Recognizing symbols in architectural CAD drawings is critical for various advanced engineering applications. In this paper, we propose a novel CAD data annotation engine that leverages intrinsic attributes from systematically archived CAD drawings to automatically generate high-quality annotations, thus significantly reducing manual labeling efforts. Utilizing this engine, we construct ArchCAD-400K, a large-scale CAD dataset consisting of 413,062 chunks from 5538 highly standardized drawings, making it over 26 times larger than the largest existing CAD dataset. ArchCAD-400K boasts an extended drawing diversity and broader categories, offering line-grained annotations. Furthermore, we present a new baseline model for panoptic symbol spotting, termed Dual-Pathway Symbol Spotter (DPSS). It incorporates an adaptive fusion module to enhance primitive features with complementary image features, achieving state-of-the-art performance and enhanced robustness. Extensive experiments validate the effectiveness of DPSS, demonstrating the value of ArchCAD-400K and its potential to drive innovation in architectural design and construction.

We study the task of panoptic symbol spotting, which involves identifying both individual instances of countable \textit{things} and the semantic regions of uncountable \textit{stuff} in computer-aided design (CAD) drawings composed of vector graphical primitives. Existing methods typically rely on image rasterization, graph construction, or point-based representation, but these approaches often suffer from high computational costs, limited generality, and loss of geometric structural information. In this paper, we propose \textit{VecFormer}, a novel method that addresses these challenges through \textit{line-based representation} of primitives. This design preserves the geometric continuity of the original primitive, enabling more accurate shape representation while maintaining a computation-friendly structure, making it well-suited for vector graphic understanding tasks. To further enhance prediction reliability, we introduce a \textit{Branch Fusion Refinement} module that effectively integrates instance and semantic predictions, resolving their inconsistencies for more coherent panoptic outputs. Extensive experiments demonstrate that our method establishes a new state-of-the-art, achieving 91.1 PQ, with Stuff-PQ improved by 9.6 and 21.2 points over the second-best results under settings with and without prior information, respectively—highlighting the strong potential of line-based representation as a foundation for vector graphic understanding.

A promising artificial intelligence (AI)-driven cross-layer modulation is proposed to perform the bit-symbol mapping for the multiple-input-multiple-output (MIMO) transmission mode in the next generation wireless communication system. Unlike the prior modulation technologies, multiple influence factors including he constellation mapping, the number of spatial layers, layer mapping and power allocation among spatial layers are simultaneously taken into account in the proposed method pursuing for spectral efficiency maximization via auto-encoder (AE) model. In this paper, a joint optimization subject to influence factors above for the cross-layer modulation is fulfilled based on the AE framework for simplifying the design complexity and achieving a better performance. Simulation results demonstrate that the block error rate (BLER) performance of the proposed cross-layer modulation is beyond the traditional scheme by around 1 dB as a result of the learned bit-symbol mapping for spatial layers.

Cross-link interference (CLI) is a prevalent type of interference in dynamic time division duplex (TDD) systems. Mitigating CLI is crucial for effective system design in contemporary wireless communication systems. Proper mitigation of CLI can enhance transmission and reception robustness, improve spectral efficiency, and overall throughput. Consequently, there is a need for innovative methods and approaches to address CLI mitigation. Therefore, this paper proposes and evaluates a novel beam nulling method based on block diagonalization (BD). The method is introduced and evaluated using numerical results. Analysis and numerical results demonstrate the superior performance of the proposed method in dynamic TDD scenarios.

Non-orthogonal multiple access (NOMA) is envisioned to be one of the promising radio access techniques for the fifth generation (5G) mobile networks. In this paper, a portable NOMA testbed based on software defined radio (SDR) is developed by transporting our NOMA system to mini personal computers (PCs). Moreover, the NOMA testbed has been enhanced from 5 MHz bandwidth to 10 MHz bandwidth. As the computation complexity grows higher with the increase of system bandwidth, the portable NOMA testbed is not competent to fully demonstrate the performance of our NOMA system due to the limitation of mini PC. For a better understanding of the NOMA testbed, the system architecture and scenario are introduced in brief. Then, the implementation of NOMA transceiver and the protocol stack are described separately. Finally, a series of experiments are carried out to evaluate its performance loss compared with the original NOMA system based on desktops. The experimental results indicate that the performance of processors may be a short slab for the development of portable SDR-based testbeds towards 5G networks.

Software Defined Radio (SDR) system has great advantages of high flexibility and better reconfigurability in simulating and verifying the new communication technologies. Among all the SDR systems based on general purpose processor (GPP), OpenAirInterface (OAI) is one of the most comprehensive and competitive open-source SDR systems. Due to the design of signal-thread processing in the system, the user equipment (UE) of OAI still cannot process the vast amounts of data when the bandwidth is up to 20 MHz in the real-time experiments. This paper provides an overview of the system architecture in the real-time experimentation of OAI and analyzes its baseband signal processing. Then, the authors of this paper propose and implement a method of multithread parallel processing on the UE of OAI. Over-the-air experiments are carried out with the proposed method for the purpose of performance evaluation in the real- time wireless environment. Results demonstrate that the proposed method enhances the performance of the OAI system and shows its potential for the fifth generation (5G) technology.