Page 92 - AI for Good-Innovate for Impact

AI for Good-Innovate for Impact

Data Interaction: The AI Prediction Model interacts with the Multicore Processor to exchange
data. This interaction likely involves providing input data for prediction generation and receiving
output predictions for further processing or action.

Radio Link Configuration: The predictions generated by the AI Prediction Model are utilized
for radio link configuration. This configuration may involve adjusting parameters such as
modulation schemes, transmission power levels, or antenna beamforming to optimize the
performance of the radio links within the network.

18�3� Use case requirements

• UC18-REQ01: Continuous CPU Utilization Monitoring - operational workload and Radio
Unit (RU) occupancy levels.
• UC18-REQ02: Data Analysis - Time series data, historical data, patterns
• UC18-REQ03: Use of statistical algorithms
• UC18-REQ04: Documentations to train the data
• UC18-REQ05: Programmable baseband is required
• UC18-REQ6: Day to day telecom operations

18�4� Sequence diagram

18�5� References

[1] https:// www .sciencedirect .com/ science/ article/ pii/ S2405959522000996

[2] https:// networkblog .global .fujitsu .com/ 2023/ 02/ 22/ ai -enabled -computing -for -o -ran
-increase s-utilization-reduces-power-consumption-and-cost/

[3] https:// ieeexplore .ieee .org/ document/ 8663985

76