This paper proposes stable virtual sensing algorithm for active noise control with sequential online modeling of the auxiliary filter and the secondary path. The two online modelings prevents system divergence and maintain high noise reduction when there is a secondary path change such as microphone relocation. The online modeling of the secondary path adjusts mismatch in the noise-control filter. The online modeling of the auxiliary filter compensates for the mismatch therein caused by the secondary path change. A simulation result with recorded signals shows that the proposed method maintains a noise reduction of 22 dB even when a secondary-path change takes place.

This paper presents a continuous-time pulse compressor (CTPC) for ultrasound imaging systems with enhanced signal intensity and axial resolution. Conventional pulse compression using a chirp signal employs matched filtering, which performs cross-correlation calculations in the digital domain. However, the calculation requires a large amount of computing resources. The proposed CTPC performs pulse compression directly in the analog domain by frequency-dependent delay. In the proposed CTPC, a second-order all-pass filter (APF) with a relatively high quality factor (Q) of 1.8 achieves a long group delay (GD) to reach a sufficiently long chirp period in the ultrasonic band. Furthermore, cascading other second-order APFs with different resonance frequencies approximately linearizes the total GD with respect to frequency. We investigated the implementation of the proposed CTPC using discrete operational amplifiers and passive components. Measurements confirmed a 6.7 dB signal enhancement from compression effects with an ideal electrical chirp signal. Further acoustic measurements using an 8×8 channel 2-D matrix array probe and an imaging prototype demonstrated improved signal enhancement and compression factor of 3.2.

Defect pattern detection in wafer bin maps (WBMs) is crucial for enhancing wafer quality, as it prevents the escalation of defects and the squandering of resources. To address this, we introduce a quantum-based Support Vector Machine (SVM) leveraging the quantum approximate optimization algorithm, termed QAOA_SVM. We employ Inception V3 to extract efficient and compact features from WBMs and apply QAOA_SVM for training and testing the data. When recognizing WBMs with two mixed defect types, our method outperforms the original mixup approach by over 7.4%, and achieves a 0.7% accuracy improvement compared to the InceptionV3+SVM method. For mixed defect samples containing more than two types of defects (three-mixed and four-mixed), we observe gains of at least 4.2% (relative to SVM) and 0.9% (relative to InceptionV3+SVM), respectively. Additionally, our method surpasses other state-of-the-art Convolutional Neural Network (CNN) methods in both single-type and mixed-type defect pattern recognition. Furthermore, QAOA_SVM requires only O(Nd) time, which is significantly more efficient than the O(N³) time complexity of traditional SVM. In summary, QAOA_SVM achieves higher accuracy with significantly reduced computational time.

The covering radius of the r-th order Reed-Muller code RM(r, n), denoted by ρ(r, n), is the maximum r-th order nonlinearity of n-variable Boolean functions. Using the Fourquet-Tavernier list-decoding algorithm and the Fourquet list-decoding algorithm, we discover, among monomial Boolean functions, 11-variable Boolean functions with second-order nonlinearity 856, and we determine that the covering radius of RM(3, 8) in RM(4, 8) is 56. Besides, it is proved that the complexity of the Fourquet algorithm for list decoding RM(r, n) is linear in the length of the code 2ⁿ given the decoding radius up to the Johnson bound. In this paper, we prove that the complexity of the Fourquet algorithm is also linear in 2ⁿ in some special cases when the decoding radius is close to 2^n-r. Moreover, following from the Carlet’s method, we improve the best proven lower bound on the third-order nonlinearity of monomial Boolean functions. In a word, the original idea of our work is to improve the lower bound on ρ(r, n) according to two categories as follows: for small r and n, we search an n-variable Boolean function with larger r-th order nonlinearity using a list-decoding algorithm for Reed-Muller codes; for large n, we study a class of quartic monomial Boolean functions to improve the best proven lower bound on its third-order nonlinearity.

