Wireless Sensor Networks (WSNs) are comprised of autonomous nodes, where the physical devices and other equipment are embedded for monitoring the environment. Yet, it consumes larger energy consumption while it fails to provide the optimal routing path. The data aggregation in WSN is accomplished by several classification approaches, but the delay and utilization of the resource are high. So, it is necessary to solve the challenges by implementing the developed model for the energy efficient routing in WSN. A distributed data mining scheme is developed based on a deep learning approach to provide greater energy efficiency and optimal load balancing in WSN. The distributed data mining is performed to distribute the appropriate data to the sensor nodes. This data distribution highly reduces the overhead at the fusion center. The data are categorized via Fused Deep Structural Network (FDSNet), where the Attention-based Dilated Deep Temporal Context Networks (DTCN) are fused with a Deep Shallow Network. The necessary data are distributed to the sensor nodes. The optimal routing is carried out based on optimal path selection using Stability Bound-based Energy Valley Optimizer (SBEVO). The optimal routing is done via the multi-objective functions such as “energy consumption, PDR, delay, and shortest path”. The energy-efficient data transmission is accomplished with the help of developed distributed data mining with a routing scheme. The developed energy-efficient optimal routing in WSN is compared to conventional routing approaches with several performance metrics to show energy efficiency. While comparing with existing methods, the developed model attains the value of 0.95% with regards of accuracy, precision, and F1-score. These findings of the developed model show accurate performance with an efficient routing mechanism in WSN.

Natural Language Processing (NLP) is an integral part of machine-to-human conversation. It requires a deep understanding of diverse human utterances. This expectation from machines encouraged the researchers to study the behavior of machines in order to understand NLP. Because humans can understand the semantics of a word during conversation, but making a machine understand it is a difficult task. Paraphrasing is a way to create multiple forms of human utterances. Thus, it is capable enough for the machines to understand the utterances while communicating with humans. Paraphrasing is a comprehensive task that involves the replacement of words or phrases and high-level syntactic rearrangement processes such as passiveness, left-splitting, and topicalization. So, this research aims to develop a novel hybrid approach that incorporates rules with machine learning techniques to generate diverse paraphrases without the use of labeled paraphrased datasets. The model is trained and evaluated using two datasets namely Quora and mobile service provider queries dataset. In this paper, two algorithms are designed; one is used to create a labeled template-based dataset, and another algorithm to produce a high level of diverse syntactic paraphrases. The evaluation shows that the proposed approaches generate more high-quality paraphrases than baselines along with an increase in the diversity of paraphrases. The proposed model attained the ROUGE-1 value of 66 which is higher than the other baseline models. Thus, it proved that the proposed approach can effectively improve its performance in both syntactic and semantic structure during paraphrase generation.

In recent times, the human agents for chat services are used to be more in the communication system among human users according to artificial intelligence. These chatbots seem to be cost-efficient and time-saving paradigm, and so it is widely developed for usage. Still, it suffers from satisfying the customer requirements, which reduces the users being less liable to conform to desires through the chatbot. The designed interactive system has to process the queries and gets a suitable reply in the chatbot. The data collection related to diverse queries is gathered. Then, the pre-processing is carried out. Further, the word to vector is conducted for extracting the features. Then, the weighted feature vector is attained through optimizing the weight by hybrid optimization named Hybrid Water Wave Barnacle Mating Optimization (HWW-BMO). Finally, detection is performed through Heuristic-based Extreme learning + Long Short-Term Memory (H-ExtLSTM) for predicting the intent related to the queries asked by mobile service providers, thus can extract the answers based on word similarity. Here, the parameters of ELM and LSTM are optimized by the HWW-BMO algorithm. Experiments show that the proposed joint models have achieved competitive results in different datasets in multiple scenarios compared with other models.

Procurement is an essential operation of every organization regardless of its size or domain. As such, aggregating the demands could lead to better value-for-money due to: (1) lower bulk prices; (2) larger vendor tendering; (3) lower shipping and handling fees; and (4) reduced legal and administration overheads. This paper describes our experience in developing an AI solution for demand aggregation and deploying it in A*STAR, a large governmental research organization in Singapore with procurement expenditure to the scale of hundreds of millions of dollars annually. We formulate the demand aggregation problem using a bipartite graph model depicting the relationship between procured items and target vendors, and show that identifying maximal edge bicliques within that graph would reveal potential demand aggregation patterns. We propose an unsupervised learning methodology for efficiently mining such bicliques using a novel Monte Carlo subspace clustering approach. Based on this, a proof-of-concept prototype was developed and tested with the end users during 2017, and later trialed and iteratively refined, before being rolled out in 2019. The final performance was 71% of past cases transformed into bulk tenders correctly detected by the engine; for new opportunities pointed out by the engine 81% were deemed useful for potential bulk tender contracts in the future. Additionally, per each valid pattern identified, the engine achieved 100% precision (all aggregated purchase orders were correct), and 79% recall (the engine correctly identified 79% of orders that should have been put into the bulk tenders). Overall, the cost savings from the true positive contracts spotted so far are estimated to be S$7 million annually.

