【作者】 朱翔；

【导师】 赵彬；

【作者基本信息】 哈尔滨工业大学 ， 信息与通信工程， 2021， 硕士

【摘要】 感知,是近些年来无线技术领域的一个热门话题。伴随着无线电技术向着物理世界不断延伸的趋势,人类的生活开始变得充满了无数种可能。Wi-Fi作为一种室内最常见的通信信号,也随着这种趋势在悄然的变革中。下一代IEEE802.11标准中,用其作为室内感知信号是一个重要的方向。一方面工业界考虑在标准的制定上为Wi-Fi信号开辟更大的带宽。另一方面,学术界也在思考如何提出有效而合理的解决方案。从目前来看,想要实现室内Wi-Fi感知面临着三大挑战:1.Wi-Fi设备收发分置,存在非相参缺陷;2.室内条件下静态杂波干扰严重;3.现有体制下带宽受限,距离分辨能力不足。因此本论文着眼于这三个问题,研究克服这些缺陷的方案和算法,并基于通用软件无线电平台设计实验,用实测结果说明其可行性。具体来说,本文的核心内容包括:(1)收发非相参噪声模型的建立,对于任意一对通信设备来说,发射机和接收机之间都有各自的本振和时钟,二者之间往往存在着较大的频偏,并且这种频偏还表现为随机特性。另一方面,通信的接收机的处理流程决定了会引入一定的定时同步误差,这些同步误差将对获得有效的目标信息是有害的。本文根据这些同步误差来源,建立了完整的相位噪声模型,并加以消除。(2)相位噪声抵消和静态杂波抑制,室内静态干扰问题,对于雷达领域来说本质上是动目标检测问题。根据被动雷达的技术体制,Wi-Fi中同样可以构造一个接收通道为参考通道,其他通道为检测通道。在根据二者的不同,在近场和远场两种条件下分别设计出合理的处理方案。远场条件下利用共轭相乘和带通滤波实现相位噪声抵消和静态杂波抑制,近场条件下设计一种复平面相位跟踪的方式实现。(3)对于带宽受限问题,本文中主要围绕两种解决方案展开。一是使用超分辨算法,对实测数据进行处理,再结合目标的角度测量结果,提高定位的精度,另外一方面可以考虑利用多个固定节点观测到的运动目标的其他信息进行轨迹跟踪,来弥补单个节点测距能力的不足。更多还原

【Abstract】 Sensing has been a hot topic in the field of wireless techno logy in recent years.With the continuous extension of radio technology to the physical world,human life has begun to become full of countless possibilities.As the most common indoor signal,Wi-Fi is also undergoing quiet changes following this trend.In the next-generation IEEE 802.11 standard,indoor perception is an important direction of exploration.On the one hand,the industry is considering opening up larger bandwidths for Wi-Fi signals and better detection performance in the formulation of labels.On the other hand,the academic community is also thinking about how to propose effective and reasonable solutions.From the current point of view,the realization of indoor Wi-Fi perception is faced with three major challenges: 1.Wi-Fi devices are sent and received separately,which has non-coherent defects;2.Static clutter interference is serious under indoor conditions;3.Existing Under the system,the bandwidth is limited and the distance resolution capability is insufficient.Therefore,this thesis focuses on these three problems,researches solutions and algorithms to overcome these shortcomings,and designs experiments based on a general software radio platform,and uses actual measurement results to illustrate its feasibility.Specifically,the core content and results of this article include:(1)The establishment of the transceiver non-coherent noise model.For any pair of communication devices,the transmitter and receiver have their own local oscillators and clocks,and there is often a large frequency deviation between the two.And this frequency deviation also shows random characteristics.On the other hand,the processing flow of the communication receiver determines that certain timing synchronization errors will be introduced.If these s ynchronization errors are not considered,effective target information will not be obtained.Based on these sources of synchronization errors,this paper establishes a complete phase noise model.(2)Phase noise cancellation and static clutter suppression,indoor static interference problems are essentially moving target detection problems for the radar field.According to the technical system of passive radar,in Wi-Fi,a receiving channel can also be constructed as a reference channel,and the other chann els can be used as a detection channel.According to the difference between the two,a reasonable treatment plan is designed under the two conditions of the near field and the far field.Under far-field conditions,conjugate multiplication and band-pass filtering are used to achieve phase noise cancellation and static clutter suppression.Under near-field conditions,a complex plane phase tracking method is designed.(3)For the problem of limited bandwidth,this article mainly focuses on two solutions.One is to use the super-resolution algorithm to process the measured data,and then combine the angle measurement results of the target to improve the accuracy of positioning.On the other hand,you can consider using other information of the moving target observed by multiple fixed nodes for trajectory tracking to compensate Insufficient ranging capability of a single node.更多还原