The targets that either have low radar cross-section typically, or their return signal has been deliberately reduced are referred to as weak targets in radar terminology. There are several algorithms for detection of a weak moving target. When such a target is in the vicinity of a large target, the side lobes of the matched filter output due to the large target mask or hide the weak target. The adaptive pulse compression filter that uses the RMMSE estimator has the ability to detect the masked weak target. However, there are at least three factors (computational load, Doppler robustness and pulse eclipsing) which limit the practical application of RMMSE. In this paper, an optimized and integrated algorithm based on adaptive post-processing is proposed to detect targets and to overcome the challenges of RMMSE in electronic defense systems. The FFL-APCR proposed algorithm when compared qualitatively to other algorithms indicates better performance for different SNRs and various target velocities, showing that it is more suitable for implementation in real-time systems. The FFL-APCR algorithm can detect high speed and pulse eclipsed weak targets with lower computational load.

In this paper‎, ‎Bayesian and expected Bayesian (E-Bayesian) estimation of premium with exponential claim size is considered‎. ‎After calculation of premium using Heilmann principal‎, ‎its Bayesian and E-Bayesian estimation under three priors of hyper-parameters are obtained‎. ‎Then‎, ‎formulas for E-MSE are provided‎. ‎A Monte Carlo simulation is conducted for comparison of E-Bayesian estimators of premium‎. ‎A real example is presented for illustration of the results‎.

In this paper, we investigate the detection of masked weak moving targets in the adjacent fast strong target by using FAPC algorithm. The matched filter of conventional pulse compression radars induces range sidelobes in surrounding a target with high SNR that could mask the smaller targets. The mismatch created in received signal by Doppler phase shift, degrades APC filter performance in side lobe suppression. In this paper, the FFL-FAPC algorithm is proposed to reduce the range side lobes using the RMMSE estimator in its post-processing method. In various scenarios, we will investigate the detection of masked targets in comparsion with previous methods. Simulation results show that the FFL-FAPC algorithm reduce significantly of computational cost, in addition to provides improved Doppler robustness.

While the Hammerstein expression is computationally attractive for modeling of nonlinear systems, the optimal calculation of filter coefficients is practically cumbersome. A proper solution to this problem is the use of adaptive algorithms. In this paper, the Hammerstein Normalized LMS algorithm is proposed by deriving the corresponding time varying optimal step-size parameter in a closed form. The convergence behavior of this algorithm is inspected using computer simulations. The results show that the proposed algorithm achieves a faster convergence speed compared to the Hammerstein LMS algorithm, practically at the cost of an acceptable increase in computations.

This paper introduces a meter of the root-Mean-Square value of deterministic and stochastic signals of an arbitrary shape that are generated over the set time interval. Such a meter involves only the Minimum number of simple arithmetic operations to obtain results, and it ensures a high degree of measurement accuracy. For this purpose, the direct calculation of the signal root-Mean-Square value is applied while the measurement of the half-period average straightened signal value is carried out by Means of the traditional measurement devices. Implementation of this meter requires neither the knowledge of what the signal period is, nor the synchronization with the processed sampling. Simulation is then carried out demonstrating the high efficiency of the proposed measurement algorithm. We analyze the characteristics of the meter operating within a wide frequency range of the measurable signals. The recommendations concerning the hardware implementation of such a meter by Means of the field programmable gate arrays are considered. The meter can be used when designing digital high-frequency AC voltmeters and ammeters and it can provide the readings that do not depend upon the signal waveform.

In this paper a low noise amplifier (LNA) in the frequency interval of [3-5] GHz is designed and optimized. The design process consists of three parts. At first, the stability of the bipolar junction transistor (BJT) is investigated. Simulation results show that it is unconditionally stable in the whole frequency interval. In the second part, the input and output matching networks are constructed in such a way that the desired transducer power gain (GT=10dB) and possible Minimum noise figure (Fn) are achieved simultaneously. In order to achieve approximately constant GT and Fn in the desired frequency interval, the input and output matching networks are finally optimized using least Mean Square Error technique. Once the optimization process is converged, variations of GT and Fn in the whole frequency interval are respectively less than 1dB and 0.5 dB.  Besides, the variation of output phase is approximately linear.

This paper presents new bounds for the left and right fractional Mean-Square stochastic integrals based on convex stochastic processes. Then a range is proposed that includes a linear combination of the left and right fractional Mean-Square stochastic integrals. Finally, the previous results presented in this subject are improved.