• Journal of Information Processing Systems
• /
• v.19 no.4
• /
• pp.439-449
• /
• 2023

This paper deals with a resource efficient artificial intelligence (AI) service architecture for multi-channel video streams. As an AI service, we consider the object detection model, which is the most representative for video applications. Since most object detection models are basically designed for a single channel video stream, the utilization of the additional resource for multi-channel video stream processing is inevitable. Therefore, we propose a resource efficient AI service framework, which can be associated with various AI service models. Our framework is designed based on the modular architecture, which consists of adaptive frame control (AFC) Manager, multiplexer (MUX), adaptive channel selector (ACS), and YOLO interface units. In order to run only a single YOLO process without regard to the number of channels, we propose a novel approach efficiently dealing with multi-channel input streams. Through the experiment, it is shown that the framework is capable of performing object detection service with minimum resource utilization even in the circumstance of multi-channel streams. In addition, each service can be guaranteed within a deadline.

• ETRI Journal
• /
• v.38 no.5
• /
• pp.847-857
• /
• 2016

Broadcasting services such as multi/single channel HDTV and 3DTV/2DTV use a multi-channel encoder that changes the bitrate and composition of the video service depending on the time. However, this type of multi-channel encoder could cause a longer latency owing to the variable bitrate and relatively bigger size of the buffers, which results in the same delay as in 3DTV even for a conventional DTV service. On the other hand, systems built based on separate encoders, each of which is optimized for the target service, might not have such latency problems. Nevertheless, there might be a distortion problem in the image and sound at the time of a switchover between two encoders with different output bitrates and group of picture structures. This paper proposes a system that can realize a seamless video service conversion using two different video encoders optimized for each video service. An overall functional description of the video service change control server, which is a main control block for the proposed system, is also provided. The experiment results confirm the seamless switchover and reduced broadcasting latency of DTV services compared with a broadcasting system composed of a multi-channel encoder system.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.227-241
• /
• 2021

Object detection techniques based on deep learning such as YOLO have high detection performance and precision in a single channel video stream. In order to expand to multiple channel object detection in real-time, however, high-performance hardware is required. In this paper, we propose a novel back-end server framework, a real-time AI vision platform (RAVIP), which can extend the object detection function from single channel to simultaneous multi-channels, which can work well even in low-end server hardware. RAVIP assembles appropriate component modules from the RODEM (real-time object detection module) Base to create per-channel instances for each channel, enabling efficient parallelization of object detection instances on limited hardware resources through continuous monitoring with respect to resource utilization. Through practical experiments, RAVIP shows that it is possible to optimize CPU, GPU, and memory utilization while performing object detection service in a multi-channel situation. In addition, it has been proven that RAVIP can provide object detection services with 25 FPS for all 16 channels at the same time.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.10 no.4
• /
• pp.78-85
• /
• 2011

In vehicle ad-hoc network (VANET) environments, traffic related information such as accident information, emergency information and real time traffic condition have to be delivered to on-board-unit (OBU) or/and road-side-equipment (RSE) for preventing traffic accidents in advance. In this paper, we introduce a Multi-Channel MAC (MCM) since the existing single channel operation may cause packet transmission delay and unexpected communication failure. To offer a seamless safety message transmission during the various services, it is necessary to manage the MAC scheduler in wireless access in vehicular environments (WAVE) systems. The MCM consists of MAC softwares and MAC hardwares where the former and the later ones are implemented with real time operation system based C language and FPGA module with VHDL language, respectively. The performance and QoS are verified by practical measurements and compared with the scheme using single channel operation.

• Journal of Broadcast Engineering
• /
• v.14 no.2
• /
• pp.178-188
• /
• 2009

In this paper, we propose the method which provides efficient sub-channel allocation for handoff and new call supporting multi-media service in AMC-based OFDMA system. Firstly, we apply the multi-band method which provides different AMC method according to the location of user terminals. Also, in OFDMA system environment that a base station has a lot of sub-channels, we adopt the sub-channel allocation scheme that provides a higher priority to handoff call and real-time service about handoff and new calls with multi-meida service. The simulation results show that the proposed scheme plays a role in increasing the number of new and handoff calls meeting the required blocking rate.

