stream and the WyZER stream, but is only capable of decoding the Main stream, (i.e., it only has a predictive decoder, such as an MPEG-4 or H.263+ or H.264 decoder) then the client will only decode the Main stream and ignore the WyZER stream. A client with both a predictive decoder and a WyZER decoder, can decode both bit-streams and hence, can receive a better quality of decoded video than a client with only a predictive decoder. [0037] While the algorithms of A. Sehgal, A. Jagmohan, and N. Ahuja, "Wyner-ziv Coding of Video: An Error-Resilient Compression Framework," IEEE Trans, on Multimedia, vol. 6, pp. 249-258, 2004; and A. Aaron, S. Rane, D. Rebollo-Monedero, and B. Girod, "Systematic Lossy Forward Error Protection for Video Waveforms," in Proc. IEEE Int. Conf. Image Proc, 2003, also use the erroneous frame reconstructed by the predictive coder as the side information at the secondary channel decoder, there are a number of important differences between the present invention and the techniques disclosed in these documents. For example, in A. Sehgal, A. Jagmohan, and N. Ahuja, "Wyner-ziv Coding of Video: An Error-Resilient Compression Framework," IEEE Trans, on Multimedia, vol. 6, pp. 249-258, 2004, the authors mark certain frames as "peg" frames. These techniques allow error propagation to occur from one peg frame to the next. At each peg frame, some extra information is sent from the encoder to the decoder which allows the decoder to correct for the errors in the peg frame. WO 20115/120079 9 PCT/US2005/019157 [0038] In contrast, aspects of the present invention stop, or reduce, the drift, or errors, as soon as possible. In this way, it is possible to maintain a more or less steady quality which is important from a visual perspective. [0039] On the other hand, in A. Aaron, S. Rane, D. Rebollo-Monedero, and B. Girod, "Systematic Lossy Forward Error Protection for Video Waveforms," in Proc. IEEE Int. Conf. Image Proc, 2003, an independent description of the video sequence is sent with each frame, but with coarser quantization. This coarser description is coded using a Wyner-Ziv framework and can be decoded with the help of the erroneous frame at the decoder serving as the side-information. Thus, the method of the above paper does not attempt to stop the drift since at best it can recover the coarse description. The techniques discussed in the above paper also work in the pixel domain and not in the DCT domain, and thus gives up on exploiting the spatial correlations. [0040] For source coding with side information, the encoder needs to encode the source within a distortion constraint, while the decoder needs to be able to decode the encoded codeword subject to the correlation noise N between the source and the side-information. While, the results proven by Wyner and Ziv are non-constructive and asymptotic in nature, a number of constructive methods to solve this problem have since been proposed wherein the source codebook is partitioned into cosets of a channel code that is matched to the correlation noise N. The number of partitions or cosets depends on the statistics of N. [0041] Figure 1 is a system block diagram illustrating aspects of encoded video over a wireless channel according to one embodiment. The system comprises a predictive encoder 10, including a frame buffer, that communicates a bit-stream to a predictive decoder 20 over a wireless main channel 25, and a secondary channel (WyZER) encoder 30 that communicates a bit-stream to a secondary channel (WyZER) decoder 40 over a wireless secondary channel 45. [0042] In Figure 1, the predictive decoder 20 receives the bit-stream from the predictive encoder 10 over a wireless main channel 25 and reconstructs a signal X which is a representation of the video signal X input to the encoder 10. The predictive decoder 20 reconstructed bit-stream, X, may be affected by errors in transmission of the bit-stream over the wireless channel. The reconstructed signal X serves as side-information to the WyZER decoder 40 which outputs X as the final reconstruction. WO 2005/120079 10 PCT/US2005/019I57 |0043] The input video signal X is the current video signal being encoded by the predictive encoder 10 in one embodiment, the predictive encoder 10 may compress the input video signal and send a compressed bit-stream based on techniques, such as, Motion Estimation (ME) and transforms of Displaced Frame Differences (DFDs). The WyZER encoder 30 sends an independent description of X using a Wyner-Ziv framework to the WyZER decoder 40. [0044] The WyZER decoder 40 then decodes the bit-stream sent over the secondary channel, and using the reconstructed signal X, outputs the final reconstruction of the video signal A" as X. The final reconstructed signal X will generally be a better reconstruction of the original video signal X than the reconstructed signal X because, as explained further below, the WyZER decoder 40 may be able to correct for some of the errors that occur during transmission of the bit-streams over the wireless channels. The final reconstructed signal X may be written back to the frame buffer of the predictive decoder 20. [0045] In this embodiment, for the purpose of designing the WyZER encoder 30, it is assumed that the rate allocated to the WyZER encoder is fixed and that the correlation structure is X = X + Z , where X represents the original input signal, X represents the reconstruction of the original signal X by a predictive decoder, and Z represents the correlation noise vector. [0046] Further, in this embodiment, because operations take place in the discrete cosine transform (DCT) domain, it is assumed that the components of the correlation noise vector Z are independent. The components can be modeled as Gaussian distributed because, of all distributions with the same variance, the Gaussian has the maximum entropy and as such provides a worst case analysis. The WyZER encoder 30 needs to find