Roadside Radar Network Deployment and Parameter Optimization in Road Environments

To enable the intelligent transportation systems (ITSs), using radars to help extracting the information of road environments is critical. However, the deployment and parameter optimization of radar networks in practical road environments has not been well-explored yet. To fill this gap, we investigate the joint deployment and parameter optimization approach for radar networks in road environments. Considering a general radar network model, we first propose a model-based approach developed under some simplifications of the general model. Then, following the optimization framework of the model-based approach and with the aid of black-box optimization, we propose a non-model-based approach that can jointly optimize the radar deployment and parameter under the general model without any simplifications. Since conducting the non-model-based approach is time-consuming, we further propose a learning-aided approach to accelerate it. We use realistic simulations to evaluate our proposed approaches. Results show that our approaches can outperform the reference schemes.