Cheul-hee Hahm

An efficient JPEG decoding and scaling method for digital TV platforms

This paper presents an efficient JPEG decoding and scaling method which reduces both of memory requirement and processing time. The proposed method overlaps decoding, horizontal scaling and vertical scaling, which results in average 92.6% reduction of memory space and average 82.8% reduction of processing time for test cases.

Enhance OS security by restricting privileges of vulnerable application

Restricting vulnerable process and its file I/O behaviors is important for securing an embedded device against intrusions, malware infection and information leakage. On embedded device like Smart TV, user can connect to internet using Web Browser process and can download many applications and games. These applications or games may have viruses or vulnerable code. If attacker with root privilege is able to launch new process (using fork () in Linux system) then the downloaded virus program will also get root privileges and hence vulnerable application with root user privileges may corrupt the whole system. This paper proposes a new method to protect the system from this kind of security threats. For experiment and result verification, proposed method uses LSM framework and TOMOYO module. It implements the LSM hook in fork/exec system call path to control the privileges of vulnerable application/binary. Android addresses this problem by writing a wrapper to fork system call in their framework, which modifies credential of new process. However they leave security hole, if application directly calls fork () system call.

Self-synthesis based super-resolution for a natural image

In this paper, we propose self-synthesis based super-resolution algorithm for natural images. The proposed method analyzes the property of the natural image and focuses on the self-similarity of natural objects. Structure and texture components are extracted from the self-image and synthesized on the interpolated image with matching weights to restore the self-similarity of interpolated image. Then, the synthesized image is compensated with MAP estimator to prevent artifacts. Experimental results show that the proposed method gives more natural texture in terms of human perception.

Memory Access Scheduling for a Smart TV

A smart TV system-on-chip (SoC) has very heavy computation and memory demands that must be met with low-cost components. As a result, there is potentially an extremely high utilization of the channel between the SoC and its memory chip. This paper presents the design of a new memory access scheduler customized for the type of memory traffic typically encountered with smart TVs. This includes special accumulated hard real-time graphics requirements, user response-sensitive soft real-time requirements, and the need to provide high memory throughput and priority-handling capabilities even under extremely heavy memory traffic conditions. The simulation results show that the proposed memory access scheduler is able to achieve up to 98% of the ideal upper bound memory throughput when faced with extremely heavy memory traffic-this is a significant improvement over previous schedulers. Novel future prediction and light-handed priority handling methods are used to achieve these results while satisfying the unique real-time requirements of smart TVs.

On the improvements of fast user interactivity in consumer electronic devices using Linux

This paper presents an improvement technique of fast user interactivity in consumer electronic devices using Linux OS. Firstly, fast user responsiveness is achieved by use of applying real time policy to dynamic boosting [1]. Second, we introduce a repetitive measurement method to evaluate the responsiveness of users continuous operations. Experimental results show that the proposed technique gives 34% better performance in application launching time and shows stable fast channel change response compared with the conventional ones in terms of automatic repetitive measurements.

An efficient device to device RPC framework using DBUS and KDBus

Due to rapid advancement of technology, embedded devices are evolving quickly and closing the gap between virtual and real work. Embedded devices should be able to use services and data offered by other embedded devices efficiently. This requires an efficient and scalable inter device communication framework which can be used to provide Internet of Things (IOT) ecosystem as well. In this paper we are proposing an efficient device to device communication framework by extending the DBus and KDBus Remote Procedure Call (RPC) mechanism. The proposed framework consists of two RPC interfaces: one is in user space for control and the other is in kernel space for data transport. User space RPC interface authenticates devices and gathers services information using DBus introspection feature to share it with other embedded devices. Kernel level RPC interface will be responsible for transferring and receiving payload with other embedded devices. Kernel RPC interface will use kernel socket to achieve better data throughput.

On the measurement of thread-level memory usage

In the embedded system, the memory profiling of an application is very critical and important to solve the problem. The memory usage is usually measured on the basis of process-level granularity. However, the process-level memory usage gives a lack of information to analyze an application, which consists of big processes with many threads. If a specific thread in the process has memory problems like memory leak or large memory consumption, then this may lead to a malfunction of an application due to the waste of an unnecessary memory by the problematic thread. This paper gives an efficient framework to get a memory usage of a process with thread level granularity and we called it Memory Usage Per Thread (MUPT). This framework would provide memory allocation information in terms of page_fault for a given multi-threaded process with thread granularity and results are expressed using new added proc interface. Thus, MUPT provides a mechanism to debug or monitor the thread memory usage and guide an application user to detect memory leak or large amount of incorrect memory allocation.

Peformance optimization utilizing heterogeneous multi-cores for smart TV applications

This paper presents an optimization technique to improve graphics performance for web-based applications taking advantage of heterogeneous multiple cores embedded in smart TV devices. The proposed technique balances workload over multiple cores by offloading jobs to suitable cores, especially GPU cores, considering the type of workload, which results in about 3.5× performance improvement to reach 60 fps.

An improvement of audio transfer QoS over 802.11 WiFi Network environment

High quality wireless audio transfer is essential for external digital audio products. There are two kinds of methods for transferring audio data. One is using proprietary RF channel and protocol for reliable audio transfer, the other is using preexisting 802.11 WiFi Network. Audio transfer via WiFi network is more general and preferred method compared to proprietary wireless transfer. But when WiFi traffic is getting heavy, it is difficult to guarantee QoS of audio Data. In this paper, we have proposed independent data transfer using additional USB endpoint for ensuring QoS of data.

On the improvement of file browsing time in single client NFS system

The Network File System (NFS) is one of the most popular remote file system protocols for accessing directories and files over the network. If NFS mount point has thousands of files (e.g. a directory containing 5000 files), there is a significant performance drop in directory browsing (traverse, open(), stat() and read()) time as compared to local file system browsing (traverse, open(), stat() and read()) time. In order to reduce the file browsing time in the NFS with single client system, this paper presents the modification of a conventional NFS protocol usage with multiple client systems. Experimental results show the proposed implementation gives 44.6% performance improvement over a conventional one.