在這篇論文中，我們利用純量量化(scalar quantizers)來做多重描述編碼(multiple description coding)，應用於比對搜尋(matching pursuit)之視訊傳輸。目前的視訊標準(如H.263,MPEG-2)，都是使用DCT來做視訊的壓縮。DCT的方法雖然有效率，但是在低位元率(low bit rates)的情況下，它會產生區塊效應(blocking artifacts)，而使用比對搜尋之視訊壓縮因為使用了過完整(over-complete)的基底，因此可避免區塊效應的產生。
我們針對殘留影像(residual image)來做比對搜尋之編碼，然後將比對搜尋後所得到的參數(atom) ，利用純量量化而得到兩個同等重要的描述(balanced descriptions)，然後將這兩個描述分別在不同的通道上傳送。為了要得到最佳的編碼器與解碼器，我們也會說明必須要滿足的一些條件。如果所收到的兩個描述都是正確的話，接收端就能解得較高品質的視訊。但是由於所傳送的描述可能會受到通道的影響而使得產生錯誤，因此若只收到其中一個描述而另一個描述沒有收到，我們可以利用所收到那一個描述去估計原本所傳送的信號，於是接收端就會解得較低但可接受的視訊品質。

實驗的結果顯示，在各種不同情況的通道下，與H.263和比對搜尋使用錯誤更正碼(forward error correction)以及參數共享(shared-atom)等方法做比較，我們所提出的方法會得到較好的結果，除了在低錯誤率(low loss rates)的情況下，比對搜尋使用參數共享的方法會得到較好的結果。

In this thesis, we proposed a method of the multiple description coding (MDC) using scalar quantizers for the matching pursuit residual coding over a diversity network with two channels. Current video compression standards, such as H.263 and MPEG-2, use block-based discrete cosine transform (DCT) to encode prediction errors. DCT based video compression scheme is efficient but it introduces undesirable blocking artifacts at low bit rates. Using Matching pursuits (MP) for residual coding does not suffer from blocking artifacts because a over-complete basis set is used to match the residual.
The parameters obtained after performing matching pursuit, called atoms, are used to encoded into two balanced descriptions using scalar quantizers. Then, the two descriptions are respectively transmitted over two separate channels. If both descriptions are received, a high-quality reconstruction can be decoded, however, if either description is lost, a low-quality, but acceptable, reconstruction can be decoded from the remaining channel. Simulation results show that our proposed method outperforms single description coding (SDC) of H.263 with FEC and MP MDC using share atoms over two-state Markov channel models.

[1] R. Neff and A. Zakhor, “Very Low Bit-rate Video Coding Based on Matching Pursuit,” IEEE Transactions on Curcuit and Systems for Video Technology, Vol. 7, No.1, pp.158-171, Feb 1997.
[2] X. Tang and A. Zakhor, “Matching Pursuit Multiple Description Coding for Wireless Video,” IEEE Transactions on Image Processing, Vol. 1, pp. 926-929, 2001.
[3] S. Mallant and Z. Zhang, “Matching Pursuit with Time-Frequency Dictionary,” IEEE Transactions on Signal Processing, Vol. 41, pp. 3397-3415, Dec 1993.
[4] B. Torresani, “Wavelet Associated with Representations of the Affine Weyl- Heisenberg Groop,” J. Math. Physics, Vol. 32, pp. 1273-1279, May 1991.
[5] Telecommunication Standardization Sector of ITU, “ITU-T Recommendation H.263,” Feb 1998.
[6] V. A. Vaishampayan, “Design of Multiple Description Scalar Quantizers,” IEEE Transactions on Information Theory, Vol. 39, No. 3, pp. 821-834, May 1993.
[7] Vivek K. Goyal, “Multiple Description Coding: Compression Meets the Network, ” IEEE Signal Processing Magazine, Vol.18, pp. 74-93, Sept. 2001.
[8] V. A. Vaishampayan; and Sam John, “Balanced Iterframe Multiple Description Video Compression,” IEEE Transactions on Image Processing, Vol. 3, pp. 812-816, 1999.
[9] Chang-Su Kim; and Sang-Uk Lee, “Multiple Description Coding of Motion Fields for Robust Video Transmission,” IEEE Transactions on Circuit and Systems for Video Technology, Vol. 11, No. 9, pp. 999-1010, March 2001.
[10] Yao Wang; Michael T. Orchard; V. A. Vaishampayan; and Amy R. Reibman, “Multiple Description Coding Using Pairwise Correlating Transforms,” IEEE Transactions on Image Processing, Vol. 10, No. 3, pp. 821-834, March 2001.
[11] M. Yajnik, S. Moon, and J. Kurose, “Measurement and Modeling of the Temporal Dependence in Packet Loss, ” INFOCOM, pp. 78-98, 1999.
[12] Apostolopoulos, J.; Wai-tian Tan; Wee, S.; and Wornell, G.W., “Modeling Path Diversity for Multiple Description Video Communication,” IEEE ICASSP, Vol. 3, pp. 2161-2164, 2002.
[13] E. N. Gilbert, “Capacity of a burst-noise channel, ” Bell Syst. Tech. J., Vol. 39, pp. 1253-1265, Sept 1960.