http://nopr.niscpr.res.in/handle/123456789/58752 2026-10-10T19:09:15Z http://nopr.niscpr.res.in/handle/123456789/58761 Title: Implementation of Neural Networks in FPGA Authors: B, Jayanthi; Kumar, Lakshmi Sutha Abstract: Artificial Intelligence (AI) refers to the recreation of human intelligence in machines that have been designed to think like humans and mimic their actions. AI has been used in many fields such as image processing, health care, education, and marketing. Machine Learning (ML) has been the sub-division of AI, and deep learning has been the subdivision of ML. Artificial Neural Network has been the most predominantly used deep learning technique. While implementing the ANN technique, knowing whether the implementation could have been done in hardware or software becomes necessary, which is essential to achieve the expected performance. This paper gives a survey on the available methods in which the ANN architecture has been implemented to achieve efficient output with minimal resources. It is vital to study and analyze various strategies for implementation and their functionality. This paper has also explained the advantages and disadvantages of different implementation techniques that allow selecting the most appropriate hardware and respective methodology for optimizing the hardware. Page(s): 57-63 2021-06-01T00:00:00Z http://nopr.niscpr.res.in/handle/123456789/58760 Title: Intelligent transportation system and smart traffic flow with IOT Authors: Kaluvan, Hariharan; Baskar, Praveen Kumar; Ramanathan, Shanmugam; Sundar, Shriram; Jeyaram, Sridhar Abstract: There has been an increase in vehicles across the globe. Also, the congestion due to traffic has leapfrogged in India. The traffic flow information has been required to find out the route with minimum congestion and forecast the traffic. And this has been a part of the Intelligent Transportation System (ITS) which would help build smart cities. A lot of work has been done on the traffic measurement system. But the integration of emerging techniques such as the Internet of Things (IoT) and cloud computing has provided a lot of research scope in ITS. This paper has proposed an IoT-based method to determine the real-time traffic flow in a road section with ultrasonic sensors, Arduino, ESP8266 Wi-Fi module, and an open-source cloud. There has been an average traffic flow every five minutes to be displayed in the cloud platform. This method can be very much cost-effective with less power consumption and improved accuracy. Hence, the proposed IoT-based technique has provided the traffic flow data, and this data shall further be used for traffic predictions using machine learning algorithms. Page(s): 64-67 2021-06-01T00:00:00Z http://nopr.niscpr.res.in/handle/123456789/58759 Title: Detecting Autism spectrum disorder with sailfish optimisation Authors: Balakrishnan, K; Dhanalakshmi, R; Khaire, Utkarsh Mahadeo Abstract: Autism Spectrum Disorder (ASD), a neurodevelopmental disorder, has been a bottleneck to several clinical researchers due to data modularization, subjective analysis, and shifts in the accurate prediction of the disorder amongst the sample population. Subjective clinical research suffers from a lengthy procedure, which is a time-consuming process. In this paper, Sailfish Optimization (SFO), a recently developed nature-inspired meta-heuristics optimization algorithm, is being utilized to detect ASD. The hunting methodology of sailfish inspires SFO. Classical SFO has examined the search space in only one direction that affects its converging ability. The Random Opposition Based Learning (ROBL) strategy enhances the exploration capacity of SFO and successfully converges the predictive model to global optima. The proposed ROBL-based SFO (ROBL-SFO) selects relevant features from autism spectrum disorder (child and adult) datasets. According to the results obtained, the proposed model outperforms the convergence capability and reduces local-optimal stagnation compared to conventional SFOs. Page(s): 68-73 2021-06-01T00:00:00Z http://nopr.niscpr.res.in/handle/123456789/58758 Title: Compact WLAN notched ultra-wideband band pass filter Authors: S, Ramkumar; R, Boopathi Rani Abstract: This paper has investigated WLAN notched Ultra-Wide Band (UWB) Band Pass Filter with a wide upper stop band. The proposed design has folded half-wave line that has been incorporated in between slot line resonators. Designed UWB band pass filter has 3-dB cut-off frequencies at 3.2 GHz and 10.1 GHz with a notched band from 5.1 GHz to 5.4 GHz. Return loss and insertion loss in the first pass band have been greater than 12 dB and less than 1.5 dB. These have been greater than 25 dB and lesser than 1.8 dB in the second pass band. In the upper transition edge, an introduced filter has accomplished pointed roll-off with attenuation of 62 dB, wide stop band with the rejection of 20 dB till 12 GHz, and greater than 17 dB from 12 GHz to 18 GHz. Higher-order harmonics have been suppressed through quarter wavelength open stubs. The overall size of the filter is 27 x 9 x 0.8 mm3. Page(s): 74-79 2021-06-01T00:00:00Z