• Journal of information and communication convergence engineering
• /
• 제20권1호
• /
• pp.22-30
• /
• 2022

The reinforcement learning algorithm has proven its potential in solving sequential decision-making problems under uncertainties, such as finding paths to route data packets in wireless sensor networks. With reinforcement learning, the computation of the optimum path requires careful definition of the so-called reward function, which is defined as a linear function that aggregates multiple objective functions into a single objective to compute a numerical value (reward) to be maximized. In a typical defined linear reward function, the multiple objectives to be optimized are integrated in the form of a weighted sum with fixed weighting factors for all learning agents. This study proposes a reinforcement learning -based routing protocol for wireless sensor network, where different learning agents prioritize different objective goals by assigning weighting factors to the aggregated objectives of the reward function. We assign appropriate weighting factors to the objectives in the reward function of a sensor node according to its hop-count distance to the sink node. We expect this approach to enhance the effectiveness of multi-objective reinforcement learning for wireless sensor networks with a balanced trade-off among competing parameters. Furthermore, we propose SDN (Software Defined Networks) architecture with multiple controllers for constant network monitoring to allow learning agents to adapt according to the dynamics of the network conditions. Simulation results show that our proposed scheme enhances the performance of wireless sensor network under varied conditions, such as the node density and traffic intensity, with a good trade-off among competing performance metrics.

• 전자공학회논문지S
• /
• 제34S권5호
• /
• pp.25-34
• /
• 1997

Relatively slow tuning speed of optical device causes the unwanted delay in high speed single-hop multiwavelength networks. To lessen the overhead, we present a channel access protocol with dynamjic message scheduling. Th eframe structure of proposed protocol adopts hybrid multiaccess scheme in which WDMA is used as a basic multiaccess technique and TDMA is used to provide subchannels within a wavelength band. This architecture has two merits : the network extention is not limited by available number of wavelengths, and the transmission delay caused by optical device tuning time can to minimize the number of tunings. It schedules messages that require same wavelength channels sequencely, so the total transmission delay is reduced by decreasement of wavelength changes. The performance of proposed protocol is evaluated through numerical analysis based on probability and queueing theory. The peformance of proposed protocol is evaluated through numerical analysis based on probability and queueing theory. The numeric results show that the peformance of proposed protocol is better than that of previous one.

• Journal of electromagnetic engineering and science
• /
• 제9권3호
• /
• pp.111-117
• /
• 2009

This paper presents an energy-efficient and bandwidth-efficient 2-1-1 relaying scheme in which a sensor node(SN) assists two others in their data transmission to a clusterhead in WSNs(Wireless Sensor Networks) using LEACH (Low-Energy Adaptive Clustering Hierarchy). We derive the closed-form BER expression of this scheme which is also a general BER one for the decode-and-forward cooperative protocol and prove that the proposed scheme performs the same as the conventional relaying scheme but obtains higher channel utilization efficiency. A variety of numerical results reveal the relaying can save the network energy up to 11 dB over single-hop transmission at BER of $10^{-3}$.

• 정보학연구
• /
• 제4권4호
• /
• pp.65-72
• /
• 2001

본 논문에서는 단일홉의 WDM 네트워크에서 멀티캐스트 서비스를 위한 그룹분할 예약할당방식(MSA l)과 그룹 예약할당방식(MSA ll)을 제안하고 그 성능을 시뮬레이션을 이용하여 확인하였다. 그룹분할 예약할당방식은 여러 슬롯으로 나누어 멀티캐스트 패킷을 송신하는 방식이다. 반면에 그룹 예약할당방식은 하나의 슬롯에서 멀티캐스트 패킷을 송신하는 방식이다. 충분한 채널이 있는 경우에는 MSA l 방식이 MSA ll 방식보다 더 좋은 성능을 제공한다. 그러나 채널의 제한되는 경우에는 트래픽이 증가함에 따라 점차 MSA ll 방식의 성능이 MSA l 보다 우수함을 입증하였다.

• 한국경영과학회지
• /
• 제34권3호
• /
• pp.137-153
• /
• 2009

In this paper, we present an analytic model that evaluates the performance of B-MAC protocol under contention transmission conditions of two senders in a single-hop wireless sensor network. Our model considers the impact of several important factors such as sleep cycle, the backoff mechanism and incoming traffic loads. After obtaining the service delay distribution of a sending node and expected delay of a receiving node, an iterative algorithm is presented for calculating the performance measures such as expected energy consumption usage per packet and latency. Simulation results show that the proposed analytic model can accurately estimate the performance measures under different traffic conditions.

