Abstract: This paper considers the jointly optimal pilot and data power allocation in single-cell uplink massive multiple-input-multiple-output systems. Using the spectral efficiency (SE) as performance metric and setting a total energy budget per coherence interval, the power control is formulated as optimization problems for two different objective functions: the weighted minimum SE among the users and the weighted sum SE. A closed form solution for the optimal length of the pilot sequence is derived. The optimal power control policy for the former problem is found by solving a simple equation with a single variable. Utilizing the special structure arising from imperfect channel estimation, a convex reformulation is found to solve the latter problem to global optimality in polynomial time. The gain of the optimal joint power control is theoretically justified, and is proved to be large in the low-SNR regime. Simulation results also show the advantage of optimizing the power control over both pilot and data power, as compared to the cases of using full power and of only optimizing the data powers as done in previous work.
Introduction
In Massive MIMO systems, a base station equipped with a large number of antennas serves multiple users simultaneously. To perform effective data detection, the base station requires accurate Channel State Information (CSI), typically obtained via uplink pilot signals. This creates a fundamental trade-off: increasing power for pilots improves channel estimation but leaves less energy for the actual data payload.
While previous research often treated pilot power as a fixed parameter, this paper introduces a framework for the joint optimization of both pilot and data power. By considering a total energy budget per coherence interval, the authors demonstrate how to balance these two components to maximize network performance.
Methodology and Complexity Breakthrough
The study models a single-cell uplink using two common linear detectors: Maximum Ratio Combining (MRC) and Zero-Forcing (ZF). A key theoretical contribution is the proof that the optimal pilot sequence length ($\tau_p$) is exactly equal to the number of users ($K$), which significantly simplifies the optimization space.
The authors address two primary network goals:
- Weighted Max-Min SE: Ensuring fairness by maximizing the performance of the user with the worst connection.
- Weighted Sum SE: Maximizing the total throughput of the network.
One of the most significant findings is that the joint optimization problem in Massive MIMO possesses a unique structure that makes it more tractable than traditional Multiuser MIMO problems.
As illustrated in the table above, while the Sum SE problem is typically NP-hard in standard Multiuser MIMO, the authors discovered a convex reformulation for the Massive MIMO case with joint pilot and data power control. This allows the problem to be solved to global optimality in polynomial time.
Key Insights and Results
The Low-SNR Advantage
The benefits of joint optimization are most critical in low Signal-to-Noise Ratio (SNR) conditions, such as at the cell edge. In these scenarios, the system intelligently "borrows" power from data transmission to strengthen the pilots. A superior channel estimate provides a greater overall boost to spectral efficiency than raw data power alone in noisy environments.
Performance Gains
- Fairness: In low-SNR scenarios (-5 dB), joint optimization doubled the spectral efficiency for the 5th percentile of users compared to standard equal power allocation.
- Throughput: For the Sum SE objective, joint optimization provided a 10% improvement over only optimizing data power at low SNR.
- Detector Independence: Interestingly, the power allocation optimal for MRC was found to be also optimal for ZF in the max-min fairness case, suggesting that users do not need to know the specific base station processing technique to benefit from optimal control.
Conclusion and Future Work
This research provides a mathematically guaranteed framework for power allocation that maximizes the potential of Massive MIMO hardware. By transforming complex interference management into solvable convex problems, it offers a practical path for improving network efficiency. Future work will look to extend these joint optimization techniques to multi-cell environments where "pilot contamination" from neighboring cells becomes a factor.
Summary updated