Part one of this FAQ series focused on the various types of components available for designing power systems for wireless Internet of Things (IoT) sensor nodes, wearables, and similar applications. This FAQ focuses on measuring and validating the performance of those power systems. There are several instruments and approaches available for measuring and testing the power consumption of wireless IoT sensor nodes. Choices include digital multimeters (DMMs), power analyzers, multi-channel probes, oscilloscopes, intelligent power supply source measurement units, and so on.
Designing appropriate hardware and software is key for low-power devices, for example implementing optimal power consumption in active mode, but also deep sleep modes or short startup/shutdown phases. Power consumption also strongly depends on the use of power-saving features, application behavior, and interaction with the wireless network. Particularly IoT devices that use wireless low-power technologies (LP-WAN) such as LTE-M or NB-IoT require best-in-class designs. They also need to consider all aspects affecting the power consumption of the operational modes.
Device and application developers require very accurate power measurement solutions with a high dynamic range – from a few microamperes to several amps. There is also a need for long-term measurements covering the different operational modes and various network configurations.
Multi-level active mode power profile
Accurate capture of the current waveform is critical for power consumption calculation and can be accomplished on the DMM7510 7½-digit graphical sampling multimeter from Keithley. The DMM7510 combines a precision, high-resolution digital multimeter (DMM), graphical touchscreen display, and high-speed, high-resolution digitizer to create the first graphical sampling multimeter. With pA-level sensitivity and 1Msample/s sampling, it accurately measures ultra-low sleep mode currents and transmits drain currents from wireless devices.
The figure below illustrates a typical multi-level active mode current pulse captured by the DMM7510. The active mode typically contains current levels as high as tens of milliamps. These multiple current levels represent events such as start-up and shut-down sequences, transmit and receive modes, sensor data conversion, etc.
High dynamic range current measurements
Long battery life is crucial for modern IoT and mobile devices. In order to keep energy consumption as low as possible, these devices typically work with special sleep modes that consume very little power and are only interrupted by very short activity phases of normal or high power consumption. For a device to succeed, its power
consumption has to be optimized in the early development phase. It needs to be measured accurately, which requires sophisticated probing solutions. The probe must be able to simultaneously measure very small currents in the μA or even nA range as well as currents up to several amperes. Handling such a high dynamic range of 106 or even up to 109 is a challenge for every measurement device.
The R&SRT-ZVC02/-ZVC04 multi-channel probes from Rhode & Schwarz are designed for battery life measurements on low power consumption devices. They simultaneously measure current with high dynamic range and high resolution in all device activity phases. To operate the multi-channel probe, an R&SRTE1000, R&SRTO2000, or R&SRTP oscilloscope is required. With up to four current and four voltage input channels, each with an 18 bit ADC resolution, the R&SRT-ZVC02/-ZVC04 multi-channel power probes provide the dynamic range needed to analyze current consumption. Three built-in shunts and an external shunt mode in combination with switchable gain factors help optimize the input current range.
Benchmarking transport layer security power consumption for IoT edge nodes
EEMBC SecureMark is an objective, standardized benchmarking framework for measuring the efficiency of cryptographic processing solutions. Within SecureMark, EEMBC plans to support the test and analysis of various security profiles for different application domains. The first of these to be available for licensing is SecureMark-TLS, which focuses on Transport Layer Security (TLS) for IoT edge nodes.
The TLS protocol provides privacy and integrity of the exchanged messages and can authenticate the communicating parties. SecureMark-TLS measures the performance and energy consumption of a physical device (development board or end-product) for a prescribed set of cryptographic functions. The benchmark uses a common IoT cipher suite comprised of elliptic curve cryptography for key exchange and digital signing, and standard primitives such as SHA256 and AES128, in both CCM and ECB modes. The energy measurements are aggregated into a final, single score that is representative of the TLS operations for an IoT edge node device.
Complementary cumulative distribution function analysis of battery drain
A complementary cumulative distribution function (CCDF) is an important part of analyzing long-term power consumption and predicting battery life in wireless IoT nodes, wearables, and similar devices. The CCDF curve shows the amount of time power consumption spends above the average power level of the measuring device, or equivalently, the probability that the power will be above the average power level. Accurately developing a CCDF requires a combination of measurement hardware and corresponding software to analyze the resulting time-domain measurements.
A source measure unit (SMU) is an intelligent power supply that can deliver power to the device under test while measuring current consumption and evaluating the results, including battery drain analysis. Portable devices use idle and sleep states to minimize current and extend battery life. An active state that transfers data can use over 1000 times more current. Seamless current ranging captures large active currents and small idle currents in the same waveform. SMUs enable designers to source and measure currents accurately down to microampere and nano ampere.
The Keysight N6705B dc power analyzer combines with the N6781A SMU module can simulate, measure, and analyze dynamic conditions including power sequencing, battery droop, and various supply variations. Because it supplies the power, it can measure it both accurately (0.025% up to 18 bits) and quickly (100kHz). The N6705B can behave like an oscilloscope so the designer has a familiar operating model to rapidly explore circuit behavior. It can also behave like a data logger to record long term circuit power consumption.
Software packages, such as the Keysight 14585A Control and Analysis Software, can add additional capabilities to the designer’s toolkit, allowing fast connection setup and measurement of the device’s most important characteristics. For example, the 14585A software can perform a CCDF analysis, which concisely displays short- and long-term battery drain measurements.
The previous FAQ in this series considered the various types of components available when designing power systems for wireless IoT nodes and wearables. This FAQ focused on measuring and validating the performance of those power systems. The next FAQ will review options for batteries and charging for wireless IoT nodes and wearables.
References
Battery Drain Analysis for Low Power IoT Devices (white paper), Keysight Technologies
Complementary cumulative distribution function, Cadence
High dynamic range current measurements on IoT devices, Rohde & Schwarz
Measuring ultra-low power in wireless sensor node applications using the model DMM7510 7½-digit graphical sampling multimeter, Keithley