This paper presents the performance analysis of compound chaotic sequence (CCS)-based noise reduction differential chaos shift keying (NR-DCSK) system under multipath Rayleigh fading channel conditions. The special characteristics of chaotic sequences are their deterministic randomness behaviour that adds security and multipath immunity to the data when used as a carrier in communication systems. In this paper, the chaotic sequences are generated by combining the outputs of chaotic maps, such as logistic map, Chebyshev map, Bernoullishift map, tent map, etc., leading to new complex sequences known as CCSs. This sequence possesses more randomness, overcomes severe interference levels encountered during transmission and provides higher multipath immunity compared with those of pseudo-noise (PN) codes. Since NR-DCSK is a spread spectrum technique, its performance in wireless multipath fading channels has important considerations. The CCS is used as a carrier in NR-DCSK systems, which leads to improved bit error rate (BER) performance. Comparisons of simulation results to theoretical BER expressions of additive white Gaussian noise (AWGN) and Rayleigh fading channels have been carried out to test the efficiency of the proposed CCS-based NR-DCSK system.

Sentiment classification plays a dominant role in day-to-day life including the political events, production areas, and commercial activities. The need for the accurate and instant classification of the user emotions is a hectic task to be solved. The traditional methods fail to address the classification of dynamic dataand huge volumes of documents. Moreover, to assure the classification accuracy and deal with huge volumes of data, the proposed method employs the MapReduce Framework. The proposed sentiment analysis involves twoprocesses, such as feature extraction and classification that is performed in the MapReduce framework using the mapper and reducer functions. The feature vector is based on the sentence-specific features, SentiWordNetbasedfeatures, and statistical features corresponding to the individual reviews that are classified as positive and negative reviews using the proposed Chronological-Brain Storm Optimization based Support Vector Neural Network (CBSO-SVNN). The analysis is progressed using four datasets obtained from the movie reviewdatabase that confirms that the proposed method outperformed the existing methods in terms of the accuracy, sensitivity, and specificity. The accuracy, specificity, and sensitivity of the proposed method are 0.8714, 0.9027,and 0.8714, respectively.