Yueh-Teng Hsu

Software Viterbi Decoder with SSE4 Parallel Processing Instructions for Software DVB-T Receiver

In this paper, we discuss the procedures how to make Viterbi decoder faster. The implementation in Intel CPU with SSE4 parallel processing instruction sets and some other methods achieves the decoding speed 47.05 Mbps (0.64 Mbps originally). The DVB-T mode used in Taiwan needs 13.27 Mbps to achieve real-time reception, so our implementation of software Viterbi decoder takes only 28% CPU loading.

GPU-Accelerated Single Image Depth Estimation with Color-Filtered Aperture

There are two major ways to implement depth estimation, multiple image depth estimation and single image depth estimation, respectively. The former has a high hardware cost because it uses multiple cameras but it has a simple software algorithm. Conversely, the latter has a low hardware cost but the software algorithm is complex. One of the recent trends in this field is to make a system compact, or even portable, and to simplify the optical elements to be attached to the conventional camera. In this paper, we present an implementation of depth estimation with a single image using a graphics processing unit (GPU) in a desktop PC, and achieve real-time application via our evolutional algorithm and parallel processing technique, employing a compute shader. The methods greatly accelerate the compute-intensive implementation of depth estimation with a single view image from 0.003 frames per second (fps) (implemented in MATLAB) to 53 fps, which is almost twice the real-time standard of 30 fps. In the previous literature, to the best of our knowledge, no paper discusses the optimization of depth estimation using a single image, and the frame rate of our final result is better than that of previous studies using multiple images, whose frame rate is about 20fps.

Reed-Solomon decoder optimization for PC-based DVB-T software radio receiver

We use look up tables, SIMD parallel processing instruction sets and loop expansion to implement RS decoder in the Intel CPU platform. Our implementation of Reed-Solomon decoder takes only 1.89% CPU loading.

A distributed quasi-orthogonal space time block code for cooperative communication with information exchange errors for decode-and-forward relay networks

In this paper, we design the full-rate full-diversity quasi-orthogonal space time block code (QOSTBC) matrix for decode-and-forward (DAF) relay networks with four independent single-antenna source nodes, a single-antenna destination node and a single receive four-trans mit-antenna relay node. In this scheme, we consider the channel states between the four users to select the elements of the DQOSTBC matrix to be decode and forward (DAF) or amplify and forward (AAF) types. Thus the proposed scheme is a DQOSTBC matrix with embedded adaptive DAF/AAF elements, based on the source-to-source channel coefficient is above or below a threshold. By the simulation results, the proposed DQOSTBC with adaptive DAF/AAF matrix elements is approximately 7dB better than the traditional DQOSTBC (all matrix elements are fixed to DAF type, by Li et. al.) at 10-4 bit error rate (BER) because the traditional DQOSTBC loses full diversity due to information exchange errors.

Turbo ICI cancellation and LDPC decoding for OFDM systems

We apply LDPC codes for turbo inter-carrier interference (ICI) cancellation (channel detecting) and decoding in OFDM systems due to mobility of channels. The simulation result shows that the proposed turbo ICI cancellation and sum-product algorithm decoding of LDPC coded OFDM systems can achieve the better error rate performance than turbo ICI cancellation and BCJR decoding of convolutional coded OFDM systems when the codeword length is 816 bits.

Iterative channel decoding for PC-based software radio DVB-T receiver

We propose a hard iterative channel decoding for a PC-based DVB-T software receiver testbed. By Reed-Solomon (RS) decoder, we make the error symbols have maximum Euclidean distance and thus no chosen in Viterbi decoding.

Improved Channel Estimation for PC-Based Software Radio DVB-T Receiver

In the paper, we propose an improved discrete Fourier transform (DFT)-based channel estimation for a PC-based software DVB-T receiver testbed. The proposed channel estimation algorithm includes an adaptive low pass filter whose passband get wider when higher moving speed. The image quality of the proposed scheme is significantly better than that of the traditional scheme in realistic mobile channels (not simulated mobile channels).

Panoramic endoscope based on convex parabolic mirrors

Abstract. We propose a parabolic mirror-based catadioptric imager for the panoramic imaging endoscope. The main difference of the structure is the use of a parabolic mirror in endoscope design. The manufacturing cost and/or size of the proposed panoramic design is lower than the previous designs. The designs by Katman and Liu use customized lens and thus cost more than the proposed in mass production. The design by Sahli uses more mirrors/lenses than the proposed. The size of the previous design by Yu, 20 mm in diameter, is larger than the standard-angle Olympus video colonoscope (9.5 mm in diameter). Our proposed design is 5.5 mm in diameter, so it is feasible to be integrated into an endoscope while the previous design by Yu is not feasible to be integrated into an endoscope.

Improve spectrum efficiency based on variable-power AMC and ARQ for OFDM systems

To improve spectrum efficiency (SE), the adaptive modulation and coding (AMC) and automatic repeat request (ARQ) scheme have been combined for OFDM systems. In this paper, we add variable power subject to power constraint in each AMC mode. We use KKT optimization algorithm to get the optimal transmit power and AMC mode boundaries. The numerical results show that the average SE is increased by about 0.3 bits/sec/Hz for OFDM systems with Nakagami fading with parameter m=2 when SNR is around 15dB and the retransmission is once.

Simpler and More Accurate Throughput Analysis of CDMA Pure ALOHA Networks with Variable Length Traffic

We propose M/M/inf model for the DS CDMA/unslotted ALOHA system with variable number of fixed-length packets. The analyzed throughput of M/M/inf queuing model is closer to the simulation throughput than the analyzed throughput of the previous queuing model is. Moreover, the proposed analysis scheme is simpler.