Recognizing student behaviors in classroom videos remains challenging due to complex backgrounds, frequent occlusions, subtle inter-class motion differences, and temporal jitter in frame-wise predictions. To address these issues, this paper proposes a hybrid student behavior recognition framework that integrates a Multi-branch Spatiotemporal Attention Network (MSTA-Net) with a Behavior State Kalman Filter (BSKF). At the perceptual level, MSTA-Net employs decoupled channel, spatial, and short-term temporal attention branches to enhance discriminative behavioral features while suppressing irrelevant background information. At the cognitive level, BSKF reformulates behavior recognition as a continuous state estimation problem in a high-dimensional probability space, where behavioral inertia is exploited to smooth noisy observations and improve temporal consistency. Experimental results on the SCB-Dataset and real-world classroom video sequences demonstrate that the proposed method achieves an accuracy of 94.7% and a real-time inference speed of 33 FPS. Compared with purely deep learning-based models, the proposed framework reduces the Action Category Switching (ACS) rate by 50%, indicating substantially improved robustness in long-term behavior recognition. These results suggest that coupling attention-based perception with Kalman-based state estimation provides an effective and efficient solution for reliable student behavior analysis in intelligent classroom environments.

Existing single-channel co-frequency signal blind separation (SCSBS) algorithms struggle to balance separation accuracy, computational complexity, and robustness, while current channel state information (CSI) estimation methods lack precision. To address these limitations, we propose a sequential detection (SD)-based iterative separation (SDIS) algorithm. SDIS incorporates a delayed unscented Kalman filter (DUKF) into an iterative decision feedback framework, jointly enhancing signal separation and CSI estimation. Simulation results show that SDIS outperforms benchmarks in separation accuracy, CSI estimation accuracy, computational efficiency, and robustness. Notably, when the mean bit error rate (MBER) drops below 10^-4, SDIS can tolerate at least 0.8 dB more noise than the benchmarks.

Practical separation of single-channel co-frequency signals (SCCFSs) is hindered by the prohibitive computational complexity of benchmark algorithms. To address this issue, we propose a low-complexity separation framework based on sequential detection (SD), in which signal separation is cast as a sequential path search over a trellis. To support both hard-decision detection and log-likelihood ratio (LLR) extraction, we develop two algorithms within this framework: the SD-based separation (SDS) algorithm and its soft-output variant (SO-SDS). Furthermore, SDS employs a windowing strategy combined with dynamic pruning to concentrate computational resources on high-probability paths, thereby enabling efficient detection of transmitted symbol sequences. Building upon SDS, SO-SDS further incorporates a state completeness verification mechanism (SCVM) to estimate bit LLRs, thus facilitating subsequent soft decoding. Numerical results show that, compared to benchmark algorithms, SDS achieves significant complexity reduction without degrading separation performance, while SO-SDS offers notable computational savings with only modest LLR accuracy loss. Notably, the computational complexity advantage of the proposed algorithms over benchmark algorithms grows substantially with increasing modulation order.

In unmanned aerial vehicle (UAV) communication systems, high-order quadrature amplitude modulation (QAM) signals are widely used for their high data rates. However, the complex UAV channel environment reduces the equalizer's processing speed, making it challenging for the equalization algorithm to converge effectively. To address this issue, this paper proposes a dual-cost function fractional spaced blind equalization algorithm. By integrating second-order statistical information with conventional blind equalization methods, the algorithm optimizes the equalizer coefficients at two distinct points within each symbol period of the fractionally spaced equalizer (FSE) using two cost functions. Additionally, a dual-mode algorithm and variable step-size function are introduced, enhancing both convergence stability and speed. Simulation results show that the proposed algorithm adapts well to various UAV communication environments. Compared to traditional blind equalization methods, it significantly improves equalization performance and interference resistance for high-order QAM signals. The steady-state residual inter-symbol interference (ISI) is greatly reduced, and the Bit Error Rate (BER) decreases by approximately one order of magnitude. This demonstrates that the proposed algorithm provides an efficient and effective blind equalization solution for broadband UAV communication.