the number of partitions of the source codebook, i.e. the number of cosets, for each component in X, for which it needs to know the variances of each component of Z. To find these variances, correlation estimation techniques can be utilized as described below. [0047] For example, in one embodiment, the correlation estimation algorithm uses some aspects of the algorithm proposed in R. Zhang, S. L. Regunathan, and K. Rose, "Optimal intra/inter mode switching for robust video communication over the Internet," in Proc. 33rd Ann, Asilomar Conf. on Sig. Syst. Comp., 1999. However, unlike the above publication, this embodiment of the present invention is implemented in the DCT WO 2005/120079 11 PCT/US2005/019I57 11 domain. For both the main channel and secondary channel, it is assumed that packet erasures on the two channels are independent, and that packets are independently decodable. It is also assumed that the probability of losing a packet is equal to the probability of losing the information for a block. When a block is lost the predictive decoder replaces the block with the block in the frame memory at the same position in a previous frame. The packet loss probability on the main channel and secondary channel are represented as p and q, respectively. [0048] As described in co-pending U.S. Patent Application Serial No. 10/396,982, entitled "Method, Apparatus and System for Encoding and Decoding Side Information for Multimedia Transmission" a WyZER encoder can communicate a coset index to the WyZER decoder. The WyZER decoder then decodes the index to identify the coset that includes the codeword that is jointly typical with the side-information. In one embodiment, the concepts detailed in S. S. Pradhan and K. Ramchandran, "Distributed source coding using syndromes (DISCUS): Design and construction," in Proc. IEEE Data Compression Conf., 1999, can be used. In general the source codebook can be partitioned into cosets of any channel code. [0049] Detailed block diagram of the WyZER encoder is shown in Figure 6. It consists of three blocks, distortion statistics calculator, coset encoder and frame statistics calculator. The WyZER encoder uses the current video frame data and the expected reconstruction of the video frame at the decoder to calculate the statistics of the correlation noise. These statistics are used by the coset encoder to generate the WyZER bitstream. After WyZER encoding the frame statistics are updated so they can be used in the next frame. [0050] Figure 7 shows the block diagram of the WyZER decoder. The WyZER bits consisting of the coset bits and the coset index information are used along with the side information (from the main decoder) to generate the WyZER decoded output. [0051] Following the notation of R. Zhang, S. L. Regunathan, and K. Rose, "Optimal intra/inter mode switching for robust video communication over the Internet," in Proc. 33rd Ann. Asilomar Conf. on Sig. Syst. Comp., 1999, let gni,k be the original value of the jthDCT coefficient in zig-zag scan order in the k'h block of the n'h frame. And let gni,k denote its encoder reconstruction, that is gni,k is the quantized representation of gni,k. WO 2005/12007 12 PCT/IJS2005/019I57 Let this coefficient be re-constructed as gni,k by the predictive decoder. The reconstruction gni,k is a random variable for the encoder. [0052] Consider the situation when there is no secondary channel. There are two cases - the block is either intra (I block) or inter (predictive, or P block) coded. An I block is self-contained, that is, it includes all of the information needed to render, or display, one complete block of data. A P block is not self-contained and will typically contain differential information relative to the previous frame, such as motion vectors and differential texture information. Consider that p is the probability that a information unit is lost due to channel errors. Further, assume that this region corresponding to the lost information unit is filled by the previously reconstructed, co-located video. Then, if the block was intra-coded, block is lost, it is simply replaced by the block in the same position in the previous frame. It should be noted that even if the predictive bitstream for a particular block is received without errors gni,k may not equal gni,k,i.e., the block may nonetheless be in error because there may have been previous errors. The expected distortion for this case is: [0053] The expected distortion, d'nk, can be calculated for all the DCT coefficients in the block and the WyZER encoder should encode for this correlation. The goal is to refine the erroneously reconstructed DCT coefficients to their error-free quantized representations to stop the drift. However, it may not be possible to encode all the WO 2005/120079 13 PC T/US2005/019157 coefficients given the rate constraints of the secondary channel. The right approach here is to use the available rate to refine the subset of DCT coefficients that would minimize the expected distortion. However, it is quite complex to perform this optimization for every block. Based on offline tests it was found that most of the time the best strategy is to refine the lower frequency coefficients before refining the higher frequency coefficients. This also makes intuitive sense from a visual perspective. Thus for the online algorithm, only the m lowest frequency DCT coefficients that can be accommodated within the given rate for the secondary channel were transmitted. For these m DCT coefficients, the WyZER decoder will recover their encoder reconstructions at the decoder when both the Main and WyZER streams get through. Thus for these m DCT coefficients, the reconstruction at the decoder will be: [0054] For the remaining DCT coefficients that are not encoded by the WyZER encoder, gni,k is given by Equation (2) above. [0055] To compute the expected distortion dni,k using Equation (3), it is needed to compute and (because and are already known at the encoder). Equations (1), (2), and (4) can be used to relate a and to and . Thus the WyZER encoder can compute and using a simple recursive algorithm (similar to that used by R. Zhang, S. L. Regunathan, and K. Rose, "Optimal intra/inter mode switching for robust video communication over the Internet," ," in Proc. 33rd Ann. Asilomar Conf. on Sig. Syst. Comp., 1999) and thus compute dni,k using Equation (3). Note that for the purpose of encoding, the predictive encoder divides the frame into a grid of non-overlapping spatial blocks. If the predictor block (the jth block in the (n - l)th frame in the above analysis) does not align with this grid of non-overlapping blocks, the WyZER encoder would not have stored the values of and while encoding the (n-1)th frame (since j is not a valid block in the (n - \f frame). In this case, the block with the maximum overlap to the j"' block WO 2005/1201179 14 PCT/US2005/019157 in the (n-\)th frame that is a valid block is used in calculation. Instead, a weighted average of the corresponding blocks in the (n-l)th frame overlapping the jth block can also be used. [0056] The correlation noise estimate (dni,k) determines the number of partitions of the source codebook or number of cosets. This information, the number of cosets, may be communicated to the WyZER decoder over the secondary' channel in the form of header information. To restrict the size of the header, the possible value that the number of cosets can take, may be limited, for each coefficient. The limit on the possible values may be based on extensive tests. [0057] Figures 2(a)-(f) are graphs of simulation results for various performance comparisons that demonstrate the effectiveness of the WyZER codec described above. For the simulations, standard video sequences, that are well known to those in the art, were encoded with an H.263+ coder used as the predictive video codec. The secondary channel is allocated about 30% of the rate on the main channel. [0058] Figure 2(a) is a plot of the peak signal to noise ratio (PSNR) verses frame number for a video sequence encoded with an H.263+ encoder 202 and the DSC techniques 204 described above. In Figure 2 (a), the 3rd frame of the standard "Flower Garden" video sequence (352x240, 15fps, 1 intra frame followed by 29 inter frames) was dropped. To make the comparison fair, H.263+ is compensated with the same amount of extra rate. Figure 2(a) shows that with the DSC techniques, the PSNR increases steadily following the dropped frame, as the drift is corrected, to about 7-8 dB better than H.263+ by the end of the thirty frames sequence. [0059] The DSC techniques described above were also tested using a wireless channel simulator that adds packet errors to multimedia data streams transmitted over wireless networks conforming to the CDMA 2000 IX standard 5. For these tests, the DSC technique was modified to allow a portion of the rate on the secondary channel to be allocated to forward correcting codes (FECs) and the remaining portion to be allocated to the WyZER stream. A motivation for this, is that the WyZER stream cannot recover from loss instantaneously (see Figure 2(a)), unlike in the case when FEC is used and the number of erasures is less than the FECs correction capability. This is important from both visual and PSNR perspectives. However, the entire rate on the side-channel is not allocated to FECs because, as mentioned before, it cannot guarantee error recovery. This is especially true for wireless channels which often have significant error bursts. WO 2005/120079 PCT/US20U5/019157 15 [0060J Figure 2(b) is a plot of the PSNR verses frame number for the a video sequence encoded with an H.263+ encoder 212, FEC 214, and a DSC technique 216 described above. The standard video sequence used in generating Figure 2(b) was the "Stefan" video sequence (both 352x240, 15 fps, 1 GOP). In Figure 2(b) the three different encoding techniques compared are: 1) H.263+ encoded at a rate equal to the total allocated to the main and secondary channels (curve 212), 2) Full secondary channel rate allocated to FEC (we use Reed-Solomon codes) (curve 214), and 3) Proposed algorithm, modified by allocating part of the rate on the secondary channel to FEC (again RS codes are used) (curve 216). [0061] For both the FEC and DSC techniques, (2) and (3) above, the latency constraint is 1 frame. Tests were conducted to find the optimal rate partition between FEC and the WyZER stream in the case of (3). In most cases it was best to allocate roughly equal rates to FEC and the WyZER stream and this is the setup for the results presented here. The Main and WyZER streams are sent over the same (simulated) wireless channel. [0062] In the case of the Stefan sequence shown in Figure 2 (b), the proposed WyZER 216 approach outperforms the others, achieving an average PSNR that is 2.9 dB higher than the FEC-only case 214, and 5.2 dB higher than baseline H.263+, (curve 212). [0063] Figure 2(c) is a plot of the PSNR verses frame number for another video sequence encoded with an H.263+ encoder 222, FEC 224, and a WyZER technique 226 described above. The standard video sequence used in generating Figure 2(c) was the "Football" video sequence (both 352x240, 15 fps, 1 GOP). Figure 2(c) show the typical performance of the three encoding schemes for the Stefan and Football sequences (both 352x240, 15 fps, 1 GOP), for a packet loss rate of 8.5%. [0064] In Figure 2(c), the PSNR of the proposed DSC technique 226 takes two dips in Frame 23 and 26, but in both cases it recovers quickly. The average PSNR for the WyZER technique 226 for this test is 3.1 dB higher than the FEC only case 224, and 5.6 dB higher than baseline H.263+ (curve 222). [0065] Figure 2(d) is a plot of PSNR verses error percentage for the video sequence "Stefan" encoded with an H.263+ encoder 232, FEC 234, and a DSC technique 236. Figure 2(e) is a plot of