Search

In this paper, we propose a novel decentralized channel resource allocation algorithm for V2V communication based on deep multi-agent reinforcement learning. Each vehicle behaves as an independent agent and uses its local observation to select the optimal resource blocks (RBs) from pre-conflgured resource pool (environment). The selected RB is considered optimal if transmission within this RB causes a minimum interference to other ongoing transmissions. We applied an actor-critic reinforcement learning algorithm to let each agent conduct a centralized training and decentralized execution. In centralized training, agents share their actions and local observations through the centralized critic network. While making a decision (choosing the optimal resource blocks), therefore, each agent can estimate the policies of other agents. In decentralized execution, each agent uses its local observation to optimize its local policy independently. Each action taken by an agent in actor network is judged by its private critic network.

Hardware accelerated modules that can continuously measure/analyze resource (frequency channels, power, etc.) utilization in real-time can help in achieving efficient network control, and configuration in cloud managed wireless networks. As utilization of various network resources over time often exhibits broad and skewed distribution, estimating quantiles of metrics to characterize their distribution is more useful than typical approaches that tend to focus on measuring average values only. In this paper, we describe the development of a real-time quantile-based resource utilization estimator module for wireless networks. The intensive processing tasks run on the FPGA, while the command and control runs on an embedded ARM processor. The module is implemented by using high level synthesis (HLS) on a Xilinx’s Zynq-7000 series all programmable system on chip board. We test the performance of the implemented quantile estimator module, and as an example, we focus on forecasting congestion with real frequency channel utilization data. We compare the results from the implemented module against the results from a theoretical quantile estimator. We show that with high accuracy and in real time, the implemented module can perform quantile estimation and can be utilized to perform forecasting of congestion in wireless frequency spectrum utilization.

