JMM Abstracts 

Vol.8 No.1 April 15, 2012

Research Articles for the Theme Issues

Study and Development of the Transmission Method for Large Multimedia File Size Using MMS Technology (001-024)
       
Andik Setyono, Md J. Alam, and C. Eswaran
Mobile technology has grown rapidly in recent years due to the availability of sophisticated mobile phones in the market. Many features offered in the mobile phones encourage people to develop mobile-phone-based systems. Internet applications also have been developed to transmit data by mobile phone in a mobile environment. In this paper, we study and develop transmission methods for large multimedia files using MMS technology. We implement file compression, splitting, masking and cropping techniques to improve the ability of MMS technology in multimedia message delivery so that it can transmit large size data files which are not possible with the existing technologies. We hope that this research can provide a valuable contribution to the development of MMS technology. The experimental results demonstrate that the proposed techniques perform better than the conventional MMS technologies in preserving the quality of data. The modified MMS technology can be used to develop practical client-server applications.

Live Sharing with Multimodal Modes in Mobile Network (025-033)
       
Xiao Zeng, Kongqiao Wang, and Da Huo

Live video sharing is a newly generated and interesting service, with which users can broadcast and view the videos being recorded by mobile phones. However, mobile network usually blocks users to enjoy that service since video transmitting is still a nontrivial task with poor bandwidth. In order to make live sharing easier in mobile environment, a novel service with multimodal modes is proposed in this paper, which could save a lot of bandwidth for sharing and is more adaptive in mobile network. To save bandwidth and introduce differentiated user experience, real-time extracted key-frames, audio or hybrid information can be transmitted instead of original video stream. Both publishers and receivers can select suitable mode according to their preference or network condition. Thanks to the key frame mode of the proposed service, detailed tagging of video content and live cooperation with other SNS can be implemented. Experimental results and user study demonstrate that the proposed multimodal live sharing service is of high adaption of mobile network and introduces direct and interesting user experience.

An Energy-adaptive Multiple Paths Routing Approach for Wireless Sensor Networks (034-048)
       
Jinglun Shi, Kang Cai, Chenghong He, Gang Wei, and Zhilong Shan
A wireless sensor network (WSN) is expected to have a significant impact on military and civil applications such as target field imaging, intrusion detection, weather monitoring. Sensors are battery-powered, and hence energy-conserving communications are essential to prolong the sensor network’s lifetime. Also, given the unreliable nature of the wireless channels and the high failure rate of the individual sensors, a fault tolerant routing protocol with energy-efficiency is getting more and more attention. In this paper, we propose an energy-adaptive multiple paths routing algorithm (EMRA) for wireless sensor networks. It consists of three elements: () gradients to disseminate data over multiple paths from a source to a sink, () rules used for setting up disjoint multiple paths, and () policies for selecting multiple paths. By limiting the maximum number of the gradients, the exploratory data messages forwarded is decreased efficiently. By using the rules for setting up disjoint multiple paths, the sink node can get a sufficient number of disjoint multiple paths, which enables EMRA to recover from a routing failure quickly. Our analysis and simulation results reveal that EMRA performs better than the existing multiple paths routing algorithms in terms of the average dissipated energy and the delay to set up multiple paths.

Music Visualization Technique of Repetitive Structure Representation to Support Intuitive Estimation of Music Affinity and Lightness (049-071)
       
Tatiana Endrjukaite and Naoko Kosugi
This paper proposes a method to identify and visualize repetitive structures in a pairwise representation of music to support people to imagine their affinity for music and the lightness of music intuitively, or in other words without listening to it. Repetitive structures in this paper are fragments that a music piece contains multiple times, and all these fragments may be slightly different but are perceived as very similar. For example, a tune might have little difference in tonality and could be performed by different kinds of musical instruments. We propose an algorithm to identify repetitive structures in a tune by using a self-similarity matrix. Identified structures are visualized on two kinds of images. One is a colored cylinder of varying diameter where colors represent repetitions and the diameter represents volume changes; the other is repetitions lines image, where different pairs of repetitions are shown on the Y-axis and the duration of each repeated pair is shown on the X-axis with a color. We selected eight tunes based on music psychology to evaluate the performance of the identification and visualization technique. Finally, we found that the amount of repetitions is related to the affinity for music, but not to the lightness of music. Volumes in both high-affinity music and high-lightness music change drastically.

Back to JMM Online Front Page