The SoCC aims to establish a world‐leading research center in high-performance, energy efficient, small form factor, and low-cost electronic systems achieved through innovation in low-power digital processing, wireless communications, power management, and sensor technologies. The Center will directly contribute toward human and intellectual capital underpinning high‐tech job creation and direct local and foreign investment. It will extend the current partnership with local and global industries to address current and future challenges.
The SoCC will harness its faculty and ETIC expertise to conduct basic and applied research to develop novel solutions in circuits and systems to serve applications in a variety of fields including military, healthcare, artificial intelligence, security, space, robotics, and sensing. The need for specialized electronic devices has been the emerging trend in research and the industry, especially for restricted technology (high frequency), mobile, and Internet of Things. The Center will provide an excellent opportunity to enable a highly skilled workforce to drive innovation and entrepreneurship in the UAE’s electronics sector in line with the Abu Dhabi 2071 vision.
Wireless sensor networks (WSNs) and Internet-of-Things (IoT) are among the main enabling technologies for smart systems (transportation, farming, health, etc.). Therefore, WSNs/IoT have received enormous attention where various communications-related performance metrics were considered such as reliability, connectivity, throughput, delay, network life-time, power efficiency, and security. In the literature, WSN design is typically optimized to maximize/minimize one or more of the aforementioned metrics under certain constraints. However, very little work has performed system/network optimization while considering the communications requirements jointly with the decision fusion process. Such approach is expected to produce novel system, network, and signal designs. Consequently, the main objective of this project is to jointly consider communication and data fusion requirements in designing an efficient WSN-IoT system/network that maximizes the control action accuracy and minimizes the system cost. The system design will focus on red palm weevil detection, which is one of the world’s most destructive palm pests.
Artificial intelligence (AI) engines have been integrated in a myriad of applications, whether they run in data centers or on edge/end-node devices. Most contemporary AI applications use the cloud to execute computationally intensive and power demanding deep learning algorithms. Moving the processing from the cloud to edge devices reduces data transfers and latency, improves security, and enables scalability. The huge computational requirements of deep learning neural networks (DNN) deem it necessary to achieve challenging tradeoffs among energy, latency, and accuracy at every application level. This project will utilize alternative numbering systems and fused primitives to create DNN architectures with low latency, low energy, and high accuracy for AI implementation. Optimization at the algorithm and architecture level will enable the proposed architectures to attain target power and performance. Moreover, optimized digital circuit primitives for functions that accelerate AI engines in FPGAs and ASICs will be developed, verified, and demonstrated in FPGA prototypes.
Artificial Intelligence (AI) solutions are taking the lead in pioneering research to develop the next-generation technologies. The next stage of AI systems promises to have improved robustness and reliability; enhanced security; reduced power, data, and performance inefficiencies; and enable common sense reasoning. AI applications include, but are not limited to, smart healthcare systems, self-driving cars, robotics, etc. Conventional CMOS technology, which has provided cheaper, faster, and lower power electronics system over the past 40-years, is facing big challenges. In addition, von Neumann architecture does not match the characteristics of machine learning algorithms, hence researchers are looking into novel technology and hardware architectures for accelerating such algorithms. In this project, DeepMem, an efficient intelligent Memristor-based hardware system, is introduced. Memristor technology enables efficient in-memory computations with low power and high density. Co-optimization of machine learning algorithms, as well as Memristor-based in-memory computing for IoT devices, is the main aim of this project.