In this paper, we use the ECMA-392 standard, the first cognitive radio networks to meet the demand for IT services in the ship. First, we learn about ECMA-392-based network and examine the consideration and the problems for assigning channels to the device. And, we propose a fair channel resource allocation to ensure efficiently channel required by device groups in ECMA-392 MAC protocol. This is a method to reallocate the channels fairly with considering the channel assignment parameters to a new device group and existing device groups. The simulation results show that the proposed scheme improves throughputs compared to the existing one. ※ 이 논문은 2014년도 정부(교육부)의 재원으로 한국연구재단의 지원을 받아 수행된 기초연구사업(NRF-2009-0093828)과 미래창조과 학부 및 정보통신산업진흥원의 IT융합 고급인력과정 지원사업의 연구결과로 수행되었음(NIPA-2014-H0401-14-1009). ※ This research was supported by Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(2012R1A1A4A01009097) First Author : Institute Research of Information Science and Engineering, Mokpo National University, rikioh75@gmail.com, 정회원 ° Corresponding Author : Department of Information & Electronics Engineering, Mokpo National University, srlee@mokpo.ac.kr, 정회원 * Department of Electronics Engineering, Mokpo National University, 083591@mokpo.ac.kr ** Department of Information & Electronics Engineering, Mokpo National University, sypark@mokpo.ac.kr 논문번호:KICS2014-10-403, Received October 8, 2014; Revised December 4, 2014; Accepted December 4, 2014 I. 서 론 최근 조선업계의 최대 관심사는 IT와의 접목이다. 선박은 점차 대형화, 특수화 되고 있고, 승무원 및 승 객들은 육상에서 사용하는 것과 같은 편리한 통신 서 비스를 제공 받기를 기대하고 있다. 이렇게 선박에 IT 기술을 융합하면 다양한 요구를 만족 시킬 수 있다. 선박을 이용하는 승무원이나 승객에게 IT 기술을 활 논문 / ECMA-392 기반 선박 내 네트워크에서 Fair QoS 제공을 위한 채널 자원 할당 방안 1281 그림 1. ECMA-392 기본 네트워크 구조 Fig. 1. ECMA-392 basic network formation 용한 맞춤식 정보를 제공하여 미연에 사고방지를 할 수 있고, 근무환경 및 이용환경의 효율 또한 높일 수 있다. 이러한 서비스에 대한 요구가 점차 증대되면서 선박 내에서 사용 가능한 적응적이고 합리적으로 무 선 주파수를 활용하는 인지무선(Cognitive Radio) 기 술에 대한 연구가 진행되고 있다 . IEEE 802.22는 광대역 무선 인터넷 서비스에 인지 무선 기술을 적용하여 54∼862 MHz 사이의 TV 주 파수 대역에서 WRAN(Wireless Regional Area Network) 서비스를 제공하기 위한 표준을 선보였다 . IEEE의 활동과는 별개로 ECMA-International은 개인/휴대기기들을 위한 인지무선 표준을 발표하였다 . IEEE 802.22 표준은 고정형 기기를 이용하여 도 시 외곽 지역에서 서비스를 제공하는 반면에, ECMA-International 은 고정형 기기와 개인/휴대기기 의 혼합형 네트워크를 통한 서비스를 제공한다. ECMA-International의 TC48-TG1 표준은 가정용 고화질 멀티미디어와 인터넷 접속 등 실내에서 인지 무선 기술을 이용하여 비디오 스트리밍 등을 전송하 는 개인/휴대기기를 위한 무선 기술 개발을 목표로 한 다. 이 표준은 개인/휴대기기의 MAC(Medium Access Control) 계층과 PHY(Physical) 계층에 대한 기술표 준을 제공한다. 한편, 고속 무선 데이터 서비스에 대한 수요가 급격 히 증가하는 반면, 대부분의 스펙트럼은 전혀 사용되 지 않고 있거나 사용률이 매우 저조한 것이 현실이다. Shared Spectrum Company의 측정 결과에 의하면 30 ∼3,000 MHz 대역을 기준으로 미국 6개 지역에서 사 용되는 비율은 평균 5.2%에 불과하다. 인구 과밀 지 역인 뉴욕시의 경우에도 스펙트럼 사용률은 기껏해야 13.1% 수준이다 . 나아가 DTV로의 전환 시기와 맞 물려 사용되지 않거나 사용률이 저조한 스펙트럼 (white space 라고 부름)을 이용하기 좋은 기회라고 할 수 있다. TV white space의 장점은 높은 주파수 대역에 비 해 전파 특성이 우수하여 넓은 커버리지를 확보할 수 있다는 점이다. TV white space의 추가적인 대역으로 인해 ISM(Industrial Scientific and Medical) 대역의 포화 문제를 어느 정도 해결할 수 있을 것으로 예상된 다. ECMA-392 표준의 대표적인 서비스 모델은 홈 네 트워크에서 고화질 비디오 전송서비스, 빌딩이나 캠퍼 스 내에서 무선 데이터 서비스, 소도시에서 넓은 커버 리지를 만족하는 무선 인터넷 접속 등이다. 넓은 커버 리지를 만족하면서도 고속의 데이터 전송 성능을 보 이기 때문에 선박 내에서 사용하기에 적절하다고 볼 수 있다. 본 논문에서는 선박 내 IT 서비스에 대한 요구에 부응하기 위해 적응적이고 합리적인 무선 주파수를 활용하는 인지무선 기술 중 최초 표준인 ECMA-392 네트워크를 사용한다. 우선, ECMA-392 기반 네트워 크에 대해서 알아보고, 디바이스에게 채널 할당을 위 한 고려사항 및 문제점을 살펴본다. 또, ECMA-392 MAC 프로토콜에서 디바이스 그룹들이 요청하는 채 널을 효율적으로 보장하기 위한 공평한 채널 자원 할 당 방법을 제안한다. 새로운 디바이스 그룹과 기존의 디바이스 그룹들에게 채널 할당 파라미터를 고려하여 공평하게 채널들을 재할당할 수 있는 방법이다. 본 논문의 구성은 다음과 같다. 2장에서는 ECMA-392 네트워크 구조에 대해 설명하고, 채널 할 당을 위한 고려사항 및 문제점을 살펴본다. 3장에서는 공평한 채널 할당을 위한 방법을 제안하고, 4장의 성 능 평가를 통해 결론을 맺는다.

