Content for TR 26.942 Word version: 19.0.0

1… 4… 4.2… 5… 6… 7… 7.3… 7.4… 7.5… 7.6… 7.7… 7.8… 7.9… 7.10… 8… 9…

7.3 Solution #2: Evaluation Framework to measure energy efficiency of a UE 7.3.1 Key Issue mapping 7.3.2 Functional description 7.3.3 Procedures 7.3.4 Summary
...

7.3 Solution #2: Evaluation Framework to measure energy efficiency of a UE p. 39

7.3.1 Key Issue mapping p. 39

This Candidate Solution addresses Key Issue #3 (Evaluation Framework) described in clause 6.3.

7.3.2 Functional description p. 39

7.3.2.1 UE Energy Efficiency metric p. 39

Similar to the definition of the Mobile Network Energy Efficiency KPI (see clause 4.2.3.6.2), one possible example to represent the UE energy efficiency is bits per Joule. A UE is considered to be more energy-efficient when the data usage (in bits) is high with less energy consumption (in Joules), assuming that battery capacity and consumption rate are fixed when assessed during the time frame (T). The metric is defined as below:

Data Volume [bit] represents the traffic volume for the UE for a given duration.
Energy Consumption [estimated Joule] represents the total energy consumption of the UE for a given duration, which is the summation of power consumption.

The data volume consumed by k sub-applications is defined as the summation of the data volume consumed by the User Equipment under investigation during the time frame T of the energy consumption assessment for these sub-applications. This could be shown as:

formula for data volume over the measurement period T

where DV_k,app, measured in bits, is the data volume over the measurement period T of the UE for a given duration.

The overall UE Energy Efficiency, EEUE is then calculated as follows:

formula for overall UE Energy Efficiency

7.3.3 Procedures p. 39

The following methodology is proposed to measure the Energy Efficiency of a test application running on a UE:

A test scenario is defined, and test conditions described:
1. Network (connection type, upload and download bandwidth, latency).
2. User devices (type, model, SoC, OS version, video player). At least 2 devices should be used.
3. Test environment (number of devices, number of iterations, etc.).
4. Anchor against which the specific features will be evaluated (i.e., 5GMS service delivering a 720p video at 2 Mbps in HEVC).
5. Reference sequence(s) used.
Launch the application under test which implements the collection of energy-related information, including either battery discharge rate or else total energy discharged and duration.
The test is done for the anchor and the implementation including the feature evaluated.
Extract data from the data collector for non-real-time analysis. Characterization is documented in terms of expected energy savings, and may include additional comparison parameters such as impact on the end user's Quality of Experience, etc.
The time duration of the measurement, denoted as T, shall be one of the following alternatives:
- Weekly measurement: T = 7 days.
- Monthly measurement: T = 30 days.
- Yearly measurement: T = 365 days.
The minimum duration is therefore one week: monthly and yearly measurements are extensions of the basic week test per the guidelines in ETSI ES 203 228 [66].
The Energy Consumption of the mobile network is measured by means of metering information provided by utility suppliers or by measurement systems integrated into the mobile network. Moreover, sensors can be used to measure the energy consumption of individual sites or pieces of equipment, following the requirements set by ETSI ES 202 336-12 [68].
The data volume is measured using network counters for data volume in the UE under test.
- For Packet-Switched traffic, the data volume is considered as the overall amount of data transferred to and from the UE under test. Data volume is measured separately for each application present in the UE.
- For Circuit-Switched traffic (e.g. CS voice or VoLTE), the data volume is considered as the number of minutes of communications during the time T multiplied by the data rate of the corresponding service and the call success rate per ETSI ES 203 228 [66].

7.3.4 Summary p. 40

The Candidate Solution described in clauses 7.2.2 and 7.2.3 has the following limitations:

The solution proposes to evaluate the energy efficiency of the entire UE as a whole device, while individual energy consuming aspects such as the media energy consumption, energy consumed during radio transmissions (4G, 5G, Wi-Fi), etc. may have additional impact on the energy efficiency of the UE.
The solution is UE implementation-specific, i.e. the same feature may result in different evaluations of energy efficiency on different UEs.
The solution may result in different evaluations of energy efficiency of the same UE depending upon the test conditions that may vary over time (e.g. environmental changes and radio conditions, etc.).