Abstract
This paper studies precoder designs for simultaneous wireless information and power transfer (SWIPT) in multi-input multi-output (MIMO) channels, where a transmitter sends information to information decoding (ID) users while satisfying the minimum energy requirement of energy harvesting users. In contrast to the previous designs focused only on maximum information rate (MIR), we propose a more general and simpler solution using the weighted minimum mean squared error (WMMSE) criterion. To solve the SWIPT-WMMSE problem which is generally non-convex, we suggest two different design schemes, separate and joint designs. Interestingly, it is shown that the joint design achieves optimal performance with a single initial point and a few iterations, while the separate design needs a large number of iterations and initial points to approach the optimum. Based on the observation, we propose a simple closed-form solution, which is shown to achieve near optimal performance with reduced complexity. The derived solution can be adopted in various pragmatic applications of MIMO communications, such as the MMSE, quality-of-service, equal error designs, as well as the MIR by adjusting the weight matrix. We also confirm that our design strategies are a great use for managing co-channel interference in multiple ID-user scenarios. Finally, simulation results demonstrate the efficiency of the proposed MIMO-SWIPT framework.
| Original language | English |
|---|---|
| Article number | 7568981 |
| Pages (from-to) | 7941-7954 |
| Number of pages | 14 |
| Journal | IEEE Transactions on Wireless Communications |
| Volume | 15 |
| Issue number | 12 |
| DOIs | |
| Publication status | Published - 2016 Dec 1 |
Fingerprint
Keywords
- Closed-form solution
- MIMO
- Multi-user
- SWIPT
- WMMSE
ASJC Scopus subject areas
- Computer Science Applications
- Applied Mathematics
- Electrical and Electronic Engineering
Cite this
Generalized Precoder Designs Based on Weighted MMSE Criterion for Energy Harvesting Constrained MIMO and Multi-User MIMO Channels. / Song, Changick; Park, Jaehyun; Clercks, Bruno; Lee, Inkyu; Lee, Kyoung Jae.
In: IEEE Transactions on Wireless Communications, Vol. 15, No. 12, 7568981, 01.12.2016, p. 7941-7954.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Generalized Precoder Designs Based on Weighted MMSE Criterion for Energy Harvesting Constrained MIMO and Multi-User MIMO Channels
AU - Song, Changick
AU - Park, Jaehyun
AU - Clercks, Bruno
AU - Lee, Inkyu
AU - Lee, Kyoung Jae
PY - 2016/12/1
Y1 - 2016/12/1
N2 - This paper studies precoder designs for simultaneous wireless information and power transfer (SWIPT) in multi-input multi-output (MIMO) channels, where a transmitter sends information to information decoding (ID) users while satisfying the minimum energy requirement of energy harvesting users. In contrast to the previous designs focused only on maximum information rate (MIR), we propose a more general and simpler solution using the weighted minimum mean squared error (WMMSE) criterion. To solve the SWIPT-WMMSE problem which is generally non-convex, we suggest two different design schemes, separate and joint designs. Interestingly, it is shown that the joint design achieves optimal performance with a single initial point and a few iterations, while the separate design needs a large number of iterations and initial points to approach the optimum. Based on the observation, we propose a simple closed-form solution, which is shown to achieve near optimal performance with reduced complexity. The derived solution can be adopted in various pragmatic applications of MIMO communications, such as the MMSE, quality-of-service, equal error designs, as well as the MIR by adjusting the weight matrix. We also confirm that our design strategies are a great use for managing co-channel interference in multiple ID-user scenarios. Finally, simulation results demonstrate the efficiency of the proposed MIMO-SWIPT framework.
AB - This paper studies precoder designs for simultaneous wireless information and power transfer (SWIPT) in multi-input multi-output (MIMO) channels, where a transmitter sends information to information decoding (ID) users while satisfying the minimum energy requirement of energy harvesting users. In contrast to the previous designs focused only on maximum information rate (MIR), we propose a more general and simpler solution using the weighted minimum mean squared error (WMMSE) criterion. To solve the SWIPT-WMMSE problem which is generally non-convex, we suggest two different design schemes, separate and joint designs. Interestingly, it is shown that the joint design achieves optimal performance with a single initial point and a few iterations, while the separate design needs a large number of iterations and initial points to approach the optimum. Based on the observation, we propose a simple closed-form solution, which is shown to achieve near optimal performance with reduced complexity. The derived solution can be adopted in various pragmatic applications of MIMO communications, such as the MMSE, quality-of-service, equal error designs, as well as the MIR by adjusting the weight matrix. We also confirm that our design strategies are a great use for managing co-channel interference in multiple ID-user scenarios. Finally, simulation results demonstrate the efficiency of the proposed MIMO-SWIPT framework.
KW - Closed-form solution
KW - MIMO
KW - Multi-user
KW - SWIPT
KW - WMMSE
UR - http://www.scopus.com/inward/record.url?scp=85006721264&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85006721264&partnerID=8YFLogxK
U2 - 10.1109/TWC.2016.2609904
DO - 10.1109/TWC.2016.2609904
M3 - Article
AN - SCOPUS:85006721264
VL - 15
SP - 7941
EP - 7954
JO - IEEE Transactions on Wireless Communications
JF - IEEE Transactions on Wireless Communications
SN - 1536-1276
IS - 12
M1 - 7568981
ER -