KyungOh Lee

Exploiting caching for realtime multimedia systems

Several caching schemes for realtime multimedia systems have been proposed but they only concentrate on the hit ratio as the performance metric, without providing any means to utilize the disk bandwidth gained through cache hits. A high cache hit ratio alone does not capture the effectiveness of a caching scheme for such systems unless hiccup-free service is guaranteed. We evaluate three caching schemes which make it possible to provide service to additional streams with the saved disk bandwidth. We first start with NIC in which no cached intervals are preempted or replaced unless they have finished their playback. Next, we propose PSIC, which preempts the cached intervals for more valuable intervals but still guarantees hiccup-free playback. Finally, we present SPT-L scheme which has a significantly greater overhead and the best hit ratio. However STP-L suffers from frequent starvation, which verifies our claim that the high cache hit ratio does not necessary result in better performance.

An effective admission control mechanism for variable-bit-rate video streams

Abstract. For admission control in real-time multimedia systems, buffer space, disk bandwidth and network bandwidth must be considered. The CBR-based mechanisms do not use system resources effectively, since media data is usually encoded with VBR compression techniques. We propose an admission control mechanism based on a VBR data model that has a dynamic period length. In our mechanism, the period can be adaptively changed to maximize the performance, considering both disk bandwidth and buffer space. To compare the performance, extensive simulations are conducted on RR, SCAN, and GSS schemes which have the dynamic period length and the static period length.

MONETA: an embedded monitoring system for ubiquitous network environments

Accurate and efficient monitoring of dynamically changing environments is one of the most important requirements for ubiquitous network environments. To exploit these ubiquitous environments, we designed and implemented a monitoring system called MONETA that can obtain sensor data transmitted from wireless sensors to hub nodes in embedded equipment. MONETA adopts Web technology for the implementation of a simple but efficient user interface that allows an operator to visualize any of the processes, elements, or related information in a convenient graphic form. We developed MONETA using a kernel wrapper mechanism. This wrapper is a software module that enables us to monitor the state information of several embedded devices without any modification to the kernel. This wrapping method has allowed us to increase the portability of the monitoring system to other platforms and reduce the time required to write monitoring software modules.

Deciding Round Length and Striping Unit Size for Multimedia Servers

In realtime multimedia storage servers, it is very important to find the optimal values for the round length and the striping unit size since they have strong relation with the throughput. In getting the optimal values, we should not overlook any factor of multimedia servers since this may produce the wrong values far from the optimal. If we use a wrong value, then we may suffer from 2 or 3 times performance degradation. We present a new scheme which considers all possible factors in multimedia servers such as disk properties, buffer space, the characteristics of multimedia objects, and users’ trends of requesting. Since we do not use simple peak data rate but the maximum segment data rate which is much smaller than the peak data rate, we can use the resources very effectively and still guarantees no starvation and no overflow.

SLAP - A Secure but Light Authentication Protocol for RFID Based on Modular Exponentiation

The advantage of using RFID technology is gaining much attention as is seen by an increase in its deployment. However, this pervasive use of RFID tags opens up the possibility for various attacks violating user privacy. This work describes simple, inexpensive, and untraceable mutual identification protocols based on modular exponentiation for RFID tags. The proposed authentication mechanism is a low-cost but high-security for protecting consumers information in RFID tag against unwanted read-attacks. We analyzed with issues of the security and privacy of RFID systems. This paper also showed the 8-bit architecture of a low-power and low die-size implementation of modular exponentiation. The 8-bit modular exponentiation implementation has a chip area of less than 1 K gates and the encryption of 64 bits requires about 1000 clock cycles. The executing time of the proposed protocol would range from 12.36 ms to 20.6 ms, with 6-8 tags authenticated per second.

