Homomorphic Encryption (HE) technology allows for processing encrypted data, breaking through data isolation barriers and providing a promising solution for privacy-preserving computation. The integration of HE technology into Convolutional Neural Network (CNN) inference shows potential in addressing privacy issues in identity verification, medical imaging diagnosis, and various other applications. The CKKS HE algorithm stands out as a popular option for homomorphic CNN inference due to its capability to handle real number computations. However, challenges such as computational delays and resource overhead present significant obstacles to the practical implementation of homomorphic CNN inference, largely due to the complex nature of HE operations. In addition, current methods for speeding up homomorphic CNN inference primarily address individual images or large batches of input images, lacking a solution for efficiently processing a moderate number of input images with fast homomorphic inference capabilities. In response to these challenges, we introduce a novel leveled homomorphic CNN inference scheme aimed at reducing latency and improving throughput using the CKKS scheme. Our proposed inference strategy involves mapping multiple inputs to a set of ciphertext by exploiting the sliding window properties of convolutions to utilize CKKS's inherent Single-Instruction-Multiple-Data (SIMD) capability. To mitigate the delay associated with homomorphic CNN inference, we introduce optimization techniques, including mask-weight merging, rotation multiplexing, stride convolution segmentation, and folding rotations. The efficacy of our homomorphic inference scheme is demonstrated through evaluations carried out on the MNIST and CIFAR-10 datasets. Specifically, results from the MNIST dataset on a single CPU thread show that inference for 163 images can be completed in 10.4 seconds with an accuracy of 98.95%, which is a 6.9× throughput improvement over state-of-the-art works. Comparative analysis with existing methodologies highlights the superior performance of our proposed inference scheme in terms of latency, throughput, communication overhead, and memory utilization.

Implementing compute-in-memory (CIM) architectures on FPGA offers an effective solution to the von Neumann bottleneck by enabling fast configuration and computation directly within memory. Traditional custom solutions rely on the modification of block RAM (BRAM) to implement memory computing. However, single-word-line activation of BRAM results in low parallelism, and the need for additional adder trees to accumulate partial sums further limits efficiency. To overcome these limitations, we propose a CIM core based on a 2T1C structure as a replacement for BRAM units. This core utilizes a charge redistribution mechanism and reuse of ADC capacitors, achieving high parallelism, low power consumption, and a compact area. By incorporating computational capabilities within a single cell, our design enables dual parallelism, further enhancing performance and efficiency. In addition, we present an automated deployment and mapping tool for deep neural networks (DNNs) on FPGA, allowing users to rapidly develop FPGA-based solutions for different network architectures. Compared to state-of-the-art solutions, our design achieves a peak throughput improvement of 4.5× and a reduction in area by 53%.

Field Programmable Gate Array (FPGA) is a versatile and programmable hardware platform, which makes it a promising candidate for accelerating Deep Neural Networks (DNNs). However, FPGA’s computing energy efficiency is low due to the domination of energy consumption by interconnect data movement. In this article, we propose an all-digital Compute-in-memory FPGA architecture for deep learning acceleration. Furthermore, we present a bit-serial computing circuit of the Digital CIM core for accelerating vector-matrix multiplication (VMM) operations. A Network-CIM-deployer ( NCIMD ) is also developed to support automatic deployment and mapping of DNN networks. NCIMD provides a user-friendly API of DNN models in Caffe format. Meanwhile, we introduce a Weight-stationary dataflow and describe the method of mapping a single layer of the network to the CIM array in the architecture. We conduct experimental tests on the proposed FPGA architecture in the field of Deep Learning (DL), as well as in non-DL fields, using different architectural layouts and mapping strategies. We also compare the results with the conventional FPGA architecture. The experimental results show that compared to the conventional FPGA architecture, the energy efficiency can achieve a maximum speedup of 16.1×, while the latency can decrease up to 40% in our proposed CIM FPGA architecture.

Seismic wave suffers from amplitude attenuation and phase distortion when propagating in the attenuating media, thus reverse time migration (RTM) for viscous media should take the attenuation effects into consideration. Compensating for the attenuation effects in RTM may occur numerical instability because of the exponential amplification of the extrapolated wavefields. To obtain stable imaging results, we have developed a stabilized Q-compensated reverse time migration in the frequency domain. This algorithm is implemented by the following steps: first, we use the Kolsky-Futterman model to derive a frequency-domain viscoacoustic wave equation, which can simulate the amplitude loss and phase dispersion effects separately. Then, we calculate the source wavefields in the viscoacoustic media. Next, treating the recorded (viscoacoustic) data as the receiver sources, we can obtain the phase-dispersion-only and viscoacoustic receiver wavefields, which can be used to construct the stabilized Q-compensated receiver wavefields. Finally, we apply the deconvolution imaging condition for obtaining a Q-compensated image. A simple anticline model and gas chimney model are used to verify the effectiveness of the proposed approach. The Q-compensated images for the noise-free data indicate the algorithmic stability and compensation accuracy of the proposed scheme. The noisy data tests for the gas chimney model demonstrate the good anti-noise property of our method. The field data applications further prove its feasibility and practicability.