Multiplexers (MUX) are essential elements in FieldProgrammable Gate Arrays (FPGA), widely used in practical applications. Due to the LUT-based architecture of FPGAs, multiplexers that switch among many signals or operate on large word sizes incur significant resource costs, as these costs scale linearly with the data word size. Vivado's automatic synthesis flow often produces sub-optimal MUX implementations, necessitating hand-crafted solutions to minimize resource overhead. Here, we present three MUX implementation schemes that reduce resource usage for various input signal counts. These optimizations enable enhanced resource efficiency in applications ranging from circuits generated by High-Level Synthesis (HLS) tools to optimized digital filters and artificial neural networks.

We present a training tool flow for deep neural networks (DNN) optimized for a hardware-efficient FPGA-implementation based on reconfigurable constant-coefficient multipliers (RCCMs). RCCMs replace the costly generic multipliers by shift-and-add operations. In previous work, it was shown that RCCMs offer a better alternative for saving FPGA area than utilizing low-precision arithmetic. This work proposes an improved tool flow that enables layer-wise weight quantization, a larger search space by additional RCCM coefficient sets and an optimized retraining. This leads to an improved accuracy compared to the previous method. In addition, hardware requirements are lower as only 1 to 3 adders per multiplication are used. This reduces the overall complexity and the required memory bandwidth simultaneously. We evaluate our tool flow using multiple networks (ResNets) on the ImageNet data set.

Low-precision arithmetic operations to accelerate deep-learning applications on field-programmable gate arrays (FPGAs) have been studied extensively, because they offer the potential to save silicon area or increase throughput. However, these benefits come at the cost of a decrease in accuracy. In this article, we demonstrate that reconfigurable constant coefficient multipliers (RCCMs) offer a better alternative for saving the silicon area than utilizing low-precision arithmetic. RCCMs multiply input values by a restricted choice of coefficients using only adders, subtractors, bit shifts, and multiplexers (MUXes), meaning that they can be heavily optimized for FPGAs. We propose a family of RCCMs tailored to FPGA logic elements to ensure their efficient utilization. To minimize information loss from quantization, we then develop novel training techniques that map the possible coefficient representations of the RCCMs to neural network weight parameter distributions. This enables the usage of the RCCMs in hardware, while maintaining high accuracy. We demonstrate the benefits of these techniques using AlexNet, ResNet-18, and ResNet-50 networks. The resulting implementations achieve up to 50% resource savings over traditional 8-bit quantized networks, translating to significant speedups and power savings. Our RCCM with the lowest resource requirements exceeds 6-bit fixed point accuracy, while all other implementations with RCCMs achieve at least similar accuracy to an 8-bit uniformly quantized design, while achieving significant resource savings.

Convolutional neural networks (CNNs) gained great success in machine learning applications and much attention was paid to their acceleration on field programmable gate arrays (FPGAs). The most demanding computational complexity of CNNs is found in the convolutional layers, which account for 90% of the total operations. The fact that parameters in convolutional layers do not change over a long time interval in weight stationary CNNs allows the use of reconfiguration to reduce the resource requirements. This work proposes several alternative reconfiguration schemes that significantly reduce the complexity of sum-of-products operations. The proposed direct configuration schemes provide the least resource requirements and fast reconfiguration times of 32 clock cycles but require additional memory for the pre-computed configurations. The proposed online reconfiguration scheme uses an online computation of the LUT contents to avoid this memory overhead. Finally, a scheme that duplicates the reconfigurable LUTs is proposed for which the reconfiguration time can be completely hidden in the computation time. Combined with a few online reconfiguration circuits, this provides the same configuration memory and configuration time as a conventional parallel kernel but offers large resource reductions of up to 80% of the LUTs.

The computational complexity of neural networks for large-scale or real-time applications necessitates hardware acceleration. Most approaches assume that the network architecture and parameters are unknown at design time, permitting usage in a large number of applications. This article demonstrates, for the case where the neural network architecture and ternary weight values are known a priori , that extremely high throughput implementations of neural network inference can be made by customising the datapath and routing to remove unnecessary computations and data movement. This approach is ideally suited to FPGA implementations as a specialized implementation of a trained network improves efficiency while still retaining generality with the reconfigurability of an FPGA. A VGG-style network with ternary weights and fixed point activations is implemented for the CIFAR10 dataset on Amazon’s AWS F1 instance. This article demonstrates how to remove 90% of the operations in convolutional layers by exploiting sparsity and compile-time optimizations. The implementation in hardware achieves 90.9 ± 0.1% accuracy and 122k frames per second, with a latency of only 29µs, which is the fastest CNN inference implementation reported so far on an FPGA.

Constant matrix multiplication (CMM), i.e., the multiplication of a constant matrix with a vector, is a common operation in digital signal processing. It is a generalization of multiple constant multiplication (MCM) where a single variable is multiplied by a constant vector. Like MCM, CMM can be reduced to additions/subtractions and bit shifts. Finding a circuit with minimal number of add/subtract operations is known as the CMM problem. While this leads to a reduction in circuit area it may be less efficient for power consumption or throughput. It is well studied for the MCM problem that a) reducing the adder depth (AD) leads to a reduced power consumption and b) pipeline resources have to be considered during optimization to enhance throughput without wasting area. This paper addresses the optimization of CMM circuits which considers both adder depth and pipelining for the first time. For that, a heuristic is proposed which evaluates the most attractive graph topologies. It is shown that the proposed method requires 12.5% less adders with min. AD and 38.5% less pipelined operations. Synthesis results for recent FPGAs show that these reductions also translate to superior results in terms of delay and power consumption compared to the state-of-the-art.

The scaling operation, i. e., the multiplication with a single constant is a frequently used operation in many kinds of numeric algorithms. The multiple constant multiplication (MCM) is a generalization where a variable is multiplied by several constants. This kind of operation is heavily used, e. g., in digital filters or discrete transforms. It was shown in recent work that small, fast and power efficient MCM implementations can be realized by using the fast carry chains of FPGAs rather than wasting specialized embedded multipliers. However, in the work so far, only common two-input adders were used. As FPGAs today support ternary adders, i. e., adders with three inputs, this work investigates the optimization of pipelined MCM circuits which include ternary adders. It is shown experimentally that 27% less operations are needed on average by using ternary adders, resulting in 15% slice (Xilinx) and 10% ALM (Altera) reductions, respectively.