By assuming signal propagation speed to be unknown, a convex rank unconstrained semidefinite programming (RUSDP) algorithm is designed to obtain the unified solution for near-field and far-field TDOA-based localization. Then the rank constrained semidefinite programming (RCSDP) is further proposed by availing of the rank constraint. The simulations show the validity of the RUSDP and RCSDP algorithms whether the source is in the near-field or in the far-field. The RCSDP performs significantly better than the RUSDP due to the tighter relaxation.

Two efficient solutions are proposed for the localization problem of an object in multiple-input multiple-output (MIMO) radar systems, when the transmitter positions and offsets are unknown. The localization problem is first recast into a convex form by applying the semidefinite relaxation (SDR) technique, the solution of which converges to global optimum. We also propose a closed-form solution warranting global convergence, in which the object position is estimated by two stages. In the stage-one solution, the auxiliary variables are introduced to transform the nonlinear problem into a linear form. The stage-two solution is further designed to refine the estimates obtained from the stage-one solution. The minimum number of receivers and the complexity are also analyzed for the proposed solutions. The simulated results show that the SDR and closed-form solutions provide good estimates for the object position and transmitter positions. Their performance can sufficiently approach the Cramér-Rao Lower Bound (CRLB) accuracy at the high signal-to-noise ratio (SNR).

Distributed Denial-of-Service (DDoS) attacks pose a serious threat to the stability and security of edge intelligence devices. To solve this issue, we first describe the cost of edge intelligence environments in detail and introduce the mean-field term, edge intelligence repair speed, and DDoS attack intensity, which provide a theoretical basis for the subsequent model construction. Secondly, based on Hamilton-Jacobi-Bellman (HJB) backward and Fokker-Planck-Kolmogorov (FPK) forward equations, a mean-field model is proposed. The optimal DDoS defense policy is then solved considering the strongest DDoS attack intensity and the best edge intelligence device repair speed. Further, the focus is turned to the interaction between DDoS attackers and the edge intelligence environment. We give the update rule of the mean-field term and construct a mean-field game (MFG) model with a value function. Finally, we propose a multi-agent deep reinforcement learning algorithm called Mean-Field Games with Dueling Double Deep Q-learning Network (MFGD3QN) to solve the optimal DDoS defense policy problem under the MFG model. In the experiment, we compare MFGD3QN with several benchmark algorithms and verify the superiority of the MFGD3QN algorithm in an edge intelligence environment. We also carry out experiments on different parameters of the MFGD3QN algorithm, lower visibility conditions of the edge intelligence environment, and different repair speeds of edge intelligence devices, which verify the robustness and feasibility of the algorithm.

With the fast development of new array technology and intelligent antenna, it is easier to obtain angle of arrival (AOA) measurements. Hybrid received signal strength (RSS) and AOA measurement techniques are proposed for the position computing in sensor networks. By converting the measurement equations and relaxing the optimization function, range-based square semidefinite programming (RLS-SDP) and squared range-based square semidefinite programming (SRLS-SDP) algorithms are put forward to obtain the source position estimate by considering the transmit power to be known or unknown. The proposed RLS-SDP and SRLS-SDP algorithms provide accurate solution to the source position estimate and avoid the initialization process of numerical calculation. The simulations show that the proposed RLS-SDP and SRLS-SDP algorithms perform better than the linear estimator and provide the accuracy performance which is very close to the Cramér-Rao Lower Bound (CRLB) of position estimation. The proposed SRLS-SDP algorithm shows its advantages in the computational complexity compared with the RLS-SDP, since the complexity of SRLS-SDP is independent of the number of anchor nodes.