This study investigates the use of Augmented Reality (AR) generated analytics to measure the number of times a child can perform the action of crossing the midline by a drumming action. Using a prototype Drum AR App with gamification techniques to motivate the drumming actions with a drumstick, we can capture and log key metrics which measure the physical drumming actions such as the timings and the quality of crossing of the midline through game play. This study is still work in progress where further research on the uses and effectiveness of AR generated analytics can give insights to the development of perceptual motor skills in young children. Generating 2D and 3D plots from the AR generated analytics gives evidence to further identify novel metrics to capture, analyse and give insights to user interactions while using AR apps in learning.

Wireless service providers often have to maximize their profit while satisfying the pre-specified Grade-of-service (GoS). However, the uneven traffic demand over a day has adverse effect to the design - the high blocked call probability during the peak-load hour limits the possible number of admitted users while low traffic during the off-peak hour results in the inefficient use of radio spectrum. The aim of this work is to study how a monopoly service provider can maximize the profit given that the daily traffic demands are time-varying. A possible way is to alter traffic demands through pricing differently at different periods of a day. Earlier study has shown that such peak-load pricing mechanism can reshape the traffic demand. As a result, the overall profit increases by moving some of the traffic to off-peak hour while at the same time improves the GoS of the group of users who are willing to pay more during the peak. There is generally a lack of theory to model the change even though some measurement has been made in the literature. We develop a simple model to predict the effect of peak-load pricing - one of the key features in the mechanism is the inter-temporal substitutability between the peak-load and off-peak period. We calibrate the model against the empirical evidences available.

We detail a code mapping scheme designed especially for devices in the enhanced wireless personal area networks operating in both the 868 MHz and 915 MHz bands for the emerging IEEE 802.15.4b standard. This carefully designed coding scheme, called cyclic odd bit inversion length 16 (COBI-16), meets all the following three essential criteria: (a) no spikes in the frequency spectrum, (b) low complexity correlation operations and (c) simple frequency offset and DC compensation operations. The scheme is tested in both AWGN and Rayleigh fading (flat and frequency selective) channels. Both coherent and non-coherent demodulations are investigated. Also investigated are the synchronous false alarm rate and synchronous miss rate in the AWGN channel. The results show that the scheme conforms to all the stringent requirements defined by the IEEE 802.15.4b protocol. The scheme has been accepted by the IEEE 802.15.4b task group as the standard for data transmission.

IEEE 802.15.4b is an enhanced standard to the IEEE 802.15.4, addressing the needs for high-data rate transmission using 868 or 915 MHz band in the wireless personal area networks. This paper details a modulation and code mapping scheme designed specifically for the standard's high rate transmission schemes operating in the 868 MHz and 915 MHz bands; which leads to efficient transmission and low implementation complexity. The code mapping scheme is called Quadrature Cyclic Odd Bit Inversion length 16 or Quad COBI-16. Among the scheme's desirable properties is that the total phase rotation in the I/Q plane accumulates to 0 degree over the complete symbol. The scheme is tested in both AWGN and Rayleigh frequency selective fading channels. The results show that it conforms to all the stringent requirements defined by the standard.

IEEE 802.15.4b is an enhanced standard to the IEEE 802.15.4, addressing the need for high-data rate transmissions in the 868 and 915 MHz band in the wireless personal area networks. This paper details a code mapping scheme designed specifically for the standard's 915 MHz band that leads to efficient transmission and low implementation complexity. The scheme is called quadrature cyclic odd bit inversion length 16 or Quad COBI-16 and one of its desirable properties is that the total phase rotation in the I/Q plane accumulates to 0 degree over the complete symbol. The scheme is tested in both AWGN and Rayleigh fading (flat and frequency selective) channels. Both coherent and non-coherent demodulations are investigated. Also investigated are the synchronous false alarm rate and synchronous miss rate in the AWGN channel. The results show that the scheme conforms to all the stringent requirements defined by the IEEE 802.15.4b protocol.

