• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1997년도 추계학술대회 학술발표 논문집
• /
• pp.111-114
• /
• 1997

Usually, Evolutionary Algorithms are considered more efficient for optimal system design, However, the performance of the system is determined by fitness function and system environment. In this paper, in order to overcome the limitation of the performance by this factor, we propose a co-evolutionary method that two populations constantly interact and coevolve. In this paper, we apply coevolution to neural network's evolving. So, one population is composed of the structure of neural networks and other population is composed of training patterns. The structure of neural networks evolve to optimal structure and, at the same time, training patterns coevolve to feature patterns. This method prevent the system from the limitation of the performance by random design of neural network structure and inadequate selection of training patterns. In this time neural networks are trained by evolution strategies that are able to apply to the unsupervised learning. And in the coding of neural networks, we propose the method to maintain nonredundancy and character preservingness that are essential factor of genetic coding. We show the validity and the effectiveness of the proposed scheme by applying it to the visual servoing of RV-M2 robot manipulators.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2020년도 추계학술대회
• /
• pp.145-147
• /
• 2020

딥러닝 기술이 발전함에 따라 High-Efficiency-Video-Coding(HEVC)와 같은 비디오 코덱을 딥러닝 기술을 사용해 발전시키는 방법이 많아 시도되었다. 가장 많이 연구된 분야 중 하나에는 코덱 내부의 필터들을 영상 복원 기술을 통해 개선시키는 연구가 있다. 본 논문에서는 그 중 Sample adaptive offset(SAO) 필터링 기법을 깊은 신경망으로 대체하는 방식을 제안한다. SAO는 주변 정보들을 통해 에러의 형태를 결정하고 그를 상쇄하는 값을 전송하는 형태를 가진다. 이 때 찾은 보상 값이 최적의 값이 아니기 때문에 제안 기법에서는 깊은 신경망을 통해서 그 값을 찾는다. 제안하는 네트워크는 최적의 보상 값을 찾는 부분과 에러의 형태를 찾는 부분으로 이루어져 있으며, 두 네트워크를 통해 비선형적이고 복잡한 형태의 에러를 제거할 수 있다. 실험 결과 제안하는 방식은 저지연 P 모드와 임의접근 모드에서 기존 HEVC 보다 좋은 성능을 낸 것을 확인할 수 있다.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2020년도 추계학술대회
• /
• pp.180-182
• /
• 2020

최근 IoT 기술이 대중화됨에 따라 커넥티드 카, 스마트 시티와 같은 machine-to-machine 기술의 활용 분야가 다양화되고 있다. 이에 따라, 기계 지향 비디오 처리 및 부호화 기술에 대한 연구분야에 산업계와 학계의 관심 역시 집중되고 있다. 국제 표준화 단체인 MPEG은 이러한 추세를 반영하여 기존 비디오 부호화 표준을 개선할 새로운 표준을 수립하기 위해 Video Coding for Machines (VCM) 그룹을 구성하여 기계 소비를 대상으로 하는 비디오 표준의 표준화를 진행하고 있다. 이에 본 논문에서는 VCM이 기계 소비를 대상으로 진행하고 있는 특징맵 부호화의 부호화 효율을 개선하기 위해 특징맵을 시간적, 공간적으로 재정렬하는 방법을 제안한다. 실험 결과, 제안 방법이 CityScapes의 검증 세트 내 일부 이미지에 대해 시간적 재정렬을 수행한 결과 random access 조건에서 최대 1.48%의 부호화 효율이 향상됨이 확인되었다.

• Journal of Communications and Networks
• /
• 제17권5호
• /
• pp.482-490
• /
• 2015

This article exploits spatial diversity for jamming to prevent wiretapping in the extreme case in which an eavesdropper is located near the source and a common jamming signal is unavailable. To address this challenge, the jamming signal is allowed to carry a random binary message. Then, it is proposed that the active intermediate node transmits this jamming signal and the decoding of this signal at both source and destination is physically secured as result of using the physical-layer security method. If the source and the destination securely and correctly decode this jamming message, the source transmits another message which is created from combining its information message and the decoded message using the network-coding method. Therefore, this method prevents the transmissions from being eavesdropped upon by the source-wiretapping.

• Journal of Communications and Networks
• /
• 제16권4호
• /
• pp.397-406
• /
• 2014

Radio frequency (RF) jamming is a denial of service attack targeted at wireless networks. In resource-hungry scenarios with constant traffic demand, jamming can create connectivity problems and seriously affect communication. Therefore, the vulnerabilities of wireless networks must be studied. In this study, we investigate a particular type of RF jamming that exploits the semantics of physical (PHY) and medium access control (MAC) layer protocols. This can be extended to any wireless communication network whose protocol characteristics and operating frequencies are known to the attacker. We propose two efficient jamming techniques: A low-data-rate random jamming and a shot-noise based protocol-aware RF jamming. Both techniques use shot-noise pulses to disrupt ongoing transmission ensuring they are energy efficient, and they significantly reduce the detection probability of the jammer. Further, we derived the tight upper bound on the duration and the number of shot-noise pulses for Wi-Fi, GSM, and WiMax networks. The proposed model takes consider the channel access mechanism employed at the MAC layer, data transmission rate, PHY/MAC layer modulation and channel coding schemes. Moreover, we analyze the effect of different packet sizes on the proposed jamming methodologies. The proposed jamming attack models have been experimentally evaluated for 802.11b networks on an actual testbed environment by transmitting data packets of varying sizes. The achieved results clearly demonstrate a considerable increase in the overall jamming efficiency of the proposed protocol-aware jammer in terms of packet delivery ratio, energy expenditure and detection probabilities over contemporary jamming methods provided in the literature.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권12호
• /
• pp.5191-5211
• /
• 2016