In this paper, we study the problem of stabilizing a linear time-invariant discrete-time system with information constraints in the input channels. The information constraint in each input channel is modelled as a sector uncertainty. Equivalently, the transmission error of an input channel is modelled as an additive system uncertainty with a bound in the induced norm. We attempt to find the least information required, or equivalently the largest allowable uncertainty bound, in each input channel which renders the stabilization possible. The solution for the single-input case, which gives a typical H∞ optimal control problem, is available in the literature and is given analytically in terms of the Mahler measure or topological entropy of the plant. The main purpose of this paper is to address the multi-input case. In the multi-input case, if the information constraint in each input channel is given a priori, then our stabilization problem turns out to be a so-called µ synthesis problem, a notoriously hard problem. In this paper, we assume that the information constraints in the input channels are determined by the network resources assigned to the channels and they can be allocated subject to a total recourse constraint. With this assumption, the resource allocation becomes part of the design problem and a modified µ synthesis problem arises. Surprisingly, this modified µ-synthesis problem can be solved analytically and the solution is also given in terms of the Mahler measure or topological entropy as in the single-input case.

In this paper, the networked control system (NCS) in continuous-time with sampled data is considered. The transmission channel is assumed to locate at the plant input, and introduce both logarithmic quantization and fading. The proper definition of channel capacity for such an NCS is presented. The capacity being greater than the topological entropy is found to be the necessary and sufficient condition for stabilizability for the single-input case. For the multi-input case, the sufficiency of the inequality is proved by means of channel resource allocation.

In this paper, we firstly introduce the hawk-pigeon game model. We simulate the model 1000 times using MATLAB. After carefully analyses the relationship between two users' gains and the probability of hawk strategy. The simulation results show that with fewer numerical hawk strategies are selected, the times of hawk-hawk are greatly reduced. Then we simulate and analyze the wireless channel resource, which introduces the hawk-pigeon game of learning mechanism. Learning mechanism is that next strategy based on several previous gains. We also add some constraints to achieve fairness of users of the learning mechanism. The simulation results demonstrate that by learning mechanism we achieve users' fair and high utilization rate of the channel.

With the development of communication technology and the severe increasing of number of wireless communication instruments, wireless channel resource becomes more crowded than many years before. It is a severe problem that how to build a wireless channel resource allocation mechanism with characters, such as: high efficient, less cost, short delay time and so on. This problem is also a hot topic in wireless communication field. There are many researchers that try to tackle this problem by using varied mechanisms. Game theory attracts increasing attention in the field of resource allocation. In this presented paper, we use an interesting game theory named grasping money game to allocate the wireless channel resource. This mechanism can usefully tackle the problem of jam in wireless channel. We study two players and multi-players (101) using grasping money game theory as the resource allocation mechanism. We use MATLAB as simulation tool. The simulation results show that the proposed mechanism is suitable in wireless channel resource allocation especially in multiuser and this mechanism can improve the efficiency of wireless channel.

In this paper, we study the stabilization of multirate networked control systems with norm bounded uncertainties in the input channels. The key idea is to use the channel resource allocation, i.e., given the overall capacity of the transmission network, we do have the freedom to allocate the capacities among different input channels. With this idea, we successfully show that a multirate networked control system could be stabilized by state feedback under an appropriate resource allocation if and only if the overall network capacity is larger than the topological entropy of the plant. We also apply the result to multirate quantized networked control systems. A sufficient condition for stabilization is obtained which involves a trade-off between the densities of time quantization and spatial quantization.

As the need of channel setting up increases, TV channel resource allocation will be managed by IPTV agent in the distributed TV system environment, not by network itself. This paper suggests an analytical model for class based admission control strategies in the distributed IPTV system.