Analysis of limited mobile networks and the potential of distributed mobile computing

The relevance of this study stems from the exponential mobile growth and the fundamental limitations of modern mobile networks, which make traditional centralized approaches to information processing inefficient. Aim. The paper analyses the disadvantages of modern mobile networks when transmitting large-scale data and an assessment for possible compensation of limitations through distributed mobile computing, including a study of the impact of thermal throttling on smartphone performance under long-term computing load. Methods. Experimental performance analysis for data transfer rates, synthetic performance tests, stress tests for throttling analysis, and a comparative analysis of mobile and stationary systems are used. Results. A comprehensive real performance analysis disclosed key challenges in 4G, 3G, and 2G networks such as: connection instability (speed fluctuations in 4G from 10-15 Mbit/s with dips up to 5 Mbit/s), high transmission delays, and a catastrophic drop in throughput in previous-generation networks. Experimental data demonstrate that transferring a 2 GB file under optimal 4G conditions takes 20 minutes, while in 3G this figure reaches 2 hours, and in 2G it becomes a virtually impossible task. As an alternative approach, we explore distributed mobile computing for the devices. Comparative performance tests between flagship smartphones (Samsung Galaxy S24) and stationary systems revealed that with single-threaded computing the gap is only 15 %, but the key limiting factor is thermal throttling, which reduces the device's performance by 45-50% after 12-18 minutes of intensive load. Conclusions. The obtained results confirm the theoretical possibility of creating efficient distributed mobile computing systems based on clusters, but point to the necessity for advanced load balancing algorithm and thermal management determining the potential avenues for future study.