It is essential to satisfy class-specific QoS constraints to provide broadband services for new generation networks. The present study proposes a QoS-driven multicast scheme for wireless networks in which the transmission rate and end-to-end delay are assumed to be bounded during a multiple multicast session. A distributed algorithm was used to identify a cost-efficient sub-graph between the source and destination which can satisfy QoS constraints of a multicast session. The model was then modified as to be applied for wireless networks in which satisfying interference constraints is the main challenge. A discrete power control scheme was also applied for the QoS-aware multicast model to accommodate the effect of transmission power level based on link capacity requirements. We also proposed random power allocation (RPA) and gradient power allocation (GPA) algorithms to efficient resource distribution each of which has different time complexity and optimality levels. Experimental results confirm that the proposed power allocation techniques decrease the number of unavailable links between intermediate nodes in the sub-graph and considerably increase the chance of finding an optimal solution.

• Journal of Internet Technology
• /
• 제21권1호
• /
• pp.203-219
• /
• 2020

Compressed video sensing (CVS) is one of the 5G application of compressed sensing (CS) to video coding. Block-based residual reconstruction is used in CVS to explore temporal redundancy in videos. However, most current studies on CVS focus on random measurements without quantization, and thus they are not suitable for practical applications. In this study, an efficient ratecontrol scheme combining measurement rate and quantization for residual reconstruction in CVS is proposed. The quantization effects on CS measurements and recovery for video signals are first analyzed. Based on this, a mathematical relationship between quantitative distortion (QD), sampling rate (SR), and the quantization parameter (QP) is derived. Moreover, a novel distortion model that exhibits the relationship between QD, SR, and QP is presented, if statistical independency between the QD and the CS reconstruction distortion is assumed. Then, using this model, a rate-distortion (RD) optimized rate allocation algorithm is proposed, whereby it is possible to derive the values of SR and QP that maximize visual quality according to the available channel bandwidth.

• 한국의학물리학회:학술대회논문집
• /
• 한국의학물리학회 2002년도 Proceedings
• /
• pp.489-491
• /
• 2002

The need for video diagnosis in medicine has been increased and real-time transfer of digital video will be an important component in PACS and telemedicine. But, Network environment has certain limitations that the required throughput can not satisfy quality of service (QoS). MPEG-4 ratified as a moving video standard by the ISO/IEC provides very efficient video coding covering the various ranges of low bit-rate in network environment. We implemented MPEG-4 CODEC (coder/decoder) and applied various compression ratios to moving ultrasound images. These images were displayed in random order on a client monitor passed through network. Radiologists determined subjective opinion scores for evaluating clinically acceptable image quality and then these were statistically processed in the t-Test method. Moreover the MPEG-4 decoded images were quantitatively analyzed by computing peak signal-to-noise ratio (PSNR) to objectively evaluate image quality. The bit-rate to maintain clinically acceptable image quality was up to 0.8Mbps. We successfully implemented the adaptive throughput or bit-rate relative to the image quality of ultrasound sequences used MPEG-4 that can be applied for diagnostic performance in real-time.

• Journal of Communications and Networks
• /
• 제11권1호
• /
• pp.47-56
• /
• 2009

We investigate the connectivity of fading wireless ad-hoc networks with a pair of novel connectivity metrics. Our first metric looks at the problem of connectivity relying on the outage capacity of MIMO channels. Our second metric relies on a probabilistic treatment of the symbol error rates for such channels. We relate both capacity and symbol error rates to the characteristics of the underlying communication system such as antenna configuration, modulation, coding, and signal strength measured in terms of signal-to-interference-noise-ratio. For each metric of connectivity, we also provide a simplified treatment in the case of ergodic fading channels. In each case, we assume a pair of nodes are connected if their bi-directional measure of connectivity is better than a given threshold. Our analysis relies on the central limit theorem to approximate the distribution of the combined undesired signal affecting each link of an ad-hoc network as Gaussian. Supported by our simulation results, our analysis shows that (1) a measure of connectivity purely based on signal strength is not capable of accurately capturing the connectivity phenomenon, and (2) employing multiple antenna mobile nodes improves the connectivity of fading ad-hoc networks.

• Journal of Communications and Networks
• /
• 제17권2호
• /
• pp.133-144
• /
• 2015

In this paper, we introduce a system and a set of algorithms for disseminating popular content to a large group of wireless clients spread over a wide area. This area is partitioned into multiple cells and there is a base station in each cell which is able to broadcast to the clients within its radio coverage. Dissemination of information in the proposed system is hybrid in nature: Each base station broadcasts a fraction of information in the form of random linear combinations of data blocks. Then the clients cooperate by exchanging packets to obtain their desired messages while they are moving arbitrarily over the area. In this paper, fundamental trade-offs between the average information delivery completion time at the clients and different parameters of the system such as bandwidth usage by the base stations, average energy consumption by the clients and the popularity of the spread information are studied. Moreover different heuristic algorithms are proposed to control and maintain a balance over these trade-offs. Also, the more complicated case of multiple sessions where each client is interested in an arbitrary subset of sessions is considered and two variants of the basic dissemination algorithm are proposed. The performance of all the proposed algorithms is evaluated via extensive numerical experiments.