Jay Bao

A New Approach For Memory Efficient ATV Decoding

In this paper we propose an efficient memory saving architecture for an ATV decoder. In this new architecture, the motion-compensation is performed in the frequency domain. This makes it possible for anchor frames to be saved as compressed bits, resulting in significant memory savings with the addition of several extra devices in the decoder. The increase in complexity is small compared to the reduction in memory requirements. The simulation results with the new decoder architecture have shown that the reconstructed image quality is close to that of the conventional decoder.

HDTV down-conversion decoder

In the near future we assume that the advanced television (ATV) broadcasts will be a mixture of HDTV and SDTV. We propose a HDTV down-conversion decoder which can decode the Grand Alliance HDTV bitstreams and display them on SDTV or NTSC monitors. The key technique is a frequency synthesis which decompresses the HDTV bitstreams without full decoding or spatial sampling processes, instead of synthesizing a set of new 8/spl times/8 coefficients from the coefficients of four original 8/spl times/8 blocks. The intent is to provide high quality reconstructed video with lower spatial resolution and less complexity, and thus less computing and memory requirements.

A calculation-efficient algorithm for decision feedback equalizers

We describe a method for decision feedback equalizers used in high speed digital receiver demodulators such as terrestrial digital television (DTV) receivers. The new algorithm uses the sparse channel characteristics of the DTV transmission. A variation of the decision feedback equalizer, namely the partial feedback equalizer, is also introduced. The sparseness of the channel is combined with the partial feedback equalizer to develop the new algorithm. Similarity with subspace based filtering is noted, which results in robust performance in the presence of noise. The new algorithm has a superior convergence speed and lower steady state mean squared error, as compared to conventional approaches.

Frequency domain down-conversion of HDTV using adaptive motion compensation

The paper investigates the possible means of down converting an incoming HDTV signal into one which is suitable for display onto an NTSC monitor. We present a novel frequency synthesis algorithm which constructs a set of 8/spl times/8 DCT coefficients from four original 8/spl times/8 DCT blocks. Also, a down-conversion decoder employing a unique adaptive motion compensation algorithm is presented. Using this fixed motion compensation algorithm, the proposed down-conversion technique is compared to the conventional method. Both yield acceptable image sequences with different trade-offs in complexity and quality.

A New Timing Recovery Method For Dtv Receivers

We report a new unsynchronized timing recovery method designed for receiving digital TV signals modulated in 256QAM, QPSK and VSB format. The new method includes a simplified and efficient interpolator design, and fast converging PLL. It can be used for fixed rate as well as variable rate applications. The algorithm was developed using SPW, and implemented using an FPGA-based prototype board using real data.

Optimization of transceiver systems for broadband fixed wireless applications

Systems designed for high data rate broadband wireless multimedia services must combat fading and multipath impairments. In this paper, the effects of these impairments on transceiver signal processing system design are analyzed, and the system architecture of spatial transmit diversity combined with adaptive modulation is presented. The performance of such a system is compared with conventional single carrier fixed modulation and OFDM systems.