PSNR verses error percentage for the video sequence "Football" encoded with an H.263+ encoder 242, FEC 244, and a DSC technique 246. As WO 2005/120079 16 PCT/l S2005/019157 illustrated in Figures 2(d) and 2(e), for the typical performance of the three methods of encoding, the methods of the present invention consistently outperform the other two. Though results are presented only for SIF (352x240) sequences, similar behaviors are observed for video sequences of other spatial sizes, such as, QCTF, CIF. |0066] Experiments have also been conducted to study the effect of burst length on performance. Figure 2(f) is a plot of PSNR verses mean burst length for the "Football" video sequence (176x144, 15fps, 30 frames) encoded with FEC 252 and DSC 254. As illustrated in Figure 2(f), when mean burst length is greater than 0.6, the techniques according to the present invention 254 outperform the FEC-only 252 case, but when the burst length is smaller than 0.6, it is inferior to the FEC only case. [0067] Thus, simulation results using techniques described above show significant gains in performance over conventional error protection schemes such as Forward Error Correction codes under reasonable latency constraints. [0068] The DSC techniques described may operate with a number of wireless devices and wireless standards. Examples of wireless devices that can implement the DSC techniques described include cellular telephones, wireless communication enabled personal computers, and personal digital assistants (PDA), and other wireless devices. Example of wireless standards that the DSC techniques described can be operated on include standards referred to as Global System for Mobile Communication (GSM), General Packet Radio Service (GPRS), Enhanced Data GSM Environment (EDGE), TIA/EIA-95-B (IS-95), TIA/E1A-98-C (IS-98), CDMA2000, Wideband CDMA (WCDMA), and others. [0069] Figure 3 is a block diagram illustrating portions of a wireless communication system that can implement the DSC techniques. As shown in Figure 1, the wireless communication system includes a first wireless communication device (WCD) 302, also referred to as a mobile station (MS). The wireless communication system includes a second wireless device 304, that can be part of a wireless infrastructure, another wireless device, or a landline device that is connected to a wireless infrastructure. [0070] The first wireless device 302 includes a WyZER encoder 306, a WyZER decoder 308, a transmitter 310 and a receiver 312. The second wireless device 304 includes a WyZER encoder 320, a WyZER decoder 322, a transmitter 324 and a receiver 324. The second wireless device 304 may receive an input signal X,, for example, a video signal. The input signal Xx is connected to the DSC encoder 320. WO 2(105/120079 17 PCT/US2W5/019157 Included within the DSC encoder 320 there is a predictive encoder 330 and a WyZER encoder 332. The input signal is encoded in accordance with the DSC techniques described, and a Main stream and WyZER stream are output from the predictive encoder 320 and the WyZER encoder 332 to the transmitter 324. [0071] The Main and WyZER streams are transmitted over a wireless communication channel (multiple logical channels can be used to transmit the two streams) to the receiver 312 in the first wireless device 302. The Main stream is output from the receiver 312 to a predictive decoder 334 within the DSC decoder 308 in the first wireless device 302. The WyZER stream is output from the receiver 312 to a WyZER decoder 336. The output X} of the predictive decoder 334 is also input to the WyZER decoder 336. In accordance with the DSC techniques described the predictive and WyZER decoders cooperate to produce and output a reconstruction signal X, of the input signal X1. [0072] In a similar manner to that just described, the first wireless device may receive an input signal X2 that is encoded by a predictive encoder 340 and a WyZER encoder 342 included within the DSC encoder 306 in the first wireless device 302. Main and WyZER streams are output from the DSC encoder 306 to the transmitter 310. The receiver 326 in the second wireless device 304 receives the bit-streams and outputs the predictive bit-stream to a predictive decoder 344 and the WyZER stream to a WyZER decoder 346 in the DSC decoder 322 in the second wireless device 304. The output X2 of the predictive decoder 344 is also input to the WyZER decoder 346. In accordance with the DSC techniques described the predictive and WyZER decoders cooperate to produce and output a reconstruction signal X2 of the input signal X2. [0073] In another embodiment, a wireless device may include only a DSC encoder or only a DSC decoder. For example, if a device is configured to broadcast, or transmit, encoded data but not to receive encoded data, then it may have a DSC encoder to encode the broadcast data but it may not have a DSC decoder because it does not receive encoded data. Likewise, a device configured to receive and decode a signal, but not to transmit an encoded signal, may include a decoder but not an encoder. An example of this type of system would be a broadcast system, such as a video broadcast system, where the broadcaster transmits encoded data to at least one user, and typically to many users, but the users do not transmit encoded data back to the broadcaster. In WO 2IKI5/120079 18 PCT/1!S20»5/»19I57 this example, the broadcaster wireless device would include a DSC encoder to encode the broadcast data, and the user wireless device would include a DSC decoder to decode the received encoded broadcast. [0074] Figure 4 is a flow diagram illustrating an embodiment for determining intra- frame bit allocation to a WyZER encoder. Flow begins in block 402 where an entire frame of data, that is partitioned into data blocks, is examined. Flow continues to block 404 where it is determined which of the data blocks should be protected. For example, the data blocks