In this paper, a thorough characterization of phase-change memory (PCM) cells was carried out, aimed at evaluating and optimizing their performance as enabling devices for analog in-memory computing (AIMC) applications. Exploiting the features of programming pulses, we discuss strategies to reduce undesired phenomena that afflict PCM cells and are particularly harmful in analog computations, such as low-frequency noise, time drift, and cell-to-cell variability of the conductance. The test vehicle is an embedded PCM (ePCM) provided by STMicroelectronics and designed in 90-nm smart power BCD technology with a Ge-rich Ge-Sb-Te (GST) alloy for automotive applications. On the basis of the results of the characterization of a large number of cells, we propose an iterative algorithm to allow multi-level cell conductance programming, and its performances for AIMC applications are discussed. Results for a group of 512 cells programmed with four different conductance levels are presented, showing an initial conductance spread under 6%, relative current noise less than 9% in most cases, and a relative conductance drift of 15% in the worst case after 14 h from the application of the programming sequence.

This paper presents an integrated peripheral unit interfaced to an embedded Phase-change Memory (ePCM) macrocell, with the aim of adding Analog In-memory Computing (AIMC) feature without any modifications to the internal structure of the memory array. The testchip has been designed and manufactured in a 90-nm STMicroelectronics CMOS technology. The unit allows the execution of signed Multiply and Accumulate (MAC) operations at the edge of the memory array exploiting the physical characteristics of memory devices. I-V characteristic non-linearity and transconductance time drift of PCM cells are overcome through a regulated bitline readout circuitry with time-coded inputs, along with a drift compensation technique based on a conductance ratio. Measurements results show 1-σ accuracy of 95.56% in MAC operations, whose decrease in time is roughly negligible at room temperature and is less than 1% after 24 hours at 85°C bake.

Matrix-Vector Multiplications (MVMs) represent a heavy workload for both training and inference in Deep Neural Networks (DNNs) applications. Analog In-memory Computing (AIMC) systems based on Phase Change Memory (PCM) has been shown to be a valid competitor to enhance the energy efficiency of DNN accelerators. Although DNNs are quite resilient to computation inaccuracies, PCM non-idealities could strongly affect MVM operations precision, and thus the accuracy of DNNs. In this paper, a combined hardware and software solution to mitigate the impact of PCM non-idealities is presented. The drift of PCM cells conductance is compensated at the circuit level through the introduction of a conductance ratio at the core of the MVM computation. A model of the behaviour of PCM cells is employed to develop a device-aware training for DNNs and the accuracy is estimated in a CIFAR-10 classification task. This work is supported by a PCM-based AIMC prototype, designed in a 90-nm STMicroelectronics technology, and conceived to perform Multiply-and-Accumulate (MAC) computations, which are the kernel of MVMs. Results show that the MAC computation accuracy is around 95% even under the effect of cells drift. The use of a device-aware DNN training makes the networks less sensitive to weight variability, with a 15% increase in classification accuracy over a conventionally-trained Lenet-5 DNN, and a 36% gain when drift compensation is applied.