The superposition of various application data streams in smart cities can intensify the load of vehicular networks in intelligent transportation systems, which can have an impact on the popularity of smart cities. To improve the performance of large amount of data transmission in telematics, this study proposes a scheme to determine the network state using congestion parameters and routing parameters and matches different data transmission amounts according to different states of the network. The scheme first considers the possibility of the network congestion state. Once congestion is judged to occur, the amount of data sent is reduced, and reducing the data backlog can further mitigate the possibility of congestion formation on the network. Secondly, after rejecting the possibility of congestion, the routing situation of the network needs to be judged, in such cases, whether a change in vehicles in the multi-hop path leads to a path change or an interruption of the data transmission path. Congestion parameters and routing parameters evaluate the state of the network, and the size of the congestion window is appropriately limited by the routing parameters to prevent excessive data volume from causing backlogs in the vehicular network. Experimental simulations show that the proposed scheme exhibits good performance in both linear and crossover vehicular networking scenarios. The research results provide a useful reference for the data transmission of telematics in smart cities.

In the research of the natural ecological environment, environmental scattering is an important research content whose results are widely used. In order to fully study the relevant characteristics of the natural grassland environment, this paper established a grassland broadband clutter model, summarized the dominant and recessive laws of the grassland environment broadband clutter, and provided a new perspective for related environmental monitoring which would be helpful to solve the problem of target detection in complex environments. The main work of this paper is as follows: The dielectric constant model of the grass was established, and the curve of the dielectric constant with frequency was obtained. The Monte Carlo method combined with the Gaussian spectrum function was used to generate a two-dimensional Gaussian surface to simulate the actual grass surface. A broadband clutter model was established, and, considering the calculation efficiency and accuracy, the Sparse Matrix/Canonical Grid (SMCG) was used to calculate the surface scattering coefficient. Then, the environmental clutter amplitude with different radar bandwidth (10, 40, 80 MHz), surface roughness (h = 0.1 m, h = 0.2 m, h = 0.4 m) and grazing angle (30°, 60°) were calculated and the probability density function (PDF) was obtained. The results show that the higher the radar resolution, the larger the incident angle, and the rougher the interface, all of which will cause the tail of the PDF to become longer which will not only reduce the detection probability of the radar, but also the tracking accuracy. The research results can be used for civilian and military field.