Search

In this chapter, the non -polynomial (NP) -hard quadratic optimization problem under the constant modulus and similarity constraints is considered. Low -complexity algorithms for both the continuous and finite alphabet phase cases are developed. Specifically, an iterative algorithm for the continuous phase case referred to as iterative algorithm for continuous phase case (IA -CPC) is introduced to successively optimize the objective function with closed -form solutions and its convergence is proven analytically. The computational complexity of IA -CPC is only linear in the number of iterations and polynomial with the size of the design variables. Moreover, for the discrete phase case, the design variables are divided into K blocks. In each iteration, the exhaustive search is leveraged to sequentially optimize each block assuming the remaining K - 1 blocks are fixed. The computational burden is related to the size of the design variables, the number of design variables in each block and the number of discrete phases. Finally, the performance of the new techniques in comparison with the existing approaches available in the open literature are assessed by considering radar code design applications. Numerical results are provided to show the effectiveness of the proposed algorithms in terms of the achieved objective function value and computational complexity. More examples related to radar applications can be found.

In this chapter, the authors cope with the joint optimization of fast-time LAF and waveform spectrum. A novel objective function with a weighting factor to trade-off the WISL with respect to range-Doppler sidelobes and the spectral stopband energy is proposed. The energy and PAR constraints are forced on the designed signal, which further generalize the problem we developed. To handle the resulting non-convex quadratic problem, a new iterative sequential quadratic optimization (ISQO) algorithm is proposed [30]. In each iteration, it transforms the optimization problem into a tractable quadratic subproblem that is further converted to a linear optimization problem with a closed-form solution by resorting to the first-order Taylor expansion. The proposed algorithm has polynomial time complexity that is linear with the number of iterations and polynomial with the size of the designed waveform and the number of the considered range-Doppler bins. The simulation results highlight the superiority of the proposed algorithm contrast to AISO algorithm in terms of the objective value, computational time, and the capability of prohibiting the sidelobes of a strong return and narrowband spectral interferences from masking weak targets. The experimental results show that the hardware limitation influences the level suppressed in both energy spectral density (ESD) and WISL and confirms the effectiveness of spectral coexistence in prohibiting the sidelobes suppressed in the LAF rising.

For multi-input multi-output (MIMO) radar system, the orthogonal waveform set plays important rule to the system's performance. This paper considers the design problem of orthogonal phase-modulated waveform set with low autoand cross-correlation side-lobes in a specified lag interval. After formulating a multi-dimensional optimization problem, we propose a new iterative approach that based on the pattern search (PS) method. Capabilities of the PS algorithm and the WeCAN are evaluated via numerical simulations. Simulation results indicate that the new method has better performance with respect to the side-lobe level in the specified lag interval and less computation load compared to the WeCAN algorithm.

We address the optimization placement problem of distributed MIMO radar in this paper. A coverage ratio is proposed as a criterion to evaluate the surveillance performance of radar system. Due to the complexity of the objective function of the problem and the challenge of the high dimensionality, we propose a solution that contains two parts: a low-complexity method to simplify the objective function, and a placement algorithm based on particle swarm optimization (PSO) to solve the high-dimensional problem. Simulation results show the reasonability of the proposed problem and the effectiveness of our solution. Our contributions includes a novel optimization placement model of distributed MIMO radar and an effective solution to the issue.

This chapter has considered classic radar detection in the presence of white Gaussian noise. The proposed approach first performs a compression of the observation space through reduction by sufficiency. Then detector design is attacked in the reduced dimensional space. The optimum NP detector is unfortunately not implementable for both the coherent and the noncoherent pulse trains. Thus, the robust GLRT-based approach is exploited to design practically implementable receivers that are canonical, namely independent of any prior assigned to the unknown parameters. This construction, depending on the coherency of the train, leads to either the classic coherent radar detector or linear and square-law integrators. Hence, a performance analysis of the obtained detection structures is conducted through analytic equations, where possible, or exploiting numerical simulations.

In this paper, we consider the problem of the multiple targets imaging behind the walls with a through wall imaging radar, where a target image might overlap with the multipath ghosts introduced by some other targets. We propose a sequential determination based algorithm to delete the multipath ghosts while keeping the true target images when the overlapping occurred. Specifically, first, we develop a through wall imaging model by addressing on the first-order side wall multipath propagation. Second, we obtain the images of the true targets and possibly the multipath ghosts. Then, we exploit all the locations of these images to form a sequence and judge if they are the multipath ghosts or not in a sequential way. Finally, we delete the marked multipath ghost images. Some simulations are provided to evaluate the performance of the proposed algorithm.

In sensor networks, resource management is very important for both improving the performance of the system and decreasing the expenses to accomplish the tasks of the networks. Efficient resource management can help us to control the networks appropriately and quickly. In this paper, we consider the problem of adaptive resource management based on target priority in the radar networks. We first propose a novel method to attribute priority to different targets in the radar networks. Second we propose a novel adaptive resource management method in which we adjust the quality of tracking performance adaptively based on the target priority and select the minimum number of radars to satisfy the tracking quality requirement. We select the radars based on the principle that the measurement error are as low as possible. By considering the correlation between the different measurements from the selected radars, we adopt Kalman filter and Covariance-Intersection (CI) information fusion to track the targets. Numerical simulations have verified that the proposed method to attribute target priority and the adaptive resource management are both efficient in radar networks.

We consider the problem of covariance matrix estimation in heterogeneous environments for radar signal processing applications, where the secondary data exhibit heterogeneity in local power and share the same covariance structure. Without any knowledge of the statistical characterization of the sample support, a class of estimators, delined as the geometric barycenters of some basic covariance matrix estimates obtained from the available secondary data set, are proposed by exploiting the characteristics of the positive-delinite matrix space. Finally, we evaluate the performance in terms of detection properties of an adaptive matched lilter (ANMF) with the proposed estimators in the presence of compound-Gaussian disturbance.

The α-μ distribution, a generalization of the Gamma distribution, includes the Chi-squared, Swerling-Chi, Exponential, Weibull and Gamma distributions as special cases, and is adopted to model the fluctuation of the target radar cross section (RCS) in this paper. We proof the capacity of the new target model using the real data collected by the McMaster IPIX radar in the sense of minimum of Cramer-Von Mises distance. The results highlight that the α-μ distribution can describe the real data better than conventional models.

This paper considers the coherent integration problem for detecting a high speed maneuvering target, involving range migration (RM) and Doppler frequency migration (DFM) within one coherent pulse interval. An efficient coherent integration method based on keystone transform (KT), fold factor phase compensation (FFPC) and fractional Fourier transform (FRFT), i.e., KT-FFPC-FRFT, is proposed. It can not only correct the RM effect by KT and FFPC, but also remove the DFM effect and achieve the coherent integration via selecting proper rotation angle. Compared with Radon-fractional Fourier transform (RFRFT) algorithm, the computational cost of the presented method is reduced significantly under the premise that the two methods have same coherent integration detection performance. Finally, several simulations are provided to demonstrate the effectiveness.

In radar networks, radar selection has played a more and more important role. Many radar selection methods have been proposed. Recently a novel method which is based on sparsity-aware matrix decomposition has been proposed to identify informative sensors for target tracking. However, this method had stringent constraints and was under the assumption of known noise parameters in the sensing field. In this paper, we first use radar measurement model to deploy the method in radar networks. Second, we consider noise with unknown parameters existing in the sensing field. Then, we propose two methods to select the informative radars based on setting a threshold for the decomposed matrix and adjusting the parameters in the matrix deflation algorithm, respectively. Finally, we verify our proposals via numerical simulations.