Secure multiparty computation (MPC) is a cryptographic technology to perform some computation on multiple parties’ input data while concealing the individual inputs from other parties. For the case of semi-honest adversaries, the security definition in Goldreich’s famous book is widely used as a standard definition. In this paper, however, we point out that there is an MPC protocol where a semi-honest adversary receives only a ciphertext of one-time pad with unknown key but the protocol is not secure under the standard security definition, which may look inconsistent with the perfect secrecy of one-time pad that its ciphertext leaks no information at all. We propose a variant of the security definition that resolves this issue. We also show that a somewhat restrictive version of the Composition Theorem holds for our modified security definition.

This letter proposes an ANC system without an error microphone introducing primary path estimation. The proposed system enables adaptively update of a noise control filter to realize the noise reduction at a desired position without placing an error microphone at the target point, by estimating the primary path to the desired position. A computer simulation using impulse responses generated by the image method demonstrates that the proposed system achieves a noise reduction of more than 10 dB at the desired position, even when the noise source moves.

The distributed space-based coherent aperture radar (DSCAR) is a radar system that utilizes a space-based platform to form a formation of radar unit (RU) for joint target detection, inspired by the concept of distributed coherent radar. However, DSCAR based on uniform linear formation (ULF) suffers from numerous grating lobe which severely impair the beamforming performance. Due to the fact that conventional grating lobe suppression method based on non-uniform spacing is not suitable for uniform formation, this letter proposes a method based on randomized angle yaw to address the issue of grating lobe suppression in DSCAR. We present the formula for the DSCAR joint pattern and use particle swarm optimization (PSO) to optimize the peak side lobe level (PSLL). Simulation results demonstrate that randomized angle yaw method has better grating suppression effect than non-uniform spacing method, and increasing the number of RUs and the number of RU antenna elements will further improve the optimal PSLL.

To meet the key performance requirements of permanent magnet synchronous motor (PMSM) speed control systems — such as high-precision control, fast response, and strong anti-disturbance, a improved power exponential reaching law (IPERL) is proposed. This approach incorporates an inverse cotangent function to adaptively adjust the power term coefficient, effectively balancing the trade-off between convergence speed and sliding mode chattering. Additionally, a dynamic linear sliding mode surface is designed based on the system error, which effectively mitigates overshoot and further enhances both the dynamic and steady-state performance of the system. Finally, the effectiveness of the proposed control strategy is validated through simulations, demonstrating superior stability and convergence speed compared to conventional methods.

In this letter, we consider an asymptotic stabilization problem for a system under saturated input which has uncertain and time-varying bounds. With our gain-scaling controller, we analytically show that the controlled system is asymptotically stabilized and the domain of operation can be arbitrarily enlarged by increasing the gain-scaling factor. Then, our proposed method is extended to a class of feedforward systems. Two numerical results are provided to show the validity of our results.

This work studies the fundamental properties of self-learning observers (LOs), which simultaneously estimate states and parameters resource-efficiently. LOs have a simple structure, the capacity to estimate system uncertainties using only one algebraic equation, and decent performance. We explore the exponential convergence property of LOs in-depth and present an explicit exponential convergence rate for the first time using Halanay’s inequality technique. This work further contributes by providing fewer conservative conditions, thereby decreasing the equality condition that must be satisfied in previous studies on LOs. The LO parameters are acquired by solving linear matrix inequalities (LMIs), and the rules for parameter tuning under the new constraints are provided.

This letter analyzes the average channel capacity of coherent free-space optical (FSO) systems employing multiple receive aperture for binary inputs and its simple upper bound over atmospheric turbulence-induced channels. The newly derived upper bound is based on the Taylor series and moment generating function, and it significantly reduces computational complexity by involving only a single infinite series, in contrast to the conventional double-integral expression. This analytical simplicity makes it especially appealing for performance evaluation and system design. Through numerical examples, we validate that the proposed upper bound closely approximates the exact average channel capacity across a wide range of receiver configurations and refractive index structure constants.