Abstract
We devise novel techniques to obtain the downlink power inducing a given load in long-term evolution (LTE) systems, where we define load as the fraction of resource blocks in the time-frequency grid being requested by users from a given base station. These techniques are particularly important because previous studies have proved that the data rate requirement of users can be satisfied with lower transmit energy if we allow the load to increase. Those studies have also shown that obtaining the power assignment inducing a desired load profile can be posed as a fixed point problem involving standard interference mappings, but so far the mappings have not been obtained explicitly. One of our main contributions in this study is to close this gap. We derive an interference mapping having as its fixed point the power assignment inducing a desired load, assuming that such an assignment exists. Having this mapping in closed form, we simplify the proof of the aforementioned known results, and we also devise novel iterative algorithms for power computation that have many numerical advantages over previous methods.
Introduction
In Long-Term Evolution (LTE) networks, the relationship between "load"—defined as the fraction of available frequency-time resource blocks in use—and "transmit power" is fundamental to network performance. However, these two variables are inextricably linked through mutual interference. When a base station increases its power to maintain a user's data rate, it inadvertently creates interference for neighboring cells. This forces those neighboring stations to either increase their own power or utilize more resource blocks, thereby increasing their load.
This research addresses a critical challenge in wireless engineering: how to precisely calculate the power levels required to achieve a specific target load. While previous research identified a "green" opportunity—where increasing network load can actually satisfy user data rate requirements using less total transmit energy—the mathematical tools to implement this efficiently were missing. This paper closes that gap by providing a closed-form interference mapping that transforms complex network optimization into a manageable fixed-point problem.
The Problem: Computational Complexity in Power Planning
Prior to this study, calculating the exact power assignment for a desired load profile was computationally expensive. Engineers typically relied on "nested" iterative methods or bisection techniques. These "loops within loops" were not only slow to converge but also lacked clear indicators of how close an estimate was to the true solution. Without an explicit mapping, the relationship between power and load remained a "black box," making it difficult to guarantee network stability or optimal energy efficiency during real-time adjustments.
Methodology: The Framework of Standard Interference Mappings
The authors leverage the Framework of Standard Interference Mappings, a robust theoretical tool in wireless communications. The primary breakthrough of this work is the derivation of an explicit, closed-form expression for a mapping where the fixed point (the state where input equals output) represents the exact power assignment required for a target load.
Key Methodological Steps:
- System Reformulation: The non-linear equations governing LTE load and power were rewritten into a form compatible with fixed-point analysis.
- Concavity and Standardness Proofs: The authors proved that the new mapping satisfies the properties of a "standard" interference mapping (scalability and monotonicity). This mathematical rigor guarantees that a unique solution exists and can be reached through simple iteration.
- The "Sandwich" Algorithm: The study introduces a novel iterative scheme (Algorithm 1). This algorithm runs two parallel sequences: one starting from zero power and another from a known upper bound. Because the mapping is monotonic, the true solution is always "sandwiched" between these two sequences. This provides engineers with a mathematically guaranteed error bound at every step of the calculation.
Key Insights and Results
Efficiency Through Load
The paper reinforces a counter-intuitive principle of modern networking: "fuller" cells can be more energy-efficient. By spreading data across a larger number of resource blocks at a lower power per block, the total energy footprint of the network is reduced. The study confirms that for any load increase, the required transmit power is strictly lower, validating the energy-saving potential of load-aware planning.
Algorithmic Superiority
By providing the mapping in closed form, the authors eliminate the need for the slow, nested iterative methods used in previous state-of-the-art approaches. The new algorithm offers:
- Guaranteed Convergence: If a valid power assignment exists, the algorithm is mathematically proven to find it.
- Precision Tracking: Unlike previous methods, this approach provides a numerical precision value ((\epsilon)) at each iteration, allowing operators to balance computational effort against required accuracy.
- Simplified Proofs: The explicit mapping allows for more elegant and simple proofs regarding network feasibility and energy optimization than were previously possible.
Limitations and Future Outlook
While the framework is a significant advancement, it operates under the assumption that a valid power assignment for the desired load is physically possible. If requested data rates exceed the hardware's physical limits, the mapping will not find a fixed point. Additionally, the current model assumes static parameters such as pathloss and noise; in highly dynamic environments, the mapping would require frequent recomputation.
Looking forward, this framework provides a foundation for extending these optimizations to 5G New Radio (NR) scenarios. It holds particular promise for implementation in Self-Organizing Networks (SONs), where base stations could use these algorithms to dynamically balance energy consumption and capacity in real-time.
Conclusion
This research provides the mathematical "recipe" previously missing from LTE power optimization. By deriving a closed-form interference mapping, the authors have moved load-aware power planning from a theoretical ideal to a practical, computationally efficient reality, offering a clear path toward more energy-efficient wireless infrastructure.
Summary updated