may be "ranked" based on the deleterious effect that the loss of the individual data block would have on a reconstruction of the frame of data at a receiver if the particular block is lost or corrupted. The blocks whose loss would be more harmful to the reconstruction having a higher rank. [0075] Flow continues to block 406 where an amount of data to be allocated to a WyZER encoder is determined. In general, it is desirable to allocate a minimum amount of data allocated to the secondary channel because the amount of data allocated has to be transmitted over the secondary channel thereby consuming additional system resources. Thus, more data will be allocated to, and transmitted over, the secondary channel depending on how desirable it is to protect a particular data block. For example, if a data block is "ranked" high, more data may be allocated to, and transmitted over, the secondary channel for that block than to a block that is "ranked" lower. [0076] Figure 5 is a flow chart illustrating an embodiment of determining a maximum number of partitions of a source codebook that will be used by a WyZER encoder. Flow begins in block 502 where a correlation noise estimate is determined. For example, the correlation noise of a transform coefficient may be determined for the case where a predictive bit-stream, representing a block of data, is successfully transmitted from a predictive encoder to a predictive decoder. Even though the block is successfully transmitted, the decoded block may have errors due to errors in previous blocks. Flow continues to block 504 where the number of partitions of a source codebook, used to represent the data block, is determined. The number of partitions of the source codebook is also referred to as coset. [0077] In one embodiment, the number of cosets may be communicated to a WyZER decoder over a wireless secondary channel communication in a header of the WyZER stream. WO 2005/120079 19 PCT/US2005/0I9157 [0078] Those of skill in the art will understand that information and signals may be represented using any of a variety of different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips that may be referenced tlroughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof. [0079] Those of skill will further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention. [0080] The various illustrative logical blocks, modules, and circuits described in connection with the embodiments disclosed herein may be implemented or performed with a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. [0081] The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, WO 2005/1201179 20 PCT/US2W5/IH9157 EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In the alternative, the processor and the storage medium may reside as discrete components in a user terminal. [0082] The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein. WO 2005/120079 21 PCT/US2005/019157 CLAIMS What is claimed is: 1. A wireless communication device comprising: a predictive encoder configured to receive an input signal, encode the signal according to a predictive encoding standard, and to output a predictive encoded bit-stream; a secondary channel encoder configured to receive the input signal, encode the signal according to a secondary channel encoding standard, and to output a secondary channel encoded bit-stream. 2. A wireless communication device as defined in Claim 1, wherein the predictive encoded bit-stream is transmitted over a main communication channel. 3. A wireless communication device as defined in Claim 1, wherein the secondary channel encoded bit-stream is transmitted over a secondary channel communication channel. 4. A wireless communication device as defined in Claim 1, wherein the predictive encoding standard is a standard compatible with MPEG-4 standard. 5. A wireless communication device as defined in Claim 1, wherein the predictive encoding standard is a standard compatible with H.26x standard. 6. A wireless communication device as defined in Claim 1, wherein the secondary channel encoding standard is a standard compatible with Wyner Ziv techniques. 7. A wireless communication device as defined in Claim 1, wherein the input signal is a video signal. 8. A wireless communication device as defined in Claim 1, wherein the input signal comprises intra-coded blocks of data and inter-coded blocks of data. WO 2005/120079 22 PCT/US20O5/O19I57 9. A wireless communication device as defined in Claim 8, wherein Wyner Ziv techniques are applied to the inter-coded data blocks. 10. A wireless communication device as defined in Claim 1, wherein the secondary channel encoder will encode the input signal using a predetermined number of cosets. 11. A wireless communication device as defined in Claim 10, wherein the number of predetermined cosets is included within a header of the secondary channel encoded bit-stream. 12. A wireless communication device comprising: a predictive decoder configured to receive a bit-stream representation of an original signal, decode the bit stream according to a predictive decoding standard, and to output a reconstructed signal; a secondary channel decoder configured to receive a bit-stream representation of the original signal and the reconstructed signal, decode the bit-stream according to a secondary channel decoding standard, and to combine the decoded secondary channel bit stream and the reconstructed signal to generate and output a final reconstruction of the original signal. 13. A wireless communication device as defined in Claim 12, wherein the bit-stream decoded by the predictive decoder is received over a main communication channel. 14. A wireless communication device as defined in Claim 12, wherein the bit-stream decoded by the secondary channel decoder is received over a secondary channel communication channel. 15. A wireless communication device as defined in Claim 12, wherein the predictive decoding standard is a standard compatible with MPEG-4 standard. 