We propose a high-efficiency , low-complexity turbo product code (TPC) decoder. A criterion is designed to reduce the candidate code set size, retaining the most effective candidate codewords based on their Euclidean distances. A novel method for calculating extrinsic information is proposed to compensate for performance loss due to the reduced candidate code set. The decoder’s top-level architecture is also improved, allocating hardware resources in an interleave-like manner to avoid memory conflicts. Simulation results show that the proposed algorithm achieves significant performance gains and consumes less hardware resources. The decoder’s throughput in the new architecture can reach 1.371 Gbps.

To address the problem of high complexity of channel estimation for orthogonal time frequency space (OTFS) systems, we propose a low complexity channel estimation method leveraging the unitary approximate message passing (UAMP) algorithm. Specifically, the estimation of OTFS channel is represented as a linear mixing model, which is handled using UAMP, avoiding the matrix inversion operations in the state-of-the-art algorithm. Numerical results demonstrate that, with significantly lower complexity, the proposed algorithm can achieve almost the same channel estimation performance, compared to the benchmark algorithm.

Traditional continuous phase modulation (CPM) detection algorithms based on principal Laurent pulses suffer significant performance degradation in long-memory scenarios. To address this challenge, we propose a turbo-iterative detection scheme that treats the combined effects of Laurent pulses, matched filtering, and noise whitening as a virtual channel. With this treatment, unitary approximate message passing (UAMP) is employed to efficiently recover pseudo-symbols with complexity that remains independent of the CPM memory length. The proposed receiver jointly performs equalization, demodulation, and decoding in an iterative manner, leading to superior performance. Simulation results demonstrate that the proposed method significantly outperforms existing techniques under long-memory CPM conditions, particularly for high-order CPMs.

Accurate blind estimation of the symbol timing offset (STO) and carrier frequency offset (CFO) in time-frequency overlapped satellite (TFOS) signals is crucial for monitoring such signals without pilot sequences. To address this challenge, this letter proposes a novel high-precision joint blind estimation algorithm for STO and CFO in TFOS signals. Specifically, we construct an objective function parameterized by STO and CFO using the cyclic correlation functions (CCFs) of TFOS signals. Furthermore, we develop the improved water flow optimizer (IWFO) method to efficiently search for the optimal solutions of the objective function, achieving superior convergence performance. Numerical results show that the proposed algorithm requires an extremely low sampling rate, supports multiple modulation schemes, and markedly outperforms benchmark algorithms in terms of estimation accuracy.

To address the issues of poor fault tolerance and high computational complexity in traditional matrix analysis methods for identifying the code length and synchronization of linear block codes, an improved identification method based on ordered Gaussian elimination was proposed. The algorithm first constructed a codeword matrix from the intercepted bit stream and divided it into a set of continuously overlapping submatrices. Ordered Gaussian elimination was then applied to solve for potential check vectors, and check vectors were selected using a threshold decision, while relevant statistical measures were calculated. Finally, the code length and synchronous position were identified based on the distribution patterns of these statistical measures. Simulation results demonstrated that, compared to existing methods, the proposed algorithm significantly reduced complexity by avoiding redundant calculations and further improved fault tolerance by optimizing decision criteria, exhibiting stronger robustness in high bit error rate environments.

Continuous phase modulation (CPM) has extensive applications in wireless communications due to its high spectral and power efficiency. However, its nonlinear characteristics pose significant challenges for detection in frequency selective fading channels. This paper proposes an iterative receiver tailored for the detection of CPM signals over frequency selective fading channels. This design leverages the factor graph framework to integrate equalization, demodulation, and decoding functions. The equalizer employs the unitary approximate message passing (UAMP) algorithm, while the unitary transformation is implemented using the fast Fourier transform (FFT) with the aid of a cyclic prefix (CP), thereby achieving low computational complexity while with high performance. For CPM demodulation and channel decoding, with belief propagation (BP), we design a message passing-based maximum a posteriori (MAP) algorithm, and the message exchange between the demodulator, decoder and equalizer is elaborated. With proper message passing schedules, the receiver can achieve fast convergence. Simulation results show that compared with existing turbo receivers, the proposed receiver delivers significant performance enhancement with low computational complexity.

