Font Size: a A A

Research On Wireless Resource Allocation Scheme For Interference Management In UDN

Posted on:2021-02-08Degree:MasterType:Thesis
Country:ChinaCandidate:Y D TanFull Text:PDF
GTID:2428330632463002Subject:Information and Communication Engineering
Abstract/Summary:PDF Full Text Request
In Ultra-Dense Network(UDN),base stations(BSs)are densely deployed,which improves the network traffic,thus UDN has become one of the key technologies of the fifth generation mobile communication technology(5G).However,densely deployed BSs in UDN cause more severe interference.Meanwhile,the increasement of users(UEs)lead to the shortage of wireless resource,and the traditional wireless resource allocation(WRA)schemes are no longer applicable.Therefore,the interference and WRA problems in UDN limit the growth of UDN' s traffic.In order to further explore the potential of UDN,this thesis focus on the WRA scheme for interference management(IM)in UDN.The main research contents of the thesis contain three aspects.1)We outline the WRA scheme for IM in UDN.First,we introduce the advantage of UDN based on the cloud RAN(C-RAN).Then we elaborate the classical WRA algorithms for IM in BS-centric and UE-centric UDN and point out the.key ideas.Last,we summarize the limitations of the exiting classical WRA algorithms and point out the possible directions to achieve performance improvement.2)A novel topology-aware wireless resource allocation(T-WRA)scheme is proposed in the BS-centric UDN,which allocates the network devices with severe interference to the same cluster.Then we use the zero-forcing precoding to eliminate the strong intra-cluster interference.The proposed T-WRA includes two algorithms,namely the BS resource allocation algorithm based on the Mean-shift(MS)clustering and the reference signal received power(RSRP)-based UE distributed access algorithm.In the first algorithm of T-WRA,the BSs who are close to each other are allocated to the same cluster.In the second algorithm of T-WRA,UE accesses the cluster autonomously to maximize sum of all UE's RSRP.Simulation shows that the system performance is improved 25%as compared with other schemes.3)A novel WRA with low backhaul overhead in UE-centric UDN is proposed,which allocates BSs and spectrum for each UE.The novel WRA with low backhaul overhead includes three algorithms,namely the T-WRA algorithm based on the shortest distance,the spectrum resource allocation based on graph theory and the wireless topology correction algorithm based on the bargaining model.In the first algorithm,each UE accesses the nearest cluster under the constraint of BSs' load.In the second algorithm,the spectrum is allocated to the UE,which only needs limited interference information and ensures that UE's useful signal is maximized and co-channel interference is minimized.In the third algorithm,the spectrum efficiency is further improved by restricting the leakage of interference from UE accessing the BS.Simulation shows that the average spectrum efficiency is improved 57%as compared with other schemes.
Keywords/Search Tags:UDN, interference management, wireless resource allocation algorithm, machine learning, graph theory
PDF Full Text Request
Related items