16. A wireless communication device as defined in Claim 12, wherein the predictive decoding standard is a standard compatible with H.26x standard. WO 2005/120079 23 PCT/US2005/019157 17. A wireless communication device as defined in Claim 12, wherein the secondary channel decoding standard is a standard compatible with Wyner Ziv techniques. 18. A wireless communication device as defined in Claim 12, wherein the original signal is a video signal. 19. A wireless communication device as defined in Claim 12, wherein the original signal comprises intra-coded blocks of data and inter-coded blocks of data. 20. A wireless communication device as defined in Claim 19, wherein Wyner-Ziv techniques are applied to the inter-coded data blocks. 21. A wireless communication device as defined in Claim 12, wherein the secondary channel decoder will decode the secondary channel bit-stream using a predetermined number of cosets. 22. A wireless communication device as defined in Claim 21, wherein the number of predetermined cosets is included within a header of the secondary channel encoded bit-stream. 23. A wireless communication device comprising: an encoder comprising: a predictive encoder configured to receive an input signal, encode the signal according to a predictive encoding standard, and to output a predictive encoded bit-stream; a secondary channel encoder configured to receive the input signal, encode the signal according to a secondary channel encoding standard, and to output a secondary channel encoded bit-stream, a decoder comprising: a predictive decoder configured to receive a bit-stream, decode the bit stream according to the predictive decoding standard, and to output a reconstructed signal; and WO 2005/120079 24 PCT/US2005/019157 a secondary channel decoder configured to receive a bit-stream and the reconstructed signal, decode the bit-stream according to a secondary channel decoding standard, and to combine the decoded bit-stream and the reconstructed signal to generate and output a final reconstruction of an original signal. 24. A wireless communication device as defined in Claim 23, wherein the predictive decoding standard is a standard compatible with MPEG-4 standard. 25 A wireless communication device as defined in Claim 23, wherein the predictive decoding standard is a standard compatible with H.26x standard. 26. A wireless communication device as defined in Claim 23, wherein the secondary channel decoding standard is a standard compatible with Wyner Ziv techniques. 27. A wireless communication device as defined in Claim 23, wherein the input signal is a video signal. 28. A wireless communication device as defined in Claim 23, wherein the wireless communication device is a cell phone. 29. A wireless communication device as defined in Claim 23, wherein the wireless device is part of a wireless infrastructure. 30. A wireless communication system comprising: a first wireless device comprising: a predictive encoder configured to receive an input signal, encode the signal according to a predictive encoding standard, and to output a predictive encoded bit-stream representation of the input signal; a secondary channel encoder configured to receive the input signal, encode the signal according to a secondary channel encoding standard, and to output a secondary channel encoded bit-stream representation of the input signal, a second wireless device comprising: L WO 2005/120079 PCT/US20O5/019157 25 a predictive decoder configured to receive the bit-stream output by the predictive encoder, decode the bit-stream according to the predictive decoding standard, and to output a reconstruction of the input signal signal; and a secondary channel decoder configured to receive the bit-stream output by the secondary channel encoder, decode the bit-stream according to a secondary channel decoding standard, and to combine the decoded secondary channel bit stream and the reconstructed signal to generate and output a final reconstruction of the input signal. 31. A wireless communication system as defined in Claim 30, wherein the predictive decoding standard is a standard compatible with MPEG-4 standard. 32 A wireless communication system as defined in Claim 30, wherein the predictive decoding standard is a standard compatible with H.26x standard. 33. A wireless communication system as defined in Claim 30, wherein the secondary channel decoding standard is a standard compatible with Wyner Ziv techniques. 34. A wireless communication system as defined in Claim 30, wherein the first wireless communication device is a cell phone. 35. A wireless communication system as defined in Claim 30, wherein the first wireless device is part of a wireless communication infrastructure device. 36. A wireless communication system as defined in Claim 30, wherein the second wireless communication device is a cell phone. 37. A wireless communication system as defined in Claim 30, wherein the second wireless device is part of a wireless communication infrastructure device. 38. A method of encoding data, the method comprising: receiving an input signal; WO 2005/120079 26 PCT/US2005/OI9J57 encoding the signal according to a predictive encoding standard, and outputting a predictive encoded bit-stream; and encoding the signal according to a secondary channel encoding standard, and to outputting a secondary channel encoded bit-stream. 39. A method of encoding as defined in Claim 38, wherein the predictive encoding standard is a standard compatible with MPEG-4 standard. 40. A method of encoding as defined in Claim 38, wherein the predictive encoding standard is a standard compatible with H.26x standard. 41. A method of encoding as defined in Claim 38, wherein the secondary channel encoding standard is a standard compatible with Wyner Ziv techniques. 42. A method of encoding as defined in Claim 38, wherein the input signal is a video signal. 