For physical-layer network coding over time-varying bi-directional relay channels, a joint channel coding and non-coherent physical-layer network coded modulation and detection scheme without channel state information was designed in multiple-antenna environment. Firstly, the spatial modulation matrix at the source was designed to achieve physical-layer network coding. Then, a differential spatial modulation was combined with physical-layer network coding and maximum a posteriori probability of superimposed signal was derived at the relay. Moreover, considering the constellation of the superimposed signal, a mapping function to map superimposed signal to broadcasting signal was designed. Lastly, taking advantage of the linear structure of channel coding, and combining bit interleaving, channel decoding and soft-input soft-output detection algorithm, an iterative detection approach for joint channel differential physical-layer network coding was obtained. Simulation results show that the proposed scheme can achieve non-coherent transmission and detection used to physical-layer network coding for two-way relay channels and can effectively enhance the throughput and spectral efficiency of the system.

An automatic segmentation method based on image patch matching strategy was proposed to realize the automatic segmentation of glomerular basement membrane automatically. First of all, according to the characteristics of the glomerular basement membrane, the search range was extended from a reference image to multiple reference images, and an improved searching method was adopted to improve matching efficiency. Then,the optimal patches were searched out and the label image patches corresponding to the optimal patches were extracted, which were weighted by matching similarity. Finally, the weighted label patches were rearranged as the initial segmentation of glomerular basement membrane, from which the final segmentation could be obtained after morphological processing. On the glomerular Transmission Electron Microscopy (TEM) dataset, the Jaccard coefficient is between 83% and 95%. The experimental results show that the proposed method can achieve higher accuracy.

In order to improve the performance of obstacle avoidance for mobile robots in continuous obstacle environment, a fuzzy algorithm of obstacle avoidance with speed feedback was proposed. Ultrasonic sensors were utilized to perceive the surroundings, and based on fuzzy control, the mobile robot adjusted its speed according to the distribution of obstacles. Then the graceful degradation was introduced combined with the improved fuzzy obstacle avoidance to enhance the robustness of the mobile robot. The experimental results show that the method can adjust the speed through interaction with the environment, control the robot in a collision-free way and optimize the obstacle avoidance path. Simultaneously, the method shows good effectiveness.

Concerning the poor positioning rate and localization precision of Grid-Scan algorithm, an improved Grid-Scan localization algorithm based on virtual beacon nodes was proposed. Three related works were mentioned as follows: Firstly, the unknown nodes which have neighbor beacon nodes located themselves by using the beacon nodes, and the located unknown nodes were upgraded to virtual beacon nodes. Secondly, the unknown nodes that do not have neighbor beacon nodes got their location through virtual beacon nodes. Finally, different communication radiuses were set between beacon nodes, virtual beacon nodes and unknown nodes to scan in order to accomplish the localization. The simulation results show that the improved algorithm’s positioning accuracy and positioning rate increase by 6.35% and 23.37% on average respectively.

Focusing on the problems of poor locating performance and ignoring anchor nodes' inaccuracy in traditional DV-Hop algorithm which is one of the range-free algorithms in Wireless Sensor Network (WSN), an intelligent algorithm for locating nodes based on Processing Strategy of Optimal Hopping Distances (PSOHD) was proposed. Fully considering the effect of dynamic topology and anchor nodes' difference, this algorithm firstly introduced two communication radii with anchor nodes to calculate the nodes of communication range respectively. Secondly, the average one-hop distance among anchor nodes was refined by means of weighted least squares estimation. Finally, the average one-hop distance used by each locating node for estimating its location was optimized through weighting the N received average one-hop distances from anchor nodes. In addition, Particle Swarm Optimization (PSO) algorithm was presented to locate the unknown nodes. The simulation results show that the improved algorithm has obviously better location precision at the cost of increasing appropriate energy. Thus, it is a practical scheme for WSN with both inaccuracy of anchor nodes and random dynamic topology.