• Journal of KIISE:Information Networking
• /
• v.34 no.5
• /
• pp.396-404
• /
• 2007

In this paper, an Ad-hoc Routing Protocol which works in wireless Ad-hoc communication networks with multiple radios and multiple channels, and controls call admission based on bandwidth measurement is proposed. Unlike the conventional Ad-hoc node with a single radio using a single channel, an Ad-hoc node of the protocol proposed, MCQosR(Multiple Channel Quality of Service Routing), has multiple radios and uses multiple channels, which makes full duplex transmission between wireless Ad-hoc nodes, and reduces the intra interference on a route. Also, a fixed channel only for reception at each node enables the measurement of the available bandwidth, which is used to control the call admission for QoS provision. The performance of MCQosR is verified by simulation.

• International Journal of Contents
• /
• v.12 no.1
• /
• pp.65-69
• /
• 2016

Wireless communication chipsets have fixed transmission rate and communication distance. Although there are many kinds of chipsets with throughput and distance purpose, they cannot support various types of wireless applications. This paper provides theoretic research results in order to support various wireless applications requiring different throughput, delayed quality-of-service (QoS), and different communication distances by using a wireless communication chipset with fixed rate and transmission power. As a performance metric, the probability for a data frame that successfully receives at a desired receiver is adopted. Based on this probability, the average number of transmission in order to make a successful frame transmission is derived. Equations are utilized to analyze the performance of a single-hop with channel coding and a dual-hop without error correction matter transmission system. Our results revealed that single-hop transmission assisted by channel coding could extend its communication distance. However, communication range extending effect of the single-hop system was limited. Accordingly, dual-hop transmission is needed to overcome the communication distance limit of a chipset.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.8A
• /
• pp.656-663
• /
• 2005

This paper presents a multi-channel enhancement scheme for the MAC protocol of IEEE 802.15.3 High-rate WPAN (Wireless Personal Area Network). The current MAC protocol of the IEEE 802.15.3 High-rate WPAN is designed for sharing a single channel among DEVs of a piconet; that is, within a single piconet, PNC prevents interference through MAC layer assignment of time slots to other DEVs using time-division multiple access. When the number of DEVs that communicate with each other frequently, is increased in a single WPAN, the size of the superframe becomes inevitably large, and this may result in a significant throughput drop or a failure to provide QoS guarantee. A multi-channel enhancement scheme for the MAC protocol of IEEE 802.15.3 High-rate WPAN is proposed to significantly increase the aguegate throughput and more reliably provide the QoS guarantees in a piconet

• Journal of Information Technology Applications and Management
• /
• v.26 no.2
• /
• pp.41-60
• /
• 2019

O2O(Online-to-Offline) commerce, a kind of multi-channel, is used in various industries such as accommodation, food and beverage, transportation, and real estate. The O2O commerce, a channel of integrating online and offline, overcomes the limitations of traditional online commerce, where a customer made a purchase decision without direct experience. Despite this advantage of O2O, if the online-offline channel is not effectively linked, customer's complaints would occur due to service failure. This study, with regard to O2O service failure circumstances, intends to investigate the effect of customer's perceived justice on their emotional response, recovery satisfaction, followed by intention to repurchase. Perceived justice in this study is composed of distributive justice, interactional justice, and procedural justice. Furthermore, this study explores the moderation effect of O2O industry types in the relationship between perceived justice and emotion. An O2O industry type is classified into accommodation, food and beverage, and transportation. A hypothesized research model was empirically tested using a structural equation model. The current study collected 433 questionnaires and the target respondents are customers who have experienced service failure in O2O commerce. The empirical results showed that O2O commerce more effectively conducts service recovery strategy and causes positive customer response by integrating online and offline channel. One of the ultimate purposes of O2O service providers is to reduce the likelihood of service failures and to recover more quickly and efficiently by linking two channels rather than using a single channel. This study suggested that the O2O channel is effective in influencing customer satisfaction and loyalty by inducing customer's positive emotions in recovering service failure.

• Journal of Broadcast Engineering
• /
• v.11 no.1 s.30
• /
• pp.107-115
• /
• 2006

In the environment of single application data broadcasting, only one application can be serviced at a specific time on one channel. To overcome this, we developed the structure and the method of operation for multi-applications which are fully conformant to the ACAP (Advanced Common Application Platform), and modified data broadcasting system to support multi-applications. In multi-application environment, broadcasters can service multiple applications simultaneously at a specific airtime on one channel so users can enjoy services selectively according to their preferences. In this paper, we present an example of multi-application service which was developed to make an experiment before servicing them to users on the air. The core of the multi-application is a manager application which manages other ordinary applications, so we describe the function and structure of the manager application, and then present the experimental results to show that the proposed method is the proper model for multi-applications.