MACHINE LEARNING TECHNIQUES FOR POWER AND PERFORMANCE OPTIMIZATION IN MOBILE SYSTEM-ON-CHIPS: A SURVEY
DOI:
https://doi.org/10.22159/ijet.2025v13.57904Keywords:
Mobile SoCs, machine learning, Power optimization, Performance enhancement, Workload prediction, Energy efficiency, Dynamic voltage and frequency scalingAbstract
The design of the system-on-chip (SoC) must be able to balance between power and performance since the demands of more individuals to have high-performance phones with extended battery lives are growing. Machine learning (ML) methods have become important to optimizing power and performance in handheld SoCs, which need to put a tradeoff between computational performance and energy efficiency in thermally limited conditions. ML can be used to make smart choices to scale in real-time to dynamic voltage and frequency, optimize schedules on the fly, and predict workloads, which results in a better battery life and responsiveness in real-time in mobile devices. ML has emerged as a groundbreaking solution to the power–performance dilemma. Intelligent decisions related to scheduling of tasks, scaling of hardware, and allocation of resources in real time can be made based on patterns of user behavior, workload, and thermal conditions by learning with ML algorithms. The survey discusses the principles of mobile SoC design, identifies the backgrounds of ML that are applied to optimization, and summarizes recent works on security, thermal management, memory layout, and clock power minimization. A comparative analysis identifies the research gaps that exist at the moment, underlining the fact that the current solutions are fragmented, and holistic and cross-layer structures are required.
References
1. Reddy S.S. Comparative analysis of CPU scheduling algorithms for performance efficiency. 2019.
2. Saba I, Arima E, Liu D, Schulz. Orchestrated co-scheduling, resource partitioning, and power capping on CPU-GPU heterogeneous systems via machine learning. In: Lecture Notes in Computer Science. Vol. 13642. Cham: Springer; 2022. p. 51-67.
3. Mittal S. A survey of techniques for improving energy efficiency in embedded computing systems. Int J Comput Aided Eng Technol 2014;6:440.
4. Liu XW, Liu LM. A mobile computing SoC design. Adv Mater Res 2012;605-607:2049-2052.
5. Kapadia HP. Cross-platform UI/UX adaptions engine for hybrid mobile apps. Int J Nov Res Dev 2020;5:30-37.
6. Popovici K, Rousseau F, Jerraya AA, Wolf M. Embedded Software Design and Programming of Multiprocessor System-on-Chip. New York, NY: Springer; 2010.
7. Duggasani AR. The integration of java-based applications in software development and AI: A conceptual review. Int J Adv Res Sci Commun Technol 2025;5:11.
8. Kumar ND, Mohan G. Design and optimization of system-on-chip (SOC). J Emerg Technol Innov Res 2015;2:234-9.
9. Kalava SP. Enhancing software development with AI-driven code reviews. North Am J Eng Res 2024;5:1-7.
10. Kumar S, Singh R, Kumar S, Gupta S. Light Weight Resnet for Detection of Wheat Yellow Rust Over Mobile Captured Images from Wheat Fields. In: 2023 3rd Asian Conference on Innovation in Technology (ASIANCON). IEEE; 2023.p. 1-4.
11. Patel R. Sustainability and energy management: Trends and technologies for a greener industrial future. Int J Adv Res Sci Commun Technol 2024;4:886-98.
12. Gupta S. Design strategies for mobile SoCs: Balancing battery life and computational demand. J Glob Res Math Arch 2025;12:12-20.
13. Maddali G. Efficient machine learning approach based bug prediction for enhancing reliability of software and estimation. Int J Res Eng Sci Manag 2025;8:1-7.
14. Vangaru KK. Accurate power modeling and estimation for smartphone SoCs using postsilicon correlation. Sarcouncil J Multidiscip 2025;5:514-23.
15. Bhat G, Mandal SK, Manchukonda ST, Vadlamudi SV, Agarwal A, Wang J, et al.Per-core power modeling for heterogenous SoCs. Electronics 2021;10:2428.
