Shaw Hwa Hwang

An Improvement to Tree-Structured Vector Quantization

This study proposes an improvement to tree-structured vector quantization (TSVQ). The selected codeword of TSVQ is employed as the reference codeword. The dynamic triangular inequality elimination (DTIE) is employed to improve the performance of TSVQ. In the DTIE-TSVQ approach, the selected codeword and SNR are identical to that of full-search vector quantization (FSVQ), and the computational saving is better than that of TIE-VQ. The experimental results confirm that DTIE-TSVQ outperforms TIE-VQ, TSVQ.

Symmetric NAT Traversal Method for Session Initial Protocol (SIP)

Although the integration of SIP-based systems with a network address translation (NAT) environment has been investigated extensively, SIP-based system operation in symmetric model NAT remains relatively unexplored. This paper studies the application of symmetric NAT traversal techniques to SIP-based systems. This study focuses on following the SIP process, a non-additive service server, and employs standard SIP commands such as “REGISTER,” “INVITE,” and “200 OK” to predict and deliver the IP addresses and port numbers of the local NAT. This study also implements RTP streaming in the client-to-client (C2C) mode. The symmetric NAT traversal method for the SIP increases the direct peer-to-peer connection rate. This approach also avoids the STUN and RTP-Relay server. Experimental results with 50 types of NAT indicate that symmetric NAT traversal performs better than the STUN solution. The RTP-Relay server bandwidth cost is likewise superior to the TURN solution. These finding have important implications for SIP-based system developers and carrier providers.

The Dynamic LMS for Line Echo Cancellation

To the best of our knowledge, this study represents the proposal using the dynamic least mean square (DLMS) algorithm to reduce the computation load of LMS. Moreover, three regions of impulse response of line echo path are also proposed to analyze the redundant coefficients. Using the DLMS method, redundant coefficients can be detected and grouped, thereby automatically reducing computation. We employed line echo cancellation (LEC) to evaluate the performance of DLMS. The pure-delay and overlong regions of impulse response of line echo path are grouped and the associated computation load is reduced. The experimental results confirm the excellent performance of DLMS achieving a 35% savings in computation. Moreover, the quality echo return loss enhancement (ERLE) of DLMS also maintains at a level nearly equal to LMS.

Performance Analysis and Improvement on SIP-Based VoIP Server

The performance improvement on SIP-based VoIP server is given in this paper. The signaling server “SIP Proxy Server” and media streaming server “Media Relay Server” are applied. Firstly, the performance of VoIP server is analyzed and evaluated in detail. Then the effective method is applied to improve the performance of VoIP server. In the analysis results of SIP Proxy Server, the CPU time for the registration and call-setup sessions need 3.241 ms and 7.985 ms respectively. Moreover, in the Media Relay Server system, when the packet size from 1 to 32 bytes, the maximum throughput of packet is about 82,000 and 16,000 per second for UDP and TCP respectively. The large the packet, the less throughput is tenable. There are three methods applied to improve the performance. The capacity by 31 times is achieved. The capacity with 3,524 calls for each Media Relay Server is achieved. The improved method proposed in this paper is reasonable.