The opportunistic relaying scheme design and symbol error rate analysis for PLC networks in smart homes

Power line communication (PLC) is treated as a retrofit technology, transmitting messages via conductors without extra infrastructure, which reduces the cost of implementation and maintenance [1]. Therefore, PLC is extensively adopted in smart grid applications. However, PLC encounters a hostile communication environment caused by the fast signal attenuation and severe impulsive noise [2]. These factors limit the reliability of PLC networks in long-distance transmission. In general, reliability is a key factor for PLC networks, measured with the symbol error rate (SER). Thus, the research on SER analysis for PLC networks has attracted great attention. The analysis on SER for PLC networks mainly focuses on two aspects, the multi-path effect and the impulsive noise [3]. For instance, Ref. [4] has developed a signal attenuation channel model over multiple paths based on the distance to analyze the performance for the PLC channel. However, the channel model is developed with specific data, which cannot be generalized easily. To solve the problem, the authors of [5] have modelled the PLC channel with a statical approach to derive the received signal-to-noise rate (SNR) and SER. Recently, a number of researches have focused on the impulsive noise, which is another important factor leading to symbol errors in PLC networks [6, 7]. Among these research works, the Bernoulli-Gaussian model is a practical one, where the PLC system is shifted between two states: the impulsive noise state and Gaussian noise state [7]. To reduce the SER, relaying technologies have been introduced into PLC networks. For example, Ref. [8] investigated a relay-based communication protocol to solve the inefficiency problem in long-distance PLC communications. Recently, opportunistic relaying technologies have become an attractive solution for reliable wireless communications. However, to the best of the authors’ knowledge, the design of the opportunistic relaying schemes for the PLC network is rarely reported. Moreover, there is little literature analyzing SER for an opportunistic relaying PLC network. In this study, an efficient opportunistic decode-andforward relaying model is designed for the PLC network in smart homes. Then, we model the PLC channel with the lognormal fading under the Bernoulli-Gaussian noise. Based on the channel model, we derive an approximate expression for the cumulative distribution function (CDF) of the received SNR. By using the Tailor extension and Gaussian approximation, we develop the approximate closed-form expression for average SER of the opportunistic relaying PLC network with the derived CDF of received SNR. System model. As shown in Figure 1(a), we design a typical PLC network for smart homes, which is composed of several sub-circuits. These sub-circuits independently connect with a switcher, denoted by K, to obtain a power supply from the power grid. To bring safe power utility and convenient maintenance for smart home users, we adopt a single-phase transformer to isolate the sub-circuits, which is suitable for smart home users owning to its cheapness and small size. Owing to the isolation with the transformer, transmitting signals over power lines in different sub-circuits do not interfere with each other. Furthermore, to cope with the unstable PLC communications, relaying nodes are installed in the sub-circuits to improve transmission reliability. As shown in Figure 1(b), we construct a corresponding logical structure for the in-home PLC network shown in Figure 1(a). In the PLC network, a source node S in the lighting sub-circuit is selected to communicate with the destination node D located in the heating sub-circuit via several relaying nodes R1, R2, . . . , RM . In our opportunistic relaying communication scheme, we dynamically select relaying nodes with the highest forwarding efficiency in different time slots as the optimal relaying node to forward messages. Then, we implement a relaying protocol by adopting a two-stage transmission scheme. In the first time slot, S transmits messages to relaying nodes Ri for i = 1, 2, . . . ,M . Thus, the received signal at Ri is