Holographic multiple-input multiple-output (HMIMO) technology holds great promise for delivering high energy and spectral efficiency, boosting system capacity, enhancing diversity, and achieving other significant performance gains. In this work, we focus on the issue of HMIMO symbol detection, which is challenging due to the non-ideal characteristics of the HMIMO near-field (NF) channel matrix introduced by the dyadic Green’s function. These characteristics, such as high-dimensional, ill-conditioned, correlated or rank-deficient, pose considerable difficulties for effective symbol detection. To tackle this problem, we propose an efficient symbol detection algorithm by leveraging the structures of HMIMO NF channel. Specifically, by exploiting the block symmetry of the fully polarized NF channel model, we first decompose the high-dimensional signal model into multiple low-dimensional sub-models, which reduces the computational complexity of preprocessing for the channel matrix compared to its predecessor, thus permitting the design of efficient symbol detection algorithms. Then, building upon these multiple sub-signal models, we formulate the symbol detection problem within a probabilistic framework and construct the corresponding factor graph. By utilizing this factor graph and unitary approximate message passing (UAMP), we propose an efficient Bayesian symbol detection algorithm. The proposed symbol detection algorithm effectively mitigates the adverse effects caused by imperfections in the HMIMO NF polarized channel matrix. Simulation results verify the proposed method outperforms the conventional symbol detector.

In recent years, the energy crisis has prompted widespread attention to electric vehicles (EVs). As a core component of electric drive system in EV, motor controller has a significant impact on vehicle energy consumption. In such situation, this paper proposes a minimum loss modulation method for all power factors and reduces current THD by dead-band compensation. Firstly, the mechanism model of the controller is established for loss calculation, and the relationship between the clamping region and the current peak is analyzed based on the traditional modulation methods. Secondly, a discontinuous pulse width modulation considering power factor angle method (CPFA-DPWM) is proposed to ensure that clamping region always falls at the peak current at all power factors. Furthermore, the adoption of current vector method instead of direct sampling for current polarity determination effectively reduces judgment error and avoids false dead-band compensation. The effectiveness of the proposed method is verified by simulation and thermal measurement experiment. Under China Light-duty Vehicle Test Cycle (CLTC), the controller loss of CPFA-DPWM is reduced by 18.61 % compared to SVPWM.

Considering the problem of blind identification of Unique Words (UW) for Time Division Multiple Access (TDMA) signals in non-cooperative communication, a blind identification algorithm for distributed UW is proposed in this paper. Different from the unique codes recognition algorithm at the bit layer, a unique words recognition algorithm at the waveform layer oriented to the correlation is proposed between different windows of the modulated data for centralised unique words and distributed unique words, respectively. The algorithm takes advantage of the consistency and correlation of the unique words and proceeds in two steps: firstly, the unique words of different burst signals are vertically aligned by eliminating the effects of frequency and phase bias between the different burst signals through differential accumulation, and then the positions and lengths of the unique words are identified by the multilayer differential conjugate fourth order correlation algorithm. The performance of the algorithm is simulated and analysed with different number of bursts, signal-to-noise ratios, and with or without frequency and phase biases, and the effectiveness of the waveform layer identification of unique words is verified, and the algorithm achieves more than 95% of the identification rate at a signal-to-noise ratio of 5dB for both centralized and distributed unique words, which is of certain value for engineering applications.

This paper studies improving the detector performance which considers the activity state (AS) temporal correlation of the user equipments (UEs) in the time domain under the uplink grant-free non-orthogonal multiple access (GF-NOMA) system. The Bernoulli Gaussian-Markov chain (BG-MC) probability model is used for exploiting both the sparsity and slow change characteristic of the AS of the UE. The GAMP Bernoulli Gaussian-Markov chain (GAMP-BG-MC) algorithm is proposed to improve the detector performance, which can utilize the bidirectional message passing between the neighboring time slots to fully exploit the temporally-correlated AS of the UE. Furthermore, the parameters of the BG-MC model can be updated adaptively during the estimation procedure with unknown system statistics. Simulation results show that the proposed algorithm can improve the detection accuracy compared with the existing methods while keeping the same order complexity.

