As a typical signal processing problem, direction-of-arrival (DOA) estimation has been adapted to a wide range of applications in radar-based systems. A high DOA resolution requires a large number of antenna elements which increases the overall cost. To minimise the cost, it is desirable to choose an optimum sub-array from a full array. To enable cognition, the subarrays are selected based on the present target scenario. By using deep learning (DL) based techniques, the authors show a cognitive sparse array selection technique. By using hardware simulations, they demonstrate the applicability of the deep learning (DL)-based sparse antenna selection network in direction-of-arrival (DOA) estimation problems. They show that the DL-based sub-arrays lead to a higher direction-of-arrival (DOA) estimation accuracy by 6 dB over random array selection.

中文翻译：

---

具有稀疏阵列天线的认知雷达的硬件原型演示

作为典型的信号处理问题，到达方向 (DOA) 估计已适应于基于雷达的系统中的广泛应用。高 DOA 分辨率需要大量天线元件，这会增加总成本。为了最小化成本，需要从全阵列中选择最佳子阵列。为了实现认知，根据当前的目标场景选择子阵列。通过使用基于深度学习 (DL) 的技术，作者展示了一种认知稀疏阵列选择技术。通过使用硬件模拟，他们证明了基于深度学习 (DL) 的稀疏天线选择网络在到达方向 (DOA) 估计问题中的适用性。他们表明，与随机阵列选择相比，基于 DL 的子阵列可提高 6 dB 的到达方向 (DOA) 估计精度。