Research On Multi-channel Allocation And Access Mechanism For High-density Wireless Network

The contradiction between the exponentially increasing nodes and the limited spectrum resource has been prevailing recently,which poses great challenge to high-density wireless networks.As the nodes' locations are usually distributed in a random manner and also change in the dynamic network environment,centralized algorithms for network optimization is not always applicable.And therefore,distributed networks have been a research focus in recent years.In distributed networks,efficient utilization of limited multiple non-overlap channels and reasonable channel allocation are promising solutions to improve the network capacity and reduce the network interference.Furthermore,channel rendezvous allows nodes that operate on multiple channels to meet on a common channel and hence enable device-to-device communication.Therefore,distributed multi-channel network have been considered as one of promising technologies to enhance the high-density wireless communication.In this thesis,in order to analyze and optimize the performance of channel allocation and spectrum access in high-density wireless networks,we consider the complex network environment(including channel heterogeneity,node mobility and energy consumption),study the network access control(including channel adaptation,independent channel sensing and channel rendezvous),and analyze the typical scenarios(including distributed data collection network and distributed transmission pairs).The main contributions are summarized as follows.We first propose a distributed channel ranking algorithm while considering quality of service and different properties of channels,including bandwidth,signal-to-interference-plus-noise ratio(SINR),coherent bandwidth,coherent time and channel energy consumption.Then,we propose a novel channel allocation algorithm involving reinforced learning.Each self-organized node will only learn from its own past experience to guide the current action.Furthermore,combined with bandwidth adaptation protocol,each node can use more than one channel to transmit at a time and achieve a collision-free channel allocation.On the other hand,the node can adapt its channel switching strategy to the network dynamics,including important aspects such as variation of channel conditions and the arrival or departure of nodes.Our approach maximizes spectrum efficiency and is robust to malicious attacks.The combination of channel allocation and channel ranking can alleviate interference while satisfy the quality of service.In sustainable high-density network,power saving is a severe challenge in energy-constrained communication.We propose a joint channel allocation and time slot optimization solution to prolong the life time of nodes.First,we propose a distributed channel allocation algorithm so that each node can choose a proper channel of lower power consumption considering the different power consumption on each channel.On the other hand,it can enable each node to simultaneously transmit along different channels to alleviate interference and reduce wasteful energy in anti-interference and data retransmission.Second,in time domain,we employ the time slot optimization to save energy.In our proposed scheme,the sleeping time and spectrum sensing time are adaptive and always keep the optimal values.The sleeping duration and spectrum sensing duration are jointly optimized to maximize the normalized throughput and satisfy energy consumption constraints simultaneously.Also,theoretical analysis and extensive simulations have validated that when applying our solution the network throughput is further improved and the power consumption is saved at the same time.Different from the former approaches,our proposed solution requires no central coordination,synchronization or any global information that each node can operate based on its own local information in a total distributed manner.To enable sender-receiver pair's communications in high density network,we propose a distributed channel allocation method with blind rendezvous.The network is operated in a totally distributed manner,neither a central controller nor a common control channel is required,moreover,multiple channels need to be allocated to a sender-receiver pair instead of individual nodes to achieve a collision-free multi-channel allocation.Also the channel allocation can consume less time in channel access and be robust to new entrant and even malicious nodes.Then,we investigate a blind rendezvous mechanism to achieve sender rendezvous with the receiver on the allocated channel.We also present the convergence property and throughput analysis based on Markov model.We combine a distributed channel allocation method with blind rendezvous to take the advantage of channel allocation in reducing interference and blind rendezvous in investigating a communication channel.Therefore,the work reported can improve the network adaptive capability and be robust to dynamic environment.