Gilles Callebaut,Jarne Van Mulders,Bert Cox,Liesbet Van der Perre,Lieven De Strycker,François Rottenberg
Abstract:Wireless power transfer (WPT) has garnered increasing attention due to its potential to eliminate device-side batteries. With the advent of (distributed) multiple-input multiple-output (MIMO), radio frequency (RF) WPT has become feasible over extended distances. This study focuses on optimizing the energy delivery to Energy Receivers (ERs) while minimizing system total transmit power. Rather than continuous power delivery, we optimize the precoding weights within specified time slots to meet the energy requirements of the ERs. Both unsynchronized (non-coherent) and synchronized (coherent) systems are evaluated. Our analysis indicates that augmenting the number of antennas and transitioning from an unsynchronized to asynchronized full phase-coherent system substantially enhances system performance. This optimization ensures precise energy delivery, reducing overshoots and overall energy consumption. Experimental validation was conducted using a testbed with84 antennas, validating the trends observed in our numerical simulations.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is: how to optimize energy transmission in a multi - antenna system to minimize the total transmit power of the system while ensuring that each energy receiver (ERs) can receive the required direct - current (DC) energy. Specifically, the paper focuses on the application scenario of wireless charging electronic shelf labels (ESLs).
### Problem Background
With the development of wireless power transfer (WPT) technology, more and more devices can be wirelessly charged via radio frequency (RF), thus eliminating the need for batteries. However, how to efficiently transfer energy to multiple receivers and minimize the total transmit power of the system while meeting the energy requirements of each receiver is a problem that needs to be urgently solved.
### Core Problems of the Paper
1. **Optimizing Energy Transmission**:
- Research on how to optimize energy transmission through distributed precoding so that each energy receiver (ER) can receive the required energy within a specific time slot.
- Different from the traditional method of maximizing the received energy, the goal of this paper is to minimize the total transmit power of the system.
2. **Comparison between Asynchronous and Synchronous Systems**:
- Evaluate the differences in energy transfer efficiency between asynchronous (incoherent) and synchronous (coherent) systems.
- The results show that increasing the number of antennas and transitioning from an asynchronous system to a synchronous system can significantly improve system performance and reduce energy waste.
3. **Experimental Verification**:
- Experimental verification was carried out using a test platform with 84 antennas, and the results are consistent with the trend of numerical simulations, further proving the effectiveness of the optimization method.
### Mathematical Model
To achieve the above - mentioned goals, the paper proposes the following mathematical models:
#### Incoherent System
For an incoherent system, the average received DC power \( \tilde{p}_{k,n} \) can be expressed as:
\[
\tilde{p}_{k,n} = \alpha \sum_{m = 1}^{M} |h_{m,k}|^2 p_{m,n}-\beta
\]
where:
- \( \alpha \) is the RF - to - DC conversion efficiency,
- \( \beta \) is the minimum received RF power required to maintain a stable voltage,
- \( h_{m,k} \) is the complex channel coefficient from antenna \( m \) to receiver \( k \),
- \( p_{m,n} \) is the transmit power of antenna \( m \) in time slot \( n \).
The optimization problem can be written as:
\[
\min_{p_{m,n}} \sum_{n = 1}^{N} \sum_{m = 1}^{M} T p_{m,n}
\]
Subject to the constraints:
\[
\sum_{n = 1}^{N} \tilde{p}_{k,n}\geq E,\quad \forall k
\]
\[
p_{m,n}\leq P_{\max},\quad \forall m,n
\]
#### Coherent System
For a coherent system, assuming that each antenna uses beam - forming weights \( w_n \), the total transmit power is \( \|w_n\|^2 \), and it is required to satisfy:
\[
\min_{w_0,\dots,w_{N - 1}} \sum_{n = 0}^{N - 1} T \|w_n\|^2
\]
Subject to the constraints:
\[
\sum_{n = 0}^{N - 1} T \alpha |h_k^T w_n|^2\geq E + NT\beta,\quad \forall k
\]
By introducing matrix \( X \), the optimization problem can be rewritten as:
\[
\min_{X\succeq 0} \text{tr}(X)
\]
Subject to the constraints:
\[
\text{tr}(G_k X)\geq E + NT\beta,\quad \forall k
\]
where \( G_k = h_k^* h_k^T \).
### Experiment and Numerical Evaluation
The paper is verified through numerical simulation and experiment.