More Like this

1. Microcontroller Fingerprinting Using Partially Erased NOR Flash Memory Cells

   https://doi.org/10.1145/3448271  

   Poudel, Prawar ; Ray, Biswajit ; Milenkovic, Aleksandar ( April 2021 , ACM Transactions on Embedded Computing Systems)

   null (Ed.)

   Electronic device fingerprints, unique bit vectors extracted from device's physical properties, are used to differentiate between instances of functionally identical devices. This article introduces a new technique that extracts fingerprints from unique properties of partially erased NOR flash memory cells in modern microcontrollers. NOR flash memories integrated in modern systems-on-a-chip typically hold firmware and read-only data, but they are increasingly in-system-programmable, allowing designers to erase and program them during normal operation. The proposed technique leverages partial erase operations of flash memory segments that bring them into the state that exposes physical properties of the flash memory cells through a digital interface. These properties reflect semiconductor process variations and defects that are unique to each microcontroller or a flash memory segment within a microcontroller. The article explores threshold voltage variation in NOR flash memory cells for generating fingerprints and describes an algorithm for extracting fingerprints. The experimental evaluation utilizing a family of commercial microcontrollers demonstrates that the proposed technique is cost-effective, robust, and resilient to changes in voltage and temperature as well as to aging effects. 

   more » « less
2. An ultra-low energy internally analog, externally digital vector-matrix multiplier based on NOR flash memory technology

   https://doi.org/10.1145/3195970.3195989  

   Mahmoodi, M. Reza ; Strukov, Dmitri ( May 2018 , ACM/IEEE Design Automation Conference (DAC'18))

   Vector-matrix multiplication (VMM) is a core operation in many signal and data processing algorithms. Previous work showed that analog multipliers based on nonvolatile memories have superior energy efficiency as compared to digital counterparts at low-to-medium computing precision. In this paper, we propose extremely energy efficient analog mode VMM circuit with digital input/output interface and configurable precision. Similar to some previous work, the computation is performed by gate-coupled circuit utilizing embedded floating gate (FG) memories. The main novelty of our approach is an ultra-low power sensing circuitry, which is designed based on translinear Gilbert cell in topological combination with a floating resistor and a low-gain amplifier. Additionally, the digital-to-analog input conversion is merged with VMM, while current-mode algorithmic analog-to-digital circuit is employed at the circuit backend. Such implementations of conversion and sensing allow for circuit operation entirely in a current domain, resulting in high performance and energy efficiency. For example, post-layout simulation results for 400×400 5-bit VMM circuit designed in 55 nm process with embedded NOR flash memory, show up to 400 MHz operation, 1.68 POps/J energy efficiency, and 39.45 TOps/mm2 computing throughput. Moreover, the circuit is robust against process-voltage-temperature variations, in part due to inclusion of additional FG cells that are utilized for offset compensation. 

   more » « less
3. MLP+NeuroSimV3.0: Improving On-chip Learning Performance with Device to Algorithm Optimizations

   Yandong Luo, Xiaochen Peng ( July 2019 , International Conference on Neuromorphic Systems (ICONS))

   On-chip learning with compute-in-memory (CIM) paradigm has become popular in machine learning hardware design in the recent years. However, it is hard to achieve high on-chip learning accuracy due to the high nonlinearity in the weight update curve of emerging nonvolatile memory (eNVM) based analog synapse devices. Although digital synapse devices offer good learning accuracy, the row-by-row partial sum accumulation leads to high latency. In this paper, the methods to solve the aforementioned issues are presented with a device-to-algorithm level optimization. For analog synapses, novel hybrid precision synapses with good linearity and more advanced training algorithms are introduced to increase the on-chip learning accuracy. The latency issue for digital synapses can be solved by using parallel partial sum read-out scheme. All these features are included into the recently released MLP + NeuroSimV3.0, which is an in-house developed device-to-system evaluation framework for neuro-inspired accelerators based on CIM paradigm. 

   more » « less
4. A Ferroelectric FET based Processing-in-Memory Architecture for DNN Acceleration

   Long, Yun Long ; Kim, Daehyun ; Lee, Edward ; Saha, Priyabrata ; Mudassar, Burhan Ahmad ; She, Xueyuan ; Khan, Asif Islam ; Mukhopadhyay, Saibal ( June 2019 , IEEE journal on exploratory solid-state computational devices and circuits)

   This paper presents a Ferroelectric FET (FeFET) based processing-in-memory (PIM) architecture to accelerate inference of deep neural networks (DNNs). We propose a digital in-memory vector-matrix multiplication (VMM) engine design utilizing the FeFET crossbar to enables bit-parallel computation and eliminate analog-to-digital conversion in prior mixed-signal PIM designs. A dedicated hierarchical network-on-chip (H-NoC) is developed for input broadcasting and on-the-fly partial results processing, reducing the data transmission volume and latency. Simulations in 28nm CMOS technology show 115x and 6.3x higher computing efficiency (GOPs/W) over desktop GPU (Nvidia GTX 1080Ti) and ReRAM based design, respectively. 

   more » « less
5. Deep Neural Network Acceleration in Non-Volatile Memory: A Digital Approach

   https://doi.org/10.1109/NANOARCH47378.2019.181297  

   Angizi, Shaahin ; Fan, Deliang ( July 2019 , 10.1109/NANOARCH47378.2019.181297)

   Latest algorithmic development has brought competitive classification accuracy for neural networks despite constraining the network parameters to ternary or binary representations. These findings show significant optimization opportunities to replace computationally-intensive convolution operations (based on multiplication) with more efficient and less complex operations such as addition. In hardware implementation domain, processing-in-memory architecture is becoming a promising solution to alleviate enormous energy-hungry data communication between memory and processing units, bringing considerable improvement for system performance and energy efficiency while running such large networks. In this paper, we review several of our recent works regarding Processing-in-Memory (PIM) accelerator based on Magnetic Random Access Memory computational sub-arrays to accelerate the inference mode of quantized neural networks using digital non-volatile memory rather than using analog crossbar operation. In this way, we investigate the performance of two distinct in-memory addition schemes compared to other digital methods based on processing-in-DRAM/GPU/ASIC design to tackle DNN power and memory wall bottleneck. 

   more » « less