SAS-SEINet: A SNR-Aware Adaptive Scalable SEI Neural Network Accelerator Using Algorithm-Hardware Co-Design for High-Accuracy and Power-Efficient UAV Surveillance.

As a potential air control measure, RF-based surveillance is one of the most commonly used unmanned aerial vehicles (UAV) surveillance methods that exploits specific emitter identification (SEI) technology to identify captured RF signal from ground controllers to UAVs. Recently many SEI algorithms based on deep convolution neural network (DCNN) have emerged. However, there is a lack of the implementation of specific hardware. This paper proposes a high-accuracy and power-efficient hardware accelerator using an algorithm-hardware co-design for UAV surveillance. For the algorithm, we propose a scalable SEI neural network with SNR-aware adaptive precision computation. With SNR awareness and precision reconfiguration, it can adaptively switch between DCNN and binary DCNN to cope with low SNR and high SNR tasks, respectively. In addition, a short-time Fourier transform (STFT) reusing DCNN method is proposed to pre-extract feature of UAV signal. For hardware, we designed a SNR sensing engine, denoising engine, and specialized DCNN engine with hybrid-precision convolution and memory access, aiming at SEI acceleration. Finally, we validate the effectiveness of our design on a FPGA, using a public UAV dataset. Compared with a state-of-the-art algorithm, our method can achieve the highest accuracy of 99.3% and an F1 score of 99.3%. Compared with other hardware designs, our accelerator can achieve the highest power efficiency of 40.12 Gops/W and 96.52 Gops/W with INT16 precision and binary precision.

[Naringin inhibits the proliferation and invasion of Eca109 esophageal cancer cells and promotes its apoptosis by blocking JAK/STAT signal pathway].

Objective To investigate the inhibitory effect of naringin (NAR) on proliferation and apoptosis of Eca109 esophageal cancer cells and its mechanism. Methods Eca109 cells were cultured with 0, 15, 30, 45 µmol/L NAR treatment for 24, 48 and 72 hours. The colony forming ability of Eca109 esophageal cancer cells was evaluated by cell colony forming assay, the cell proliferation activity was detected by MTT assay, and the invasion ability of cancer cells was detect by TranswellTM assay; Apoptosis was detected by annexin V-FITC/PI double labeling combined with flow cytometry. Western blot analysis was used to detect the protein expression of B-cell lymphoma factor 2 (Bcl2), Bcl2 related X protein (BAX), cytochrome C (CytC), caspase-3, caspase-9, Janus kinase 2 (JAK2), phosphorylated JAK2 (p-JAK2) and signal transducer and activator of transcription 3 (STAT3), phosphorylated STAT3 (p-STAT3), protein kinase B (AKT), phosphorylated AKT(p-AKT). Results NAR inhibited the proliferation and colony formation of Eca109 cells, suppressed the invasion of Eca109 cells and promoted the apoptosis of Eca109 cells; NAR promoted the expression of BAX, CytC, caspase-3, caspase-9, p-STAT3, p-AKT, AKT and p-JAK2 and inhibited the expression of Bcl2. Conclusion NAR can inhibit the proliferation, invasion, and colony formation of Eca109 cells and promote its apoptosis by blocking the activation of JAK/STAT signal pathway.