This paper addresses the issue of blind channel estimation (BCE) in a SIMO FIR channel. This is an issue arising in e.g. mobile communication systems using digital signaling. It also arises in the presence of frequency selective fading, when training (reference/pilot) symbols are unavailable. In this contribution we analyze the performances of two types of channel coding schemes; turbo coding and serial concatenated convolutional coding with two types of second order statistical BCE algorithms. The first BCE algorithm is the subspace decomposition algorithm (SSDA) and the second is the outer product decomposition algorithm (OPDA). The object of this contribution is to investigate the performances of the coded scheme with these two types of BCE algorithms in order to characterize a possible and realistic scenario. The results show that the coded schemes using the SSDA can outperform the scheme using the OPDA by as much as 7 dB

We detail a cyclic odd bit inversion length 16 (COBI-16) code mapping scheme designed especially for devices in the enhanced wireless personal area networks operating in the 915 MHz for the emerging IEEE 802.15.4b standard. This carefully designed coding scheme meets all the following three essential criteria: (a) no spikes in the frequency spectrum, (b) low complexity correlation operations and (c) simple frequency offset and DC compensation operations. The scheme is tested in both AWGN and Rayleigh flat and frequency selective fading channels. Both coherent and noncoherent demodulations are investigated. Also investigated are the synchronous false alarm rate and synchronous miss rate in the AWGN channel. The results show that the scheme conforms to all the stringent requirements defined by the IEEE 802.15.4b protocol. The scheme has been accepted by the IEEE 802.15.4b task group as the standard for data transmission

This paper addresses the issue of blind channel estimation (BCE) in a SIMO FIR channel. This is an issue arising in e.g. mobile communication systems using digital signaling. It also arises in the presence of frequency selective fading, when training (reference/pilot) symbols are unavailable. In this contribution we analyze the performances of a 1/2 and 1/3 rate turbo code of various frame lengths with two types of second order statistical BCE algorithms. The first algorithm is the subspace decomposition algorithm (SSDA) and the second is the outer product decomposition algorithm (OPDA). The object of this contribution is to investigate the performances of the turbo coded system with these two types of BCE algorithms in order to characterize a possible and realistic scenario. The results show that the turbo coded schemes using the SSDA can outperform the scheme using the OPDA by as mush as 7 dB.

We detail a cyclic odd bit inversion length 16 (COBI-16) code mapping and modulation scheme designed especially for devices in the enhanced wireless personal area networks (WPAN) operating in the 868 MHz for the emerging IEEE 802.15.4b standard. This COBI-16 scheme has been accepted and incorporated in the draft IEEE 802.15.4b standard. This carefully designed coding and modulation scheme meets all the following three essential criteria: (a) no spikes in the frequency spectrum, (b) low complexity correlation operations and (c) simple frequency offset and DC compensation operations. The transmitted signal meets all the IEEE standard's stringent PSD mask requirements. The system is tested in both AWGN and Rayleigh fading (flat and frequency selective) channels. Both coherent and non-coherent demodulations are investigated. Also investigated are the synchronous false alarm rate and synchronous miss rate in the AWGN channel

Blind channel estimation (BCE) of multi channel FIR filters is an issue arising in e.g. mobile communication systems using digital signaling. It also arises in the presence of frequency selective fading, when training (reference/pilot) symbols are unavailable. In this contribution we analyze the performances of two types of channel coding schemes; turbo coding and serial concatenated convolutional coding (SCCC), with BCE. The BCE technique used here is a second order statistics algorithm based on subspace decomposition (SSD). The coding schemes are combined with this BCE technique to characterize a realistic scenario: in that we do not assume the transmitted signal to be white. Two types of blind equalizers are investigated; the zero forcing and the minimum mean squared error equalizers. The results show that the turbo-coding scheme performs better than the SCCC scheme by an average of 1-1.5dB.

This paper investigates the performance of a turbo coded system transmitted over correlated flat fading channels using quaternary partial response continuous phase modulation (CPM). The system provides an attractive option for spectral efficient communications systems. We consider both Rayleigh and Rician fading and compare the performances of a maximum a posteriori (MAP) algorithm and a soft output Viterbi algorithm (SOVA). The paper considers the hybrid concatenation of a parallel designed turbo encoder and a partial response quaternary CPM modulator. Based on the decomposition theory of CPM, the quaternary CPM modulator is decomposed into a quaternary ring continuous phase encoder (CPE) and a memoryless modulator (MM). This allows a quaternary convolutional encoder to be combined with the CPE without the need of a mapper. The hybrid design combined with iterative decoding yield very good performances. The results show the considerable superiority of the MAP algorithm to that of the SOVA.