The flexibility of traditional image processing system is limited because those system are designed for specific applications. In this paper, a new TMS320C64x-based multi-DSP parallel computing architecture is present...The flexibility of traditional image processing system is limited because those system are designed for specific applications. In this paper, a new TMS320C64x-based multi-DSP parallel computing architecture is presented. It has many promising characteristics such as powerful computing capability, broad I/O bandwidth, topology flexibility, and expansibility. The parallel system performance is evaluated by practical experiment.展开更多
面向差异化业务需求,电力物联网(Electric Internet of Things,EIoT)需要设计与之适配的数据处理架构,该架构将引入数据缓存、边缘处理等功能,并且涵盖EIoT中数据的清洗、过滤和融合等关键步骤。此外,在该架构基础上,需要同时满足大规...面向差异化业务需求,电力物联网(Electric Internet of Things,EIoT)需要设计与之适配的数据处理架构,该架构将引入数据缓存、边缘处理等功能,并且涵盖EIoT中数据的清洗、过滤和融合等关键步骤。此外,在该架构基础上,需要同时满足大规模数据传输需求,尤其是将电力终端的能源效率(Energy Efficiency,EE)作为保障测量、监控、控制等多个电力运行环节超可靠低延迟通信(Ultra-Reliable and Low-Latency Communication,URLLC)的重要依据。在URLLC中,功率分配被认为是提高能效与数据处理效率的有效方法。然而,由于URLLC的特殊要求,传统香农公式在其中并不适用。因此,需要使用有限块长度编码理论来确保超可靠和低延迟的通信。文中解决了EIoT中URLLC的能效优化问题,并引入自适应深度神经网络,该技术可以根据不同电力设备接入数量,动态优化深度神经网络参数。深度神经网络将要优化的功率分配函数参数化,以无监督的方式离线训练,并可以在线部署以实现实时的功率分配结果。最后,仿真结果表明了所提方法在数据处理效率方面的有效性。展开更多
基金This project was supported by the National Natural Science Foundation of China (60135020).
文摘The flexibility of traditional image processing system is limited because those system are designed for specific applications. In this paper, a new TMS320C64x-based multi-DSP parallel computing architecture is presented. It has many promising characteristics such as powerful computing capability, broad I/O bandwidth, topology flexibility, and expansibility. The parallel system performance is evaluated by practical experiment.
文摘面向差异化业务需求,电力物联网(Electric Internet of Things,EIoT)需要设计与之适配的数据处理架构,该架构将引入数据缓存、边缘处理等功能,并且涵盖EIoT中数据的清洗、过滤和融合等关键步骤。此外,在该架构基础上,需要同时满足大规模数据传输需求,尤其是将电力终端的能源效率(Energy Efficiency,EE)作为保障测量、监控、控制等多个电力运行环节超可靠低延迟通信(Ultra-Reliable and Low-Latency Communication,URLLC)的重要依据。在URLLC中,功率分配被认为是提高能效与数据处理效率的有效方法。然而,由于URLLC的特殊要求,传统香农公式在其中并不适用。因此,需要使用有限块长度编码理论来确保超可靠和低延迟的通信。文中解决了EIoT中URLLC的能效优化问题,并引入自适应深度神经网络,该技术可以根据不同电力设备接入数量,动态优化深度神经网络参数。深度神经网络将要优化的功率分配函数参数化,以无监督的方式离线训练,并可以在线部署以实现实时的功率分配结果。最后,仿真结果表明了所提方法在数据处理效率方面的有效性。
