State-of-the-art performance for many emerging edge applications is achieved by deep neural networks (DNNs). Often, these DNNs are location and time sensitive, and must be delivered from an edge server to the edge device rapidly and efficiently to carry out time-sensitive inference tasks. In this paper, we introduce AirNet, a novel training and transmission method that allows efficient wireless delivery of DNNs under stringent transmit power and latency constraints. We first train the DNN with noise injection to counter the channel noise. We then employ pruning to reduce the network size to the available channel bandwidth, and perform knowledge distillation from a large model to improve the performance. We show that AirNet achieves significantly higher test accuracy compared to digital alternatives under the same bandwidth and power constraints. We further improve the performance of AirNet by pruning the network below the available bandwidth, and using channel expansion to provide better robustness against channel noise. We also benefit from unequal error protection (UEP) by selectively expanding more important layers of the network. Finally, we develop an ensemble training approach, which trains a whole spectrum of DNNs, each of which can be used at different channel condition, resolving the impractical memory requirements.
翻译:许多新兴边缘应用的最新性能是通过深神经网络(DNN)实现的。这些DNN通常具有位置和时间敏感性,并且必须从边缘服务器迅速高效地交付到边缘设备,以便执行时间敏感推导任务。在本文件中,我们引入了AirNet,这是一个新颖的培训和传输方法,允许在严格的传输功率和潜伏限制下高效无线传送DNN;我们首先用噪音注入来训练DNNN,以对抗频道噪音;然后我们使用螺旋线,将网络规模缩小到现有的频道带宽,并从一个大模型中进行知识蒸馏,以改善性能。我们显示AirNet在相同的带宽和功率限制下,与数字替代方法相比,其测试精度要高得多。我们进一步改进AirNet的性能,将网络控制在现有的频带下,并利用频道扩展来提供更好的抗声响。我们还受益于不平等的错误保护(UEP),方法是有选择地扩大网络的重要层。最后,我们开发一个混合的培训方法,以培养DNNNM的全频谱,每个频道都可用于不切不切不现实的存储。