• 한국경영과학회지
• /
• 제22권4호
• /
• pp.17-35
• /
• 1997

This paper considers the efficient design of routing path and wavelength assignment in the single-hop WDM optical transport networks. The connection demands between node-pairs are given and a connection between a pair of nodes must be made by only one lightpath. It is assumed that no wavelength conversion is allowed and the physical topology of the network and the number of available wavelengths of a physical link are also given. This paper proposes the method to route a path and assign a wavelength to the path, which guarantees fairness to every node pair. In order to guarantee fairness and maximize connections, an optimization model is established, which successively maximizes minimum connection ratio to the demand of each node pair. A heuristic algorithm with polynomial iterations is developed for the problem since the problem is NP-hard. The heuristic algorithm is implemented and applied to the networks with real problem sizes. The results of the application are compared with GAMS/OSL.

• 한국정보통신학회논문지
• /
• 제6권2호
• /
• pp.248-256
• /
• 2002

본 논문에서는 WDM 단일홉 Broadcast-and-select망에서 멀티캐스트 서비스를 지원하기 위한 스케줄링 알고리즘을 제안하였다 기존의 멀티캐스트 스케줄링 알고리즘은 수신기의 이전상태를 고려하지 않는 방법이지만, 본 논문에서 제안하는 H-EAR, PGM 기법은 튜닝이 가능한 송신기와 수신기의 상태정보, 그리고 메시지전송시간과 튜닝지연시간을 이용하여 서브그룹을 나누는 알고리즘이며, 컴퓨터 시뮬레이션을 통해 제안된 알고리즘의 성능이 기존의 멀티캐스트 스케줄링 알고리즘에 비하여 지연시간과 서브 그룹수의 측면에서 우수함을 보였다.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 1999년도 가을 학술발표논문집 Vol.26 No.2 (3)
• /
• pp.584-586
• /
• 1999

본 논문에서는 파장 재사용 이중 단일홉 WDM(Wavelength Division Multiplexing) 통신망에서 전체 노드의 수가 사용 가능한 파장의 수보다 클 경우에 트래픽의 지역성(locality)을 고려하여 시간 분할 다중 접속 (TDMA) 방식으로 파장을 이용하는 다중 접근 알고리즘을 제안한다. 사용 가능한 파장을 그룹내에서 사용하는 전역 파장과 그룹간의 통신에 사용하는 지역 파장으로 나누고 전역 파장은 통신망에 속하는 모든 노드가 공유하고 지역 파장은 그룹내의 노드가 공유한다. 따라서 지역 파장은 모든 그룹 내에서 재사용 가능하기 때문에 전체 파장의 사용율이 증가한다. 모든 노드는 전역파장와 지역 파장으로 동조 가능한 하나의 동조 송신기와 전역 파장 중에 하나의 파장으로 수신 파장이 고정된 고정 수신기와 지역 파장 중에 하나의 파장으로 수신 파장이 고정된 고정 수신기를 가진다. 제안하는 알고리즘은 노드에서 패킷의 전송 요구가 필요한 만큼 슬롯 시간을 할당받아 모든 패킷을 연속해서 전송하는 슬롯 예약 방식을 채택한다. 이 방식은 동조 시간을 최소화함으로써 슬롯 시간을 모드 노드가 균일하게 할당하여 전체 동조 시간이 길어지는 고정 할당 방식보다 파장 사용율이 효율적이다.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2012년도 제45차 동계학술발표논문집 20권1호
• /
• pp.105-108
• /
• 2012

Wireless mesh networks are considered a promising solution to last mile broadband. The unique characteristics of WMN impose unique requirements on designing routing protocols and metrics for WMN. However, existing routing schemes that are designed for single-channel multi-hop wireless networks may lead to inefficient routing paths in multichannel. This paper focuses on the routing problem for multi-radio multichannel WMNs. We list the challenges in designing routing algorithms for multi-radio multichannel WMNs. Then we examine the requirements and considerations for designing routing metrics according to the characteristics of multi-radio multichannel WMNs. Finally we survey and investigate the existing routing metrics in terms of their ability to satisfy these requirements.

• ETRI Journal
• /
• 제18권3호
• /
• pp.127-145
• /
• 1996

A new analytic framework based on a linear algebra approach is proposed for examining the performance of a direct sequence spread spectrum (DS/SS) slotted ALOHA wireless communication network systems with delay capture. The discrete-time Markov chain model has been introduced to account for the effect of randomized time of arrival (TOA) at the central receiver and determine the evolution of the finite population network performance in a single-hop environment. The proposed linear algebra approach applied to the given Markov problem requires only computing the eigenvector ${\prod}$ of the state transition matrix and then normalizing it to have the sum of its entries equal to 1. MATLAB computation results show that systems employing discrete TOA randomization and delay capture significantly improves throughput-delay performance and the employed analysis approach is quite easily and staightforwardly applicable to the current analysis problem.