43. A method of encoding as defined in Claim 38, wherein the input signal comprises intra-coded blocks of data and inter-coded blocks of data. 44. A method of encoding as defined in Claim 43, wherein Wyner Ziv techniques are applied to the inter-coded data blocks. 45. A method of encoding as defined in Claim 38, wherein the secondary channel encoder will encode the input signal using a predetermined number of cosets. 46. A method of encoding as defined in Claim 45, wherein the number of predetermined cosets is included within a header of the secondary channel encoded bit-stream. 47. A method of decoding data, the method comprising: receiving a predictive encoded bit-stream representation of an original signal; WO 2005/120079 PCT/US2005/019157 27 decoding the predictive encoded bit stream according to a predictive decoding standard, and to output a reconstructed signal; receiving a secondary channel encoded bit-stream representation of the original signal and the reconstructed signal; and decoding the secondary channel encoded bit-stream according to a secondary channel encoding standard, and combining the decoded secondary channel bit stream and the reconstructed signal to generate and output a final reconstruction of the original signal. 48. A method of decoding as defined in Claim 47, wherein the predictive decoding standard is a standard compatible with MPEG-4 standard. 49. A method of decoding as defined in Claim 47, wherein the predictive decoding standard is a standard compatible with H.26x standard. 50. A method of decoding as defined in Claim 47, wherein the secondary channel decoding standard is a standard compatible with Wyner Ziv techniques. 51. A method of decoding as defined in Claim 47, wherein the original signal is a video signal. 52. A method of decoding as defined in Claim 47, wherein the original signal comprises intra-coded blocks of data and inter-coded blocks of data. 53. A method of decoding as defined in Claim 52, wherein Wyner Ziv techniques are applied to the inter-coded data blocks. 54. A method of decoding as defined in Claim 47, wherein the secondary channel decoder will decode the secondary channel bit-stream using a predetermined number of cosets. WO 2005/120079 28 PCT/US2005/U19157 55. A method of decoding as defined in Claim 54, wherein the number of predetermined cosets is included within a header of the secondary channel encoded bit-stream. 56. A method of determining intra-frame bit allocation to a secondary channel encoder, the method comprising: examining an entire frame of data; determining which blocks of data within the frame of data need protection; and determining an amount of data within the frame to be allocated to, and transmitted by, a secondary channel based upon the blocks of data that need protection. 57. A method of determining partitions of a source codebook, the method comprising: determining a correlation noise estimate; and determining a number of partitions of a source cookbook based upon the estimate of correlation noise estimate. 58. A computer readable media embodying a method of encoding data, the method comprising: receiving an input signal; encoding the signal according to a predictive encoding standard, and outputting a predictive encoded bit-stream; and encoding the signal according to a secondary channel encoding standard, and to outputting a secondary channel encoded bit-stream. 59. A computer readable media embodying a method of decoding data, the method comprising: receiving a first bit-stream representation of an original signal; decoding the first bit stream according to a predictive decoding standard, and to output a reconstructed signal; receiving a second bit-stream representation of the original signal and the reconstructed signal; and WO 2005/120079 29 PCT/US2005/019157 decoding the second bit-stream according to a secondary channel decoding standard, and combining the secondary channel decoded bit stream and the reconstructed signal to generate and output a final reconstruction of the original Signal. 60. A method as defined in Claim 59, wherein the first bit-stream is encoded according to a predictive encoding standard. 61. A method as defined in Claim 59, wherein the second bit-stream is encoded according to a secondary channel encoding standard. 62. A wireless communication device comprising: means for receiving an input signal; means for encoding the signal according to a predictive encoding standard, and outputting a predictive encoded bit-stream; and means for encoding the signal according to a secondary channel encoding standard, and to outputting a secondary channel encoded bit-stream. 63. A wireless communication device comprising: means for receiving a first bit-stream representation of an original signal; means for decoding the bit stream according to a predictive decoding standard, and to output a reconstructed signal; means for receiving a second bit-stream representation of the original signal and the reconstructed signal; and means for decoding the second bit-stream according to a secondary channel encoding standard, and combining the secondary channel decoded bit stream and the reconstructed signal to generate and output a final reconstruction of the original signal. Dated this 21st day of December, 2006 ABSTRACT "METHOD, APPARATUS AND SYSTEM FOR ENHANCING ROBUSTNESS OF PREDICTIVE VIDEO CODECS USING A SIDE-CHANNEL BASED ON DISTRIBUTED SOURCE CODING TECHNIQUES" A method, apparatus and system for providing distributed source coding techniques that improve data coding performance, such as video data coding, when channel errors or losses occur. Errors in the reconstruction of the data is eliminated or reduced by sending extra information. Correlation between a predicted sequence and an original sequence can be used to design codebooks and find the co-sets required to represent the original image. This information may be sent over another channel, or a secondary channel.