16. Pan Z, Mishra P. A survey on hardware vulnerability analysis using machine learning. IEEE Access 2022;10:49508-27.
17. Prajapati N. The role of machine learning in big data analytics: Tools, techniques, and applications. ESP J Eng Technol Adv 2025;5:16-22.
18. Dattangire R, Vaidya R, Biradar D, Joon A. Exploring the Tangible Impact of Artificial Intelligence and Machine Learning: Bridging the Gap between Hype and Reality. In: 2024 1st International Conference on Advanced Computing and Emerging Technologies (ACET). IEEE; 2024. p. 1-6.
19. Majumder RQ. Machine learning for predictive analytics: Trends and future directions. Int J Innov Sci Res Technol 2025;10:3557-64.
20. Patel R. Optimizing communication protocols in industrial iot edge networks: A review of state-of-the-art techniques. Int J Adv Res Sci Commun Technol 2023;4:1-12.
21. Jung H, Pedram M. Supervised learning based power management for multicore processors. IEEE Trans Comput Des Integr Circuits Syst 2010;29:1395-408.
22. Molnos A, Lesecq S, Mottin J, Puschini D. Investigation of Q-learning applied to DVFS management of a System-on-Chip. IFAC Pap Online 2016;49:278-84.
23. Panchal V. Mobile SoC power optimization: Redefining performance with machine learning techniques. Int J Innov Res Sci Eng Technol 2024;13:1-17.
24. Sakthivel P, Narayanasamy P. Test design and optimization for multiple core systems-on-a-chip using genetic algorithm. Int J Comput Sci Netw Secur 2006;6:121-9.
25. Efthymiou A, Bainbridge J, Edwards, DA. Adding Testability to an Asynchronous Interconnect for GALS SoC. In: Proceedings of the Asian Test Symposium; 2004.
26. Ravi S, Lakshminarayana G, Jha NK. Testing of core-based systems-on-a-chip. IEEE Trans Comput Des Integr Circuits Syst 2001;20:426-39.
27. Lee E, Park K, Lee J, Park DJ. Security Architecture for Heterogeneous Chiplet-Based Mobile SoC. In: 2025 IEEE International Conference on Consumer Electronics (ICCE); 2025. p. 1-6.
28. ChoY, Kim H, Lee K, Im Y, Lee H, Kim M. Thermal Aware Floorplan Optimization of SoC in Mobile Phone. In: 2023 22nd IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm); 2023. p. 1-7.
29. ImY, Jung G. Lee M, Gangrade A, Kim S. Thermal Modeling and Optimization of Mobile Device using modified LPV ROM. In: 202322nd IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm); 2023. p. 1-8.
30. Mishra S, Venkateswarlu S, Vermeersch B, Brunion M, Lofrano M, Abdi DB, et al. Towards Chip-Package-System Co-optimization of Thermally-limited System-On-Chips (SOCs). In: 2023 IEEE International Reliability Physics Symposium (IRPS); 2023. p. 1-7.
31. Agrawal S, Ghosh P, Kumar G, Radhika T. Memory Footprint Optimization for Neural Network Inference in Mobile SoCs. In: 2023 IEEE Women in Technology Conference (WINTECHCON); 2023. p. 1-6.
32. Lee JG, Jeon H, Choi Y, Kim A. Fully Automated Hardware-Driven Clock-Gating Architecture with Complete Clock Coverage for 5nm Exynos Mobile SoC. In: 2022 IEEE International Solid-State Circuits Conference (ISSCC); 2022. p. 216-8.
33. Hort M, Kechagia M, Sarro F, Harman M. A survey of performance optimization for mobile applications. IEEE Trans Softw Eng 2022;48:2879-904.
Published
How to Cite
Issue
Section
Copyright (c) 2026 SACHIN MANEKAR
This work is licensed under a Creative Commons Attribution 4.0 International License.