Abstract This paper considers an intelligent reflecting surface (IRS) assisted multi-input multi-output (MIMO) power splitting (PS) based simultaneous wireless information and power transfer (SWIPT) system with multiple PS receivers (PSRs). The objective is to maximize the achievable data rate of the system by jointly optimizing the PS ratios at the PSRs, the active transmit beamforming (ATB) at the access point (AP), and the passive reflective beamforming (PRB) at the IRS. We propose a joint optimization framework (JPSAPBO) that reformulates the non-convex problem using Lagrangian dual and fractional programming transforms. An alternating optimization algorithm is developed, utilizing dual sub-gradient methods and penalty-based majorization-minimization to achieve nearly closed-form solutions. Numerical results demonstrate significant performance gains and the effectiveness of IRS in enhancing SWIPT systems.
1. Introduction
The expansion of the Internet of Things (IoT) necessitates efficient energy solutions for billions of low-power devices. Simultaneous Wireless Information and Power Transfer (SWIPT) is a key technology that enables a single Radio Frequency (RF) signal to carry both information and energy. However, SWIPT systems often suffer from severe path loss, limiting their range and efficiency.
To overcome these hurdles, this research integrates an Intelligent Reflecting Surface (IRS)—a passive array capable of "steering" signals by adjusting their phase shifts. This creates additional propagation paths, or "virtual links," that can be tuned to satisfy energy harvesting requirements without compromising data rates.
2. System Architecture
The proposed system model involves an Access Point (AP) equipped with multiple antennas, an IRS with $N$ reflecting elements, and several Power Splitting Receivers (PSRs).
The IRS controller dynamically adjusts the phase of each reflecting element to optimize the signal environment for the PSRs, which are located within the cell.
3. The Power Splitting Trade-off
A critical component of the system is the PSR. Each receiver must divide the incoming RF signal between its Information Decoder (ID) and its Energy Harvester (EH).
This division introduces a fundamental trade-off: increasing the power for data decoding (to improve the data rate) inherently reduces the power available for energy harvesting. The optimization challenge lies in the fact that the transmitter's beamforming, the IRS's reflections, and the receivers' splitting ratios are all intricately coupled, resulting in a complex non-convex problem.
4. Optimization Methodology
The authors tackle this complexity through a framework named JPSAPBO (Joint Power Splitting, Active and Passive Beamforming Optimization). The approach uses Alternating Optimization (AO) to decompose the problem into three manageable sub-problems:
- Receiver Optimization: Deriving nearly closed-form solutions for the optimal Power Splitting (PS) ratios.
- Access Point Optimization: Using Successive Convex Approximation (SCA) and dual sub-gradient methods to determine the best active transmit beamforming (ATB) vectors.
- IRS Optimization: Employing a penalty-based Majorization-Minimization (MM) approach to handle the "constant modulus" constraint of the IRS elements.
The algorithm is proven to satisfy Karush-Kuhn-Tucker (KKT) conditions, ensuring convergence to a locally optimal solution.
5. Key Results and Insights
Numerical simulations verify the framework's effectiveness:
- Rapid Convergence: The JPSAPBO algorithm typically converges within 30 iterations, making it computationally feasible for practical deployment.
- Performance Gains: The joint optimization significantly outperforms standard benchmarks, including systems without IRS or those using fixed PS ratios.
- Robustness: As energy harvesting requirements ($e_k$) increase, the performance gap between the IRS-assisted system and traditional setups grows, highlighting the IRS's role in supporting power-hungry IoT devices.
6. Conclusion and Future Work
This research provides a robust mathematical foundation for next-generation SWIPT systems. While the results are promising, the authors note that future work should address the challenges of imperfect Channel State Information (CSI) and the hardware constraints of implementing iterative algorithms on low-power IoT devices.
Summary updated