Coarse-grained Reconfigurable Arrays (CGRAs) are domain-agnostic accelerators that enhance the energy efficiency of resource-constrained edge devices. The CGRA landscape is diverse, exhibiting trade-offs between performance, efficiency, and architectural specialization. However, CGRAs often overprovision communication resources relative to their modest computing capabilities. This occurs because the theoretically provisioned programmability for CGRAs often proves superfluous in practical implementations.

Recent studies have demonstrated the significant potential of Large Language Models (LLMs) in generating Register Transfer Level (RTL) code, with notable advancements showcased by commercial models such as GPT-4 and Claude3-Opus. However, these proprietary LLMs often raise concerns regarding privacy and security. While open-source LLMs offer solutions to these concerns, they typically underperform commercial models in RTL code generation tasks, primarily due to the scarcity of high-quality open-source RTL datasets. To address this challenge, we introduce OriGen, a fully open-source framework that incorporates self-reflection capabilities and a novel dataset augmentation methodology for generating high-quality, large-scale RTL code. Our approach employs a code-to-code augmentation technique to enhance the quality of open-source RTL code datasets. Furthermore, OriGen can rectify syntactic errors through a self-reflection process that leverages compiler feedback.

Cognitive Radio (CR) enabled vehicles in Vehicle Networks can use multiple Radio Access Networks (RANs) for data transmission. The simultaneous use of multiple RANs for transmission requires the design of a specific multi-path transmission protocol. Many scholars have studied the scheduling algorithm to improve the quality of multi-path transmission. However, most of the existing scheduling algorithms are difficult to deal with the challenges brought by the diversity and heterogeneity of the vehicle network. To deal with these challenges, this paper proposes an IP layer Deep Reinforcement Learning (DRL) multi-path transmission scheduling algorithm named Adaptive Dynamic Scheduling Algorithm (ADSA), which can dynamically generate the optimal scheduling policy through the interaction between agent and network environment. This paper first models the data packet scheduling strategy of multi-path transmission into an optimization problem of multi-path transmission efficiency. Then this paper transforms the optimization problem into a DRL problem and finds the optimal scheduling strategy through DRL model training. This paper evaluates the network performance of ADSA in different network scenarios compared with traditional scheduling algorithms. Simulation results show that ADSA increases the throughput by 8.9 Mbps compared with the three traditional scheduling algorithms and reduces the transmission delay by 4.3 ms.