Abstract:To address the coexistence of high-voltage excitation residuals and weak echo signals in high-frequency ultrasonic microscopic measurement, a miniaturized ultrasonic pulser-receiver and a field-programmable gate array control method are designed. The receiving chain integrates limiting protection, programmable gain, power management, and mode switching. The field-programmable gate array is used to coordinate triggering, gain configuration, receiver blanking, and sampling window control. Test results show that the limiting output is approximately ±1.6 V, the gain range is approximately -10 dB to 70 dB, and the system can perform A-scan acquisition using transducers from 50 MHz to 200 MHz. In addition, sample scanning imaging is achieved with a 120 MHz transducer and a scanning step of 4 μm. The results indicate that the proposed design can support echo protection, amplitude conditioning, and synchronous acquisition in high-frequency ultrasonic microscopic testing.
Abstract:To overcome the limitations of traditional MCU/DSP-based systems in multi-target position acquisition, this paper presents a high-precision monitoring system utilizing an FPGA. The design employs the Xilinx XC7Z020 SoC to implement three concurrent EnDat 2.2 protocol channels for synchronous data reading from Heidenhain encoders. Hardware-based mechanisms for data synchronization and CRC verification ensure reliability. Experimental results show the system achieves a 60 kHz update rate per channel with a bit error rate below 10⁻⁹, reducing response latency by 85% compared to serial methods and significantly improving tracking performance.
Abstract:Focusing on the deep space communication channel coding standards recommended by the Consultative Committee for Space Data Systems (CCSDS), this paper proposes a resource-efficient QC-LDPC encoding architecture tailored for resource-constrained scenarios. Based on a thorough analysis of the QC-LDPC algorithm and its generator matrix construction, a pipelined Parallel Shift-Accumulation Register (PSAR) coding architecture is designed, utilizing the quasi-cyclic characteristics through a "serial-input/parallel-computation" approach. By optimizing the ROM-based memory mapping strategy, this architecture effectively addresses the excessive resource consumption typically found in traditional fully parallel encoders. Finally, a coding circuit with a block length of 4096 bits and a code rate of 1/2 is implemented and verified on the Xilinx Kintex-7 FPGA platform using Verilog HDL. Experimental results demonstrate that under a 200 MHz system clock, the proposed encoder achieves a high throughput of 100 Mb/s while consuming minimal hardware resources (occupying only about 4% of Slice LUTs and 7% of Block RAM). The design maximizes resource efficiency without sacrificing throughput, featuring low latency, low power consumption, and strong reusability. It is highly suitable for power- and area-constrained deep space exploration communication terminals.
Abstract:Accurately controlling the output status of gate drive signal pins is the key technology for in-system programming in multi-level control systems. To overcome the inherent drawback of traditional watchdog chips in monitoring the in-system programming process, a watchdog masking circuit with a window period ranging from 1 µs to 33.55 s has been developed. By using the pin level variation patterns of the FPGA during the process, combined with the edge delay characteristics of the LTC6994-1 chip, the output of the gate drive signal is enabled. In order to verify the feasibility of this system, a physical hardware and a simulation model of the FPGA in-system programming watchdog masking circuit based on the LTC6994-1 are constructed. The experimental results show that during the in-system programming process, the system can not only control the output status of the gate drive signal, but also reliably reset program maintenance faults.
Abstract:With the development of artificial intelligence technology, object detection algorithms have been widely applied in various fields. To address the problem that complex scenarios cannot adaptively adjust according to complex and variable environments, a method combining it with target detection preprocessing is proposed in conjunction with dynamic reconfigurable technology. Firstly, preprocessing modules with different functions are designed, and then corresponding dynamic reconfiguration design is carried out, achieving a flexibly reconfigurable system architecture. This system architecture can achieve adaptive loading and switching of preprocessing modules under different complex scenarios, reducing the system resource usage by 67%. The results show that the system can efficiently deploy different preprocessing functions at the edge without affecting real-time inference.