To avoid the failure of data forwarding, brought by some characteristics of Vehicular Ad Hoc Networks (VANET), uniform distribution of vehicles, frequent network partition and mergence, etc., a new data delivery method based on Weak State Routing (WSR) from Traffic Control Center (TCC) to driving vehicles, called Weak State Forwarding (WSFD), was introduced in VANET. Firstly, a data packet collected by TCC was delivered to an Access Point (AP) along the direction of the destination vehicle. Secondly, the data packet was forwarded to the destination vehicle by AP within its communication range, at the same time, the location information of destination vehicle was carried by the data packet. Then, after comparing all the mapping information owned by the vehicle which received the data packet, the most deterministic map information was chosen by the vehicle and compared to the location information carried by the data packet so as to ensure the next forwarding direction. If the confidential level was quite high, the data packet was revised to move towards the mapping's corresponding central area, meanwhile, the information of destination vehicle carried by the data packet was updated. Otherwise, the original direction would be kept. Lastly, through several times' forwarding and revising, the data packet would be gradually approached to the area where the destination vehicle located, and the whole data delivery would be finally completed. Compared with Trajectory-based Statistical Forwarding for multihop infrastructure-to-vehicle data delivery (TSF) and Greedy Perimeter Stateless Routing (GPSR) algorithm, the WSFD algorithm could reduce the delivery delay to 5 seconds or less and elevate the delivery rate to 0.92 or more generally in the experiment of data transmission in 30 km*30 km square area. The experimental results show that the WSFD algorithm can improve safety of drivers and alleviate the traffic jam effectively.

Hyperspectral image has high spectral dimension, vast data and altitudinal interband redundancy, which brings problems to image classification. To effectively reduce dimensionality and improve classification precision, a new extraction method of nonlinear manifold learning feature based on Supervised Local Linear Embedding (SLLE) for classification of hyperspectral image was proposed in this paper. A data point's k Nearest Neighbours (NN) were found by using new distance function which was proposed according to prior class-label information. Because the intra-class distance is smaller than inter-class distance, classification is easy for SLLE algorithm. The experimental results on hyperspectral datasets and UCI data set demonstrate the effectiveness of the presented method.