Seungkwon Cho

Improving Power Savings by Using Adaptive Initial-Sleep Window in IEEE802.16e

In this paper, we propose an adaptive initial-sleep window scheme applicable to both power saving class of type I and type II in IEEE802.16e. Based on the estimated inter-arrival time of packet burst, the presented scheme adaptively updates the length of the initial-sleep window in order to maximize power-saving effect while trying to minimize packet buffering delay. The computer simulations are carried out for Web browsing and streaming video, the most typical BE service in power saving class of type I and RT-VR service in power saving class of type II, respectively. The simulations show that the proposed scheme improves the power-saving effect in terms of the awake rate and the normalized active time while suppressing the increase in the mean packet delay.

Linux BYTEmark Benchmarks: A Performance Comparison of Embedded Mobile Processors

This paper describes a performance comparison of three well-known embedded mobile processors, Intel SA1110, Intel PXA255, and Intel PXA270. Even though these processors are widely used in commercial handheld devices and consecutively developed in order to improve processing capability, there has been a lack of experimental study investigating their comparative performance. We present here the results of an experimental study into performance comparison based on Linux BYTEmark benchmark suit. The benchmark results show that PXA255 has a significant performance improvement, especially in floating-point operations compared with the other two processors.

A fully distributed coordination scheme based on orthogonal requests and responses

Device-to-device (D2D) communication is a communication technology that enables one device to transmit data directly to the other device in proximity instead of conveying data via infrastructure. Without the coordination of infrastructure, multiple access for D2D communication is performed in a distributed manner. Even though contention-based distributed multiple access scheme is widely adopted in the existing D2D systems, the excessive contention overhead with a large number of devices makes contention-based scheme lose scalability. In this paper, we present the rationale behind the design of a new fully distributed multiple access scheme based on contention-free resource request and response message exchange, which is proposed for IEEE 802.15.8. Through the results of simulations with full buffer and VoIP traffic model, we show that the presented scheme achieves scalability with the help of orthogonal requests and responses.

MAC HW/SW partitioning for aggregation in IEEE 802.11n based on a interrupt latency measurement

Hardware and software partitioning for A-MSDU and A-MPDU in IEEE 802.11n requires many considerations due to the timing requirements of the MAC functions. We propose a MAC architecture suitable for adopting both MSDU aggregation and MPDU aggregation based on the measured interrupt latency statistics.

Processing Acceleration of Broadband Wireless MAC in a Portable Terminal

Among the broadband wireless communication standards utilized to satisfy the demand for multimedia services, time division duplexing (TDD) is satisfactory for the asymmetric data transmission emphasized in Internet services. In this system, the transition between receiving a frame and transmitting a response must be bounded for an effective use of radio resources. However, the minimized inter-frame space-time requires high processing power. The aim of the present paper is to gain insight into the time latency at the turn-around time of a TDD operation. We also propose a simplified new processor, which is a terminal device-friendly architecture that includes prediction and preparation to support processing of burst-type traffic.