One of the main methods for link quality prediction is to predict the physical layer parameters first, and then evaluate the link quality based on the mapping models between such parameters and packet reception ratio (PRR). However, existing methods often ignore the temporal correlations of physical layer parameter series and rarely consider the influence of link fluctuations, which lead to more errors under moderate and sudden changed links with larger fluctuations. In view of these problems, this paper proposes a more effective link quality prediction method RNN-LQI, which adopts Recurrent Neural Network (RNN) to predict the Link Quality Indicator (LQI) series, and then evaluates the link quality according to the fitting model of LQI and PRR. This method accurately mines the inner relationship among LQI series with the help of short-term memory characteristics of RNN and effectively deals with link fluctuations by taking advantage of the higher resolution of LQI in the transitional region. Compared with similar methods, RNN-LQI proves to be better under different link qualities. Especially under moderate and sudden changed links with larger fluctuations, the prediction error reduces at least by 14.51% and 13.37%, respectively. Therefore, the proposed method is more suitable for low power wireless links with more fluctuations.

Accurate state-of-charge (SOC) estimation is essential for maximizing the lifetime of battery-powered wireless sensor networks (WSNs). Lightweight estimation methods are widely used in WSNs due to their low measurement and computation requirements. However, accuracy of existing lightweight methods is not high, and their adaptability to different batteries and working conditions is relatively poor. This paper proposes a lightweight SOC estimation method, which applies Peukert’s Law to estimate the effective capacity of the battery and then calculates the SOC by subtracting the cumulative current consumption from the estimated capacity. In order to evaluate the proposed method comprehensively, different primary batteries and working conditions (constant current, constant resistance, and emulated duty-cycle loads) are employed. Experimental results show that the proposed method is superior to existing methods for different batteries and working conditions, which mainly benefits from the ability of Peukