Search

Traffic signals are vital for traffic management. It is essential to increase the efficiency of the traffic controller to alleviate traffic congestion. In this paper, the traffic light control system using fuzzy logic is designed to minimize queue count (QC) and waiting time (WT) for vehicles at the intersection. Real-time traffic is generated and a fuzzy logic controller is implemented to control the traffic light system. In the proposed fuzzy logic controller, trapezoidal membership functions are combined with the rules to generate green light time for input QC and WT. The improvement in QC, WT, and tailback is witnessed by using Static Phase Scheduling Traffic Light System (SPSTLS) as a benchmark to measure the performance of the proposed controller. The performance comparison records a significant improvement of 81.68% in QC, 87.04% in average WT and 18.05% in the tailback. The results show that fuzzy logic controlled Dynamic Phase Scheduling Traffic Light System (DPSTLS) has the potential to resolve the problem of QC, WT, travel cost, accident, and traffic congestion.

One of the pivotal issues which must be tackled when an effective brain-computer interface (BCI) is to be designed, is to reduce the enormous space of features extracted from fNIRS signals. BCI researchers often use genetic algorithms (GA) as the technique to extract features. The classic genetic algorithm obtains a feature set with the high classification accuracy; however, it is unable to create a reliable classifier due to the large dimensionality of this set. Herein, we propose a refined version of the genetic algorithm, which chooses a small number of features and determines the optimal feature combination for classification of fNIRS signals for a BCI system.

In this paper, we have compared and analyzed the classification accuracies using two different classifiers for mental arithmetic and rest task using functional near-infrared spectroscopy (fNIRS) signals. Multi-channel continuous-wave imaging system was used to extract the signals from the prefrontal cortex of the brain of seven healthy subjects. The extracted signals were then converted into change in concentration of oxygenated and deoxygenated hemoglobin using modified Beer-Lambert Law. Three features were extracted from the signals after removing the physiological noises. The classification accuracies using 2-feature combinations (signal variance and signal skewness; signal slope and signal variance; signal skewness and signal slope) for k-nearest neighbour (kNN) classifier are 93.8, 50.7 and 50.7% respectively and for artificial neural network (ANN) classifier are 97.9, 89.9, and 97.8% respectively.

The capabilities of Wireless Sensor Network (WSN) are unveiling with each passing day. Extensive use of WSNs is giving rise to different types of threats. To defend against the threats proper security schemes are required. Limited area, power and memory of WSNs implement strict constraints on the selection of cryptographic techniques. Elliptic Curve Cryptography (ECC) is the best candidate due to its smaller key size. High security despite of smaller key size results in area and power efficient crypto systems. This paper describes the hardware implementation of Elliptic Curve based asymmetric cryptosystem for wireless sensor networks. The field used for elliptic curve is defined over prime number. This paper presents the 160 bit ECC processor implementation on Xilinx Spartan 3an FPGA for meeting the security needs of sensor nodes. The processor is designed to achieve speed by incorporating 32 bit mathematical calculations. The design also provides excellent area results.