Reconfigurable intelligent surface (RIS) has emerged as a promising technology for improving capacity and extending coverage of wireless networks. In this letter, we consider RIS-aided millimeter wave (mmWave) multiple-input and multiple-output (MIMO) communications, where acquiring accurate channel state information is challenging due to the high dimensionality of channels. To achieve efficient channel estimation, fully exploiting the structures of the channels is crucial. To this end, we formulate the channel estimation as a hierarchically structured matrix recovery problem, and design a low-complexity message passing algorithm to solve it, leveraging the unitary approximate message passing. Simulation results demonstrate the superiority of the proposed algorithm with performance close to the oracle bound.

Broadband satellite signals generate a large amount of data at high sampling rates, and spectral data compression is required to reduce transmission bandwidth pressure. The current compression algorithm has a high compression rate for spectral data, this paper proposes a preprocessing algorithm based on the characteristics of large redundancy of broadband satellite spectral data, which differentiates the current frame from the reference frame, and then interleaves the differentiated data to increase the redundancy between the data, and proposes an adaptive coding scheme for the problem of large coding redundancy when the offset is small, which further reduces the compression rate by using fewer bits to represent the coding result. The test results show that the improved algorithm in this paper reduces the compression rate by about 30 % relative to the traditional LZSS algorithm, which can effectively reduce the pressure of data transmission and has practical application value.

In this paper, we design a low-complexity direction-of-arrival (DOA) estimation algorithm based on the unitary approximate message passing (UAMP) and Bernoulli-Gaussian (BG) prior. We first show that the estimation of DOA can be transferred into a sparse signal recovery problem, where we turn to UAMP with damping technique to solve this problem. Furthermore, we assume the BG prior on the sparse vector to be estimated, resulting in the fast estimation of the positions of non-zero elements. Moreover, expectation maximum (EM) is leveraged to automatically learn the BG parameters. Compared to the state-of-the-art UAMP-based algorithm with sparse Bayesian learning (SBL), the proposed approach can achieve the same DOA mean square error (MSE) performance with a much faster convergence speed.

Aiming at the problem of insufficient accuracy of existing carrier parameter estimation algorithms for digital video broadcasting-satellite-second generation extensions (DVB-S2X) signals, this paper proposes a two-field parameter estimation framework with the field of the start of the frame (SOF) and the field of physical layer signaling code (PLSC) based on the maximum likelihood criterion. First, for the known SOF field and the unknown PLSC field, the modulation information is removed using correlation and phase duality of pi/2BPSK, respectively. Then, the residual phase margin of the PLSC field can be further cleaned up using the nonlinear transform. By removing modulation information and residual phase margin. the available data for parameter estimation can be greatly extended from the original SOF field to the whole PLHEADER. Finally, the frequency offset is estimated by fast Fourier transform and trilinear magnitude method. Simulation results show that the proposed algorithm can greatly improve the accuracy of frequency offset estimation on the basis of guaranteeing the estimation range.

Utilizing unmanned aerial vehicles (UAVs) as mobile access points or base stations has emerged as a promising solution to address the excessive traffic demands in wireless networks. This paper investigates improving the detector performance at the unmanned aerial vehicle base stations (UAV-BSs) in an uplink grant-free non-orthogonal multiple access (GF-NOMA) system by considering the activity state (AS) temporal correlation of the different user equipments (UEs) in the time domain. The Bernoulli Gaussian-Markov chain (BG-MC) probability model is used for exploiting both the sparsity and slow change characteristic of the AS of the UE. The GAMP Bernoulli Gaussian-Markov chain (GAMP-BG-MC) algorithm is proposed to improve the detector performance, which can utilize the bidirectional message passing between the neighboring time slots to fully exploit the temporally correlated AS of the UE. Furthermore, the parameters of the BG-MC model can be updated adaptively during the estimation procedure with unknown system statistics. Simulation results show that the proposed algorithm can improve the detection accuracy compared to existing methods while keeping the same order complexity.