Focused on the issue that Point-Of-Interest (POI) recommendation has low recommendation accuracy and efficiency, with deep analysis of the influence of social factors and geographical factors in POI recommendation, a POI recommendation algorithm combining location influence was presented. Firstly, in order to solve the sparseness of sign-in data, the 2-degree friends were introduced into the collaborative filtering algorithm to construct a social influence model, and the social influence of the 2-degree friends on the users were obtained by calculating experience and friend similarity. Secondly, by deep consideration of the influence of geographical factors on POI, a location influence model was constructed based on the analysis of social networks. The users' influences were discovered through the PageRank algorithm, and the location influences were calculated by the POI sign-in frequency, obtaining overall geographical preference. Moreover, kernel density estimation method was used to model the users' sign-in behaviors and obtain the personalized geographical features. Finally, the social model and the geographic model were combined to improve the recommendation accuracy, and the recommendation efficiency was improved by constructing the candidate POI recommendation set. Experiments on Gowalla and Yelp sign-in datasets show that the proposed algorithm can quickly recommend POIs for users, and has high accuracy and recall rate than Location Recommendation with Temporal effects (LRT) algorithm and iGSLR (Personalized Geo-Social Location Recommendation) algorithm.

Focused on the issue that the existing algorithms are unable to accurately detect active hotspots from trajectory data, a novel Trajectory Structure-based Hotspots discovery (TS_HS) algorithm was proposed. TS_HS consisted of the following two algorithms:Candidate Hotspots Discovery (CHSD) algorithm and Hotspots Filter (HSF) algorithm. First, trajectory dense regions were detected by the grid based clustering method CHSD as candidate hotspots. Second, the active hotspots region of moving objects were filtered by using HSF algorithm according to moving feature and time-varying characteristic of trajectories. The experiments on the Geolife dataset show that TS_HS is an effective solution for multi-density active hotspot problem, compared with Global Density threshold based Hot Region discovery (GD_HR) and Spatio-temporal Hot Spot Region Discovering (SDHSRD). The simulation results show that the proposed framework can detect active hotspots effectively based on the structure feature and time-varying characteristic of trajectory.

The current wireless positioning technology can not adapt to complex environment and has low positioning accuracy. In order to solve the problems, a mobile terminal positioning method driven by road test data was proposed. Firstly, based on the location algorithm of base station and the description algorithm of base station signal coverage, the location-coverage model of base station base was established. By matching the initial parameters of the mobile terminal with the model base, the initial range of the mobile terminal was obtained. Secondly, the road classification database was established based on the extraction algorithm of road feature, and the wireless signal feature matching algorithm was used to match the road information of the mobile terminal. Finally, the model base of longitude-latitude and intensity mapping was established and the precise position of the mobile terminal was determined by using the terminal signal comparison algorithm. The theoretical analysis and experimental results show that the probability of 2 m localization accuracy of the base station reaches 60%, the probability of 3 m reaches 77%, which are improved respectively by about 39% and 12% than those before whitening, and the description algorithm of base station signal coverage can also describe the coverage of base station signal more accurately. The accuracy improvement of the two parts can improve the final positioning accuracy.

Focusing on low prediction accuracy of the low-order Markov model and high sparsity rate of the high-order Markov model, a moving object location prediction algorithm based on Markov Model and Trajectory Similarity (MMTS) was proposed. The moving object's historical trajectory was modeled by using Markov thinking, and trajectory similarity was acted as an important factor of location prediction. With the result set predicted by Markov model as candidate set, the trajectory similarity factor was combined to get the final prediction. The experimental results show that, compared with the k-order Markov model, the predictive capability of the MMTS method is not greatly affected with the change of training sample size and the value of k, and the average accuracy is improved by more than 8% while significantly reducing the sparsity rate of k-order Markov model. So, the proposed method not only solves the problem of high sparsity rate and low prediction accuracy of the k-order Markov model, but also improves the stability of prediction.