Mutual-Authentication Mechanism for RFID Systems

The biggest challenge for current RFID technology is to provide the necessary benefits while avoiding any threats to the privacy of its users. Although many solutions to this problem have been proposed, almost as soon as they have been introduced, methods have been found to circumvent system security and make the user vulnerable. We are proposing an advanced mutual-authentication protocol between a tag and the back-end database server for a RFID system to ensure system security integrity. The three main areas of security violations in RFID systems are forgery of the tags, unwanted tracking of the tags, and unauthorized access to a tag’s memory. Our proposed system protects against these three areas of security violations. Our protocol provides reader authentication to a tag, exhibits forgery resistance against a simple copy, and prevents the counterfeiting of RFID tags. Our advanced mutual-authentication protocol uses an AES algorithm as its cryptograph primitive. Since our AES algorithm has a relatively low cost, is fast, and only requires simple hardware, our proposed approach is feasible for use in RFID systems. In addition, the relatively low computational cost of our proposed algorithm compared to those currently used to implement similar levels of system security makes our proposed system especially suitable for RFID systems that have a large number of tags.

An energy efficient routing protocol using transmission range and direction for sensor networks

Sensors in sensor networks are operated by their embedded batteries and they can not work any more if the batteries run out. The data collected by sensors should be transferred to a sink node through the efficient routes. Many energy efficient routing algorithms were proposed. However, the previous algorithms consume more energy since they did not consider the transmission range and direction. In this paper we propose an algorithm TDRP(Transmission range and Direction Routing Protocol)that considers the transmission range and direction for the efficient data transmission. Since TDRP does not produce clusters or grids but four quadrants and send data to the nodes in one quadrant in the direction of the sink node, it has less network overhead. Furthermore since the proposed algorithm sends data to the smaller number of nodes compared to the previous algorithms, the energy efficiency is better than other algorithms in communication node fields that are located in packet transmit directions

An Energy Efficient Hybrid Routing Protocol Based on LEACH and PEGASIS

Since all sensor nodes in wireless sensor networks work by their own embedded batteries, if a node runs out of its battery, the sensor network can not operate normally. In this situation we should employ the routing protocols which can consume the energy of nodes efficiently. Many protocols for energy efficient routing in sensor networks have been suggested but LEACH and PEGASIS are most well known protocols. However LEACH consumes energy heavily in the head nodes and the head nodes tend to die early and PEGASIS - which is known as a better energy efficient protocol - has a long transfer time from a source node to sink node and the nodes close to the sink node expend energy sharply since it makes a long hop of data forwarding. We proposed a new hybrid protocol of LEACH and PEGASIS, which uses the clustering mechanism of LEACH and the chaining mechanism of PEGASIS and it makes the life time of sensor networks longer than other protocols and we improved the performance 33% and 18% higher than LEACH-C and PEGASIS respectively.

Paged segment striping scheme for the dynamic scheduling of continuous media servers

An innovative dynamic scheduling scheme is proposed to improve the efficiency of video-on-demand servers. We first introduce a paged segment striping model that makes dynamic scheduling possible. Based on this striping scheme, we propose a dynamic scheduling scheme that adapts to frequently changing workloads. In particular, we can change the round length without any additional disk access so that it can be adapted to changing request trends with a negligible cost in performance. This dynamic scheduling scheme always shows better performance than the static scheduling scheme in simulation. Although the dynamical scheme introduces additional scheduling overhead, it is very small when compared with the performance degradation in the static scheme.

Striping and scheduling for large scale multimedia servers

When designing a multimedia server, several things must be decided: which scheduling scheme to adopt, how to allocate multimedia objects on storage devices, and the round length with which the streams will be serviced. Several problems in the designing of large-scale multimedia servers are addressed, with the following contributions: (1) a striping scheme is proposed that minimizes the number of seeks and hence maximizes the performance; (2) a simple and efficient mechanism is presented to find the optimal striping unit size as well as the optimal round length, which exploits both the characteristics of VBR streams and the situation of resources in the system; and (3) the characteristics and resource requirements of several scheduling schemes are investigated in order to obtain a clear indication as to which scheme shows the best performance in realtime multimedia servicing. Based on our analysis and experimental results, the CSCAN scheme outperforms the other schemes. It is believed that the results are of value in the design of effective large-scale multimedia servers.