Research On End-to-Edge Transmission Protocols In Wide Area Networks

With the rapid development of wireless networks,more and more intelligent terminals join them.Intelligent terminals deliver the obtained information to the core network based on wireless networks for further processing.As a medium for information delivery,wireless networks play an important role.In the physical space,different kinds of wireless networks coexist.For example,In wireless local area networks(WLANs),Zig Bee based wireless networks can be employed for short distance low power sensor networks,while wireless wide area networks(WWANs)based on 4G,5G,and Lo Ra techniques can support data traffic delivery in WANs.However,coupled with the rapid increase of intelligent terminals involved in wireless networks,there exists an explosive growth of data traffic in wireless networks,which poses a great challenge to traditional wireless networks such as the frequent packet collision and long transmission delay.Especially in WWANs,there is a high probability of packet collision due to Wide coverage and non-trivial channel state detection.For instance,in Lo Ra networks,one Lo Ra gateway usually connects with multiple Lo Ra nodes.Consequently,multiple Lo Ra nodes convey the sensed information such as temperature and humidity to the gateway,and then the Lo Ra gateway delivers these information to the core network for further processing.Due to the star network structure adopted in Lo Ra communications,the packet collision case frequently happens,which thus prolongs the transmission delay time coupled with wide coverage range.However,the long delay is unacceptable for delay-sensitive applications.For example,in mobile WWANs,vehicles achieve data traffic delivery with nearby 4G/5G base stations by establishing links with them,in order to satisfy the requirements of vehicular applications.In order to maintain these links,mobile vehicles need to switch between different base stations.Yet,the handover delay in Wi Fi and4 G networks is at the level of hundred milliseconds.With more stringent requirements on the delay,how to reduce the handover delay is of great concern to wireless networks.Another typical problem in mobile scenarios is the limited activity range of mobile terminals due to the transmission power difference between the base station and terminal node.For example,when swarm robots patrol or carry out tasks in outdoor scenarios,the coverage range of swarm robots is restricted due to the small transmission power of single robot.To deal with this problem,existing research work proposes to densely deploy base stations,yet which adds the implementation and maintenance cost of large-scale networks.Consequently,how to enhance the uplink coverage range is critical to the overall performance of wireless networks.With regard to different challenges in WWANs,the research of this paper has been focused on the packet collision problem,the handover problem in 5G mm Wave networks,and the uplink coverage extension based on swarm robots,which mainly consists of the following three aspects.Recently,as one of the wireless techniques designed for low-power wide-area networks,Lo Ra communications have received widespread attentions.Based on the unique physical-layer modulation technique,Lo Ra communications can achieve the aims of low-power and wide-area.In Lo Ra networks,one Lo Ra gateway or base station connects with multiple Lo Ra nodes.Meanwhile,in order to reduce the complexity,Lo Ra communications adopt the ALOHA protocol in MAC layer.Consequently,at the Lo Ra receiver side,the pack collision case frequently happens due to packets from different Lo Ra nodes arriving at the receiver simultaneously.The collision results in the packet reception failure and reduces the network efficiency.To overcome this challenge,we propose a novel packet collision resolution protocol for Lo Ra networks.In this protocol,we propose a preamble detection scheme based on the unique structure of Lo Ra packets.According to this scheme,we can obtain the chirp-level time offset between the collided packets.Meanwhile,we can also obtain the collision-free samples within the chirp-level time offset.Combining the collision-free samples with the chirp spread spectrum feature,we can obtain the other part of collided samples within the same chirp.Subsequently,subtracting the derived samples from added-up samples,we can acquire the corresponding collided samples from the other packet.Similarly,we can derive the samples within the same chirp in this packet,which then constructs a complete chirp.Iteratively,this protocol can resolve the two-packet collision case.According to the two-packet collision resolution process,we employ the collision resolution strategy to the three or more packet collision cases.Finally,we propose some design enhancements for noise mitigation and CFO elimination to further improve the performance of this protocol.Experimental results demonstrate that Lo Ra communications based on the proposed protocol can resolve the two-packet and three-packet collision cases.Secondly,With the advent of 5G Era,5G mm Wave networks recently have become the research focus and gradually been deployed.In 5G mobile scenarios,vehicles will experience frequent handover between mm Wave base stations.Compared to traditional wireless networks,densely deployed mm Wave base stations and high-speed vehicle velocity will result in more frequency handover.In convention,the handover delay is at a level of hundred milliseconds,which is unacceptable for delay-sensitive vehicular applications such as navigation system and anti-collision system.In order to reduce the handover delay between mm Wave base stations,we propose a novel seamless handover protocol based on large-scale mm Wave antenna arrays,since large-scale mm Wave antenna arrays can be integrated in a small physical space due to mm Wave’s short wavelength.First,we devise a partition scheme for large-scale mm Wave antenna arrays,which divides the antenna arrays into two parts.One part is utilized for data transmission with the currently connected base station,and the other part is applied for pre-connection with nearby base stations.When handover is triggered,the operation only needed to be executed is to activate one of the pre-connected links,which thus minimizes the handover delay.Meanwhile,we have proposed a scheme related to how to pre-connect with nearby base stations,in order to reduce the handover times.Since mm Wave links can be easily blocked by obstacles,we also consider the handover problem in this case.When mm Wave links are broken down,the received signal strength at the base station side drops rapidly.When the received signal strength is below a set threshold,the base station will inform of activating one pre-connected link,which can simply complete the handover process and avoid obstacles.Experimental results verify that the handover delay based on the proposed protocol is at a level of microseconds,compared to the hundred milliseconds of handover delay in Wi Fi and LTE networks.Thirdly,with the rapid development of robots,swarm robots gradually appear in the daily life,instead of human resource.Compared to single robot,swarm robots demonstrate a superior performance in terms of collaboration efficiency.However,in communication scenarios between swarm robots and base station,the uplink coverage range is limited due to the power constraint of single robot,which then restricts the activity range of swarm robots.In order to enhance the uplink coverage range,we propose a collaborative communication protocol based on swarm robots.The core idea of this protocol is to achieve synchronization between swarm robots and then perform distributed beamforming.Finally,the uplink coverage range can be extended.In addition,we also deal with the abnormal cases in the process of distributed beamforming in order to improve the robustness of this protocol.Experimental results verify that with a small amount of time overhead,this protocol can greatly improve the coverage range of swarm robots.For example,this protocol can improve the uplink coverage range of swarm robots from less than200 meters to around 1000 meters.