Search

low power consumption and high efficiency of heterogeneous systems improves processing power and enables the implementation of real-time applications. Deep learning, as one of the hottest topics of today, plays an important role in solving difficult problems such as machine vision. The use of traditional methods for solving visual machine problems requires the engineering of features by humans, which makes it difficult to create a comprehensive model for a problem. The use of revolutionary deep learning in the machine vision, which along with the embedded systems can be useful in many today's issues. Convolutional neural networks have shown a high degree of efficiency in the task of categorizing images and detecting objects. An important feature in neural networks is the intrinsic parallelism of its structure, which results in the use of embedded heterogeneous systems that can provide excellent performance in the implementation of neural networks. Implementing real-time objects detection systems in enclosed environments with limited computing resources and memory is challenging. This paper presents a method for implementing the MobileNet-SSD object detection system on the Jetson TK1, which attempts to improve performance by changing the network's convoys and dividing tasks between the central and the graphics processor.

IoT technology is growing very fast and one of the requirements of this technology is integrating different communications protocols and networks. In an IoT network, security of such a heterogeneous and large network is very important. Transport systems are part of this super network and in-vehicle protocols are used in such systems. Unfortunately, the Controller Area Network (CAN) protocol, the most popular protocol in the systems, designed without security in mind. In this paper, the Advanced Encryption Standard (AES), an encryption algorithm, is used to prevent masquerade and replay attacks in order to secure CAN protocol to an appropriate level. The paper has a plan to explore for an efficient implementation of AES encryption algorithm for the communication protocol. These implementations have been evaluated on an FPGA ML605 development board. The best implementation of the AES among 8-, 16-, 32- and 64-bit data paths has been investigated. The most important criteria for the protocol in these AES designs are such as consumed power, area, and cost in addition to providing better throughput. The 64-bit structure of the designed AES is selected which has the frequency of 21.4 MHz, significant throughput of 412.39 Mbps, reasonable area of 784 slice on Spartan III FPGA.

Nowadays, GPUs are known as one of the most important, most remarkable, and perhaps most popular computing platforms. In recent years, GPUs have increasingly been considered as co-processors and accelerators. Along with growing technology, Graphics Processing Units (GPUs) with more advanced features and capabilities are manufactured and launched by the world's largest commercial companies. Unified memory is one of these new features introduced on the latest generations of Nvidia GPUs which allows programmers to write a program considering the uniform memory shared between CPU and GPU. This feature makes programming considerably easier. The present study introduces this new feature and its attributes. In addition, a model is proposed to predict the execution time of applications if using unified memory style programming based on the information of non-unified style implementation. The proposed model can predict the execution time of a kernel with an average accuracy of 87.60%.