Road traffic congestion, which occurs when the traffic demand exceeds the operational capacity of the roadway, has become a serious problem all over the world. Traffic signals are effective in managing the traffic demand if properly designed. Traditionally, signal design has been carried out based on the objective of minimizing the overall queue and associated delays at upstream sections and has not considered the effect on downstream sections and midblock regions. Efficiency of signals may be improved if their design takes this factor into account, which is attempted in the present study. In this paper, a control scheme that regulates the traffic density in the downstream section was developed using a macroscopic-model-based approach with traffic density and speed as the state variables of interest. These state variables were estimated based on a Kalman filter approach. The performance of the estimation scheme was corroborated using field data. A closed-loop system with the developed control scheme was implemented in a simulated road stretch developed by integrating microscopic traffic-simulation software and mathematical simulation software. The results from the implementation of the closed-loop system demonstrated the effectiveness of the developed traffic-control scheme. It was observed that the objective of maintaining the density inside the section within the required limit was satisfied. © 2016 American Society of Civil Engineers.