Documents:

1605-mumnp-2006-abstract(15-01-2009).pdf

1605-MUMNP-2006-ABSTRACT(15-1-2009).pdf

1605-mumnp-2006-abstract.doc

1605-mumnp-2006-abstract.pdf

1605-mumnp-2006-cancelled pages(15-01-2009).pdf

1605-MUMNP-2006-CANCELLED PAGES(15-1-2009).pdf

1605-MUMNP-2006-CLAIMS(15-1-2009).pdf

1605-mumnp-2006-claims(granted)-(15-01-2009).pdf

1605-mumnp-2006-claims.doc

1605-mumnp-2006-claims.pdf

1605-MUMNP-2006-COPY OF US ASSIGNMENT(15-1-2009).pdf

1605-MUMNP-2006-CORRESPONDENCE(14-07-2008).pdf

1605-mumnp-2006-correspondence(15-06-2009).pdf

1605-MUMNP-2006-CORRESPONDENCE(15-1-2009).pdf

1605-mumnp-2006-correspondence(ipo)-(23-01-2009).pdf

1605-mumnp-2006-correspondence-others.pdf

1605-mumnp-2006-correspondence-received.pdf

1605-mumnp-2006-description (complete).pdf

1605-MUMNP-2006-DESCRIPTION(COMPLETE)-(15-1-2009).pdf

1605-mumnp-2006-drawing(15-01-2009).pdf

1605-MUMNP-2006-DRAWING(15-1-2009).pdf

1605-mumnp-2006-drawings.pdf

1605-MUMNP-2006-FORM 1(22-12-2006).pdf

1605-mumnp-2006-form 1(22-12-2009).pdf

1605-mumnp-2006-form 18(22-12-2006).pdf

1605-mumnp-2006-form 2(15-1-2009).pdf

1605-mumnp-2006-form 2(granted)-(15-01-2009).pdf

1605-MUMNP-2006-FORM 2(TITLE PAGE)-(15-1-2009).pdf

1605-mumnp-2006-form 26(15-01-2009).pdf

1605-MUMNP-2006-FORM 26(15-1-2009).pdf

1605-MUMNP-2006-FORM 3(14-07-2008).pdf

1605-mumnp-2006-form 3(15-01-2009).pdf

1605-MUMNP-2006-FORM 3(15-1-2009).pdf

1605-mumnp-2006-form 3(21-12-2006).pdf

1605-MUMNP-2006-FORM 5(15-1-2009).pdf

1605-mumnp-2006-form 5(22-12-2006).pdf

1605-mumnp-2006-form-1.pdf

1605-mumnp-2006-form-18.pdf

1605-mumnp-2006-form-2.pdf

1605-mumnp-2006-form-26.pdf

1605-mumnp-2006-form-3.pdf

1605-mumnp-2006-form-5.pdf

1605-mumnp-2006-form-pct-ib-304.pdf

1605-mumnp-2006-form-pct-ib-311.pdf

1605-mumnp-2006-form-pct-ib-332.pdf

1605-mumnp-2006-form-pct-isa-210(22-12-2006).pdf

1605-mumnp-2006-form-pct-isa-220.pdf

1605-mumnp-2006-other documents(04-02-2009).pdf

1605-mumnp-2006-pct search report.pdf

1605-mumnp-2006-petition under rule 137(15-01-2009).pdf

1605-MUMNP-2006-PETITION UNDER RULE 137(15-1-2009).pdf

abstract1.jpg