An Optimization Algorithm-Based Coherence Method for the DCAR Tracking Process

Distributed coherent aperture radar (DCAR) is a critical advancement in next-generation radar technology. Traditional DCAR relies on quadrature signals to gather target parameters before transmitting coherent signals, facing the challenge of time synchronization and parameter acquisition accuracy. This article tackles the challenges by transforming the coherence problem into an optimization issue, achieved by quantifying the degree of coherence of the echo signal. We employ an optimization algorithm to adjust the transmission time of each node, thereby ensuring that signals coherently accumulate at the target location. Furthermore, we propose a smoothing optimization technique to counter the fitness function's “multipeak” and “black box” attributes. By employing the barycentric interpolation to construct a pseudofunction, this technique facilitates the smoothing of the fitness function, speeds up convergence, and minimizes the probability of falling in local optima. For scenarios involving moving targets, a Kalman filter is embedded within the optimization algorithm to facilitate target motion prediction and preemptively resolve temporal errors in parameter acquisition, thereby enhancing real-time tracking capabilities. Theoretical analyses and simulation results demonstrate the efficacy of the proposed method in facilitating coherent target tracking. This method improves the signal-to-noise ratio while reducing the need for precise carrier synchronization. In addition, the proposed smoothing optimization technique accelerates convergence rates, reduces the possibility of convergence to a local optimum, and can be applied to optimization problems with similar properties.