On optimal binary signed digit representations of integers

Binary signed digit representation （BSD-R） of an integer is widely used in computer arithmetic, cryptography and digital signal processing. This paper studies what the exact number of optimal BSD-R of an integer is and how to generate them entirely. We also show which kinds of integers have the maximum number of optimal BSD-Rs.

Unlicensed Personnel’s Experience of Digital Signing for Medication Administration in Municipal Healthcare

Aim: To describe unlicensed personnel’s experience of digital signing lists via a smartphone application for the distribution of medication in municipal healthcare in Western Sweden. Design: A qualitative and quantitative design was used. Methods: The study included 48 unlicensed personnel, 28 of whom answered an open-ended questionnaire, while an additional 20 volunteered for individual interviews. The material was analysed by qualitative content analysis. Results: The results indicate that digital signing lists via a smartphone application are feasible, and efficient and facilitate the work. However, some aspects negatively affected the sense of security, meetings with patients and quality of care, such as an insufficient internet signal in some rural areas, difficulty remembering the password, as well as the change of focus from patient to smartphone. To improve quality of care and the meeting with the patient, it is crucial that the technology works and that unlicensed personnel develop technical skills.

A Versatile 1.4-mW 6-bits CMOS ADC for Pulse-Based UWB Communication Systems

An ADC （analog to digital converter） using the low duty-cycle nature of pulse-based UWB （ultra wide-band） communications to reduce its power consumption is proposed. Implemented in CMOS （complementary metal-oxide-semiconductor） 180 nm technology, it can capture a 5 ns window at 4 GS/s each 100 ns, which corresponds to the acquisition of one UWB pulse at the pulse repetition rate of 10 Mpps （mega pulses per second）. By using time-interleaved RSD （redundant signed digit） ADCs, the complete ADC occupies only 0.15 mm2 and consumes only 1.4 mW from a 1.8 V power supply. The ADC can be operated in two modes using the same core circuits （OTA （operational transconductance amplifier）, comparators, etc.）. The first mode is the standard RSD doubling mode, while the second mode allows improving the signal-to-noise ratio by adding coherently the transmitted pulses of one symbol. For example, for audio applications, a 300 kbps data rate and processing gain up to 15 dB can be achieved at a clock frequency of 10 MHz.

Secure and efficient implementation of facial emotion detection for smart patient monitoring system

Background:Machine learning has enabled the automatic detection of facial expressions,which is particularly beneficial in smart monitoring and understanding the mental state of medical and psychological patients.Most algorithms that attain high emotion classification accuracy require extensive computational resources,which either require bulky and inefficient devices or require the sensor data to be processed on cloud servers.However,there is always the risk of privacy invasion,data misuse,and data manipulation when the raw images are transferred to cloud servers for processing facical emotion recognition(FER)data.One possible solution to this problem is to minimize the movement of such privatedata.Methods:In this research,we propose an efficient implementation of a convolutional neural network(CNN)based algorithm for on-device FER on a low-power field programmable gate array(FPGA)platform.This is done by encoding the CNN weights to approximated signed digits,which reduces the number of partial sums to be computed for multiply-accumulate(MAC)operations.This is advantageous for portable devices that lack full-fledged resourceintensivemultipliers.Results:We applied our approximation method on MobileNet-v2 and ResNet18 models,which were pretrained with the FER2013 dataset.Our implementations and simulations reduce the FPGA resource requirement by at least 22%compared to models with integer weight,with negligible loss in classification accuracy.Conclusions:The outcome of this research will help in the development of secure and low-power systems for FER and other biomedical applications.The approximation methods used in this research can also be extended to other imagebasedbiomedicalresearchfields.