Abstract
Wireless sensor networks (WSNs) play a key role in automation and consumer electronics applications. This paper deals with joint design of the source precoder, relaying matrices, and destination equalizer in a multiple-relay amplify-and-forward (AF) cooperative multiple-input multiple-output (MIMO) WSN, when partial channel-state information (CSI) is available in the network. In particular, the considered approach assumes knowledge of instantaneous CSI of the first-hop channels and statistical CSI of the second-hop channels. In such a scenario, compared to the case when instantaneous CSI of both the first- and second-hop channels is exploited, existing network designs exhibit a significant performance degradation. Relying on a relaxed minimum-mean-square-error (MMSE) criterion, we show that strategies based on potential activation of all antennas belonging to all relays lead to mathematically intractable optimization problems. Therefore, we develop a new joint relay-and-antenna selection procedure, which determines the best subset of the available antennas possibly belonging to different relays. Monte Carlo simulations show that, compared to conventional relay selection strategies, the proposed design offers a significant performance gain, outperforming also other recently proposed relay/antenna selection schemes.
Highlights
AND SYSTEM MODELWith the advent of massive Internet of Things and massive machine-type communications, especially in the domain of 5G consumer electronics, there is a need to further enhance physical-layer performance of wireless sensor networks (WSNs)
Besides the single-relay selection method described in Subsection III-B, referred to as ‘‘1-R Selection’’, and the joint antenna-and-relay selection scheme developed in Subsection III-C, referred to as ‘‘JAR Selection’’, we report the performance of [26, channel-state information (CSI) Assumptions I and II] in the case of single-antenna nodes (i.e., N = 1) and that of [30] for both single- and multiple-antennas nodes (i.e., N ∈ {2, 3})
As a reference lower bound, we include in all the plots the ASEP curves of the Full CSI (F-CSI) design proposed in [15], whose design relies on the additional knowledge of the ith second-hop channel matrix Gi at the ith relay, for i ∈ {1, 2, . . . , NC}
Summary
With the advent of massive Internet of Things and massive machine-type communications, especially in the domain of 5G consumer electronics, there is a need to further enhance physical-layer performance of wireless sensor networks (WSNs). While the dual-hop channel matrix C can be directly estimated at the destination by training, separate acquisition of the first- and second-hop matrices H and G is more complicated to achieve, both in terms of communication resources and signal overhead, especially in multiple-relay WSNs. since channel estimation errors occur in practical situations, robust optimization designs are needed [18], [19], which further complicate system deployment. The optimization problem in [28] does not admit a closed-form solution and is solved by using a line search algorithm It has been shown in [29] that P-CSI relay selection approaches for MIMO nodes, based only on the instantaneous knowledge of H, do not fully exploit the diversity arising from the presence of multiple relays.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
