Recent years have witnessed a huge demand for ubiquitous communications services from continuously moving users.In order to provide seamless network services to high-mobility users,one of the promising solution propos...Recent years have witnessed a huge demand for ubiquitous communications services from continuously moving users.In order to provide seamless network services to high-mobility users,one of the promising solution proposed by 3 GPP is the deployment of moving-relays.In this article,we introduce the concept of Moving-Personal-Cell(mPC),which is a type of moving-relays.mPC is a user-centric network,which aims to provide reliable network services to moving users.A mPC receives data-traffic from eNB and its neighboring mPCs via wireless backhaul and sidehual links respectively and forwards the received data to its serving users.In addition to this,mPC can also increase the network capacity by caching and distributing the popular contents to its serving users.Besides these pros,the mPC also has some limitations,as its performance is highly affected by cross-tier and co-tier interferences.In this article,we analyze the effect of these interferences on mPCs performance.Our results show that the performance of mPC network is equally affected by the capacity of wireless backhaul,sidehaul,and access links.Moreover,since mPCs accommodate data traffic from wireless backhaul,sidehaul links,and content cache,their performance is also affected by the ratio of data-traffic delivered via these links.展开更多
Since the different characteristics of various network services determine that their requirements for network are also disparate, the performance of one network varies according to the services running on it. However,...Since the different characteristics of various network services determine that their requirements for network are also disparate, the performance of one network varies according to the services running on it. However, most of previous network performance evaluation （NPE） researches conduct evaluations based on the network parameters, but without considering from the perspective of specific service running on the network. In view of this issue, a novel service-oriented NPE framework is proposed. First, the characteristics discrepancy among different types of services are investigated. Next, in order to conduct comprehensive evaluation of multiple services, an enhanced low-complexity adaptive （LA）-fuzzy analytical hierarchy process （FAHP） is introduced; meanwhile by applying the experts-construct-directly （ECD） algorithm proposed later, the consistency check required in previous studies can be omitted, thereby significantly reducing the computation complexity and assessment workload for experts. Then, in accordance with the features of each service, corrections are made to their respective membership functions, thus making the proposed LA-FAHP adaptive to various service evaluation scenarios. The subsequent comparison with other NPE methods well proves the effectiveness and high sensitivity of proposed framework, and the analysis verifies the low computation complexity of the proposed algorithms as well.展开更多
基金supported by Institute for Information and communications Technology Promotion(IITP)grant funded by the Korea government(MSIP)(No.R0101-15-244,Development of 5G Mobile Communication Technologies for Hyperconnected smart services)
文摘Recent years have witnessed a huge demand for ubiquitous communications services from continuously moving users.In order to provide seamless network services to high-mobility users,one of the promising solution proposed by 3 GPP is the deployment of moving-relays.In this article,we introduce the concept of Moving-Personal-Cell(mPC),which is a type of moving-relays.mPC is a user-centric network,which aims to provide reliable network services to moving users.A mPC receives data-traffic from eNB and its neighboring mPCs via wireless backhaul and sidehual links respectively and forwards the received data to its serving users.In addition to this,mPC can also increase the network capacity by caching and distributing the popular contents to its serving users.Besides these pros,the mPC also has some limitations,as its performance is highly affected by cross-tier and co-tier interferences.In this article,we analyze the effect of these interferences on mPCs performance.Our results show that the performance of mPC network is equally affected by the capacity of wireless backhaul,sidehaul,and access links.Moreover,since mPCs accommodate data traffic from wireless backhaul,sidehaul links,and content cache,their performance is also affected by the ratio of data-traffic delivered via these links.
基金supported by the Hi-Tech Research and Development Program of China (2014AA01A701)the Ministry of Education-CMCC research fund (MCM 20120132)Beijing Municipal Science and technology Commission research fund project "The Design of Radio Access Network Architecture in 5G communication system"
文摘Since the different characteristics of various network services determine that their requirements for network are also disparate, the performance of one network varies according to the services running on it. However, most of previous network performance evaluation （NPE） researches conduct evaluations based on the network parameters, but without considering from the perspective of specific service running on the network. In view of this issue, a novel service-oriented NPE framework is proposed. First, the characteristics discrepancy among different types of services are investigated. Next, in order to conduct comprehensive evaluation of multiple services, an enhanced low-complexity adaptive （LA）-fuzzy analytical hierarchy process （FAHP） is introduced; meanwhile by applying the experts-construct-directly （ECD） algorithm proposed later, the consistency check required in previous studies can be omitted, thereby significantly reducing the computation complexity and assessment workload for experts. Then, in accordance with the features of each service, corrections are made to their respective membership functions, thus making the proposed LA-FAHP adaptive to various service evaluation scenarios. The subsequent comparison with other NPE methods well proves the effectiveness and high sensitivity of proposed framework, and the analysis verifies the low computation complexity of the proposed algorithms as well.
