Search

Text-based emotion prediction is an important task in the field of affective computing. Most prior work has been restricted to predicting emotions corresponding to a few high-level emotion classes. This paper explores and experiments with various techniques for fine-grained (27 classes) emotion prediction $^\dagger$ † . In particular, 1) we present a method to incorporate multiple annotations from different raters, 2) we analyze the model's performance on fused emotion classes and with sub-sampled training data, 3) we present a method to leverage the correlations among the emotion categories, and 4) we propose a new framework for text-based fine-grained emotion prediction through emotion definition modeling. The emotion definition-based model outperforms the existing state-of-the-art for fine-grained emotion dataset GoEmotions. The approach involves a multi-task learning framework that models definitions of emotions as an auxiliary task while being trained on the primary task of emotion prediction. We model definitions using masked language modeling and class definition prediction tasks. We show that this trained model can be used for transfer learning on other benchmark datasets in emotion prediction with varying emotion label sets, domains, and sizes. The proposed models outperform the baselines on transfer learning experiments demonstrating the model's generalization capability. [ABSTRACT FROM AUTHOR]

In this letter, a combined encryption and error-control coding scheme based on turbo codes (TCs) is proposed to enhance the security of the transmission. It relies on convolutional encoders that operate over Galois fields ${\textrm {GF}}(4)$ and include variable encryption polynomials, whose coefficients are randomly selected every $L$ steps from a set of optimal polynomials. This type of encoder has been studied in previous papers, where a certain correspondence between the randomness of the encoded sequence and the performance of the TC was conjectured. In this work, a scheme with a varying length $L$ is proposed to increase the output randomness and, therefore, the security of the sequence to be transmitted. That is to say, the number of steps $L$ , over which one of the polynomials is iterated, is a random variable. The randomness of the output and the security of the TC are quantified using differential block entropy, showing that it is possible to improve the security of the transmitted data without compromising the corresponding bit error rate (BER). [ABSTRACT FROM AUTHOR]

B5G networks are envisioned to meet higher spectral efficiencies through advanced transmission technologies. Consequently, it is important to comprehend the typical propagation characteristics of wireless communication medium that experience multipath effects, fading, shadowing, and nonlinear distortions. Most of these challenging issues arise under non-line of sight conditions concerning the propagation medium, which deteriorates the quality of the signal received due to uncontrollable channel variations. The recent advances in Intelligent Reflecting Surface (IRS), enable a dynamically alterable wireless environment to achieve flexible reconfigurable channels. The deployment of IRS can evidently bypass the obstacles through smart reflections and shall mitigate channel fading impairments. Motivated by the potentials of the IRS, this paper investigates the performance of IRS-assisted multi-input-multi-output (MIMO) systems to analyze the effect of nonlinear distortions and system adaptation over IID and non-IID Gaussian channel variations. The simulations show the distortion levels with the help of power spectral density for correlated or uncorrelated channel conditions. Furthermore, Monte Carlo simulations are performed to understand the behavior of the wireless system through the analysis of bit error rate (BER) vs signal-to-noise ratio (SNR), Ergodic Channel capacity vs SNR in the presence of IID-Gaussian/ non-IID Gaussian and in-band nonlinear distortions. Additionally, in realistic channel conditions, it is observed that the effect of increasing the number of IRS elements increases the BER performance of ~8dB in both IID and non-IID Gaussian variations. Finally, improvement in the performance of ergodic capacity of 5 bps/Hz has been observed at a fixed SNR of 5dB.

The major problems in an underwater wireless optical communication (UWOC) link are turbulence induced fading and pointing errors. In this paper, we have investigated the bit error rate (BER) performance of spatially coupled low-density parity-check (SC-LDPC) coded horizontal UWOC link over a strong turbulent channel model with pointing errors. The performance of this link for different channel and code parameters has been studied using simulations. It has been observed that a rate 1/2 ARJA protograph based SC-LDPC code with graph lifting factor of 256 gives a coding gain of 47 dB at a BER of ${10}^{-4}$ for strong turbulence channel model with pointing errors. An analytical BER expression for an uncoded UWOC link under strong turbulence with pointing errors for the On-Off Keying modulation technique has been derived. A multidimensional protograph based extrinsic information transfer algorithm has been developed to obtain the decoding thresholds for different channel parameters and code rates. We have also studied the SC-LDPC coded vertical UWOC link performance for some specific strong turbulence channel parameters with pointing errors and observed that as the link length increases from $20 m$ to $40 m$ , the performance gap between the hypothetical and cascaded channel models increases from $1.1 dB$ to $5.5 dB$ .