This paper proposes a new method for power telemetry and wireless data transfer to biomedical implants. Confinement of the transferred energy within a predefined area, hence not influencing the neighboring sensitive circuitry on the microsystem, and also allowing for more than one telemetry link on the same platform and even operating at the same carrier frequency are among the advantages of the proposed method over the traditional approaches for power and data telemetry.
In the rapidly evolving field of autonomous cars, advanced deep learning systems have ushered in a new era of innovation, enabling the integration of unique features into vehicles. These advancements span various areas, including pedestrian and vehicle detection, recognition of road signs and driving patterns, identification of drivable roads and scenes, and improved mapping and routing techniques. However, the high computational requirements of deep learning networks present a significant challenge, especially for embedded systems like FPGAs (Field-Programmable Gate Arrays) that have limited capacity. Addressing this challenge, this article presents a comprehensive survey of the methodologies employed in implementing Convolutional Neural Networks (CNNs) on resource-constrained processors, within the domain of self-driving car applications. Our survey encompasses a thorough review of the existing literature in the field of deep learning applied to autonomous cars, from perception to localization, with a specific emphasis on implementations utilizing embedded hardware such as FPGAs and embedded GPUs. Furthermore, we present and analyze results that elucidate the intricate trade-offs between latency, energy consumption, and the judicious selection of the underlying platform. These insights are crucial for researchers and practitioners in the field, as they provide a clear direction for optimizing the performance of deep learning networks on resource-constrained platforms, ultimately contributing to the advancement of self-driving car technologies.
6 May 2019 in International Journal of Control, Automation, and Systems
This paper presents an online secondary path modeling method in active noise cancellation based on an enhanced Filtered-X Least Mean Squares (FxLMS) algorithm, implemented on an Altera Cyclone II board. The main specifications of the proposed algorithm are: (1) it avoids overdriving noise in ear by applying a limiter block, (2) it updates online secondary path modeling filter coefficients, utilizes error signal power, and minimizes system error by injecting random white noise at an appropriate time, which can modify the system changes, and (3) it synthesizes the hardware architecture using VHDL Hardware Description language for an Altera Cyclone II board. In this architecture, only 22 of 300 embedded multipliers and 6% of the logical elements in the FxLMS algorithm have been used. The proposed method has used less than 1% of the Field-programmable Gate Arrays’ board memory. The experiments illustrate convincing results in terms of accuracy, convergence speed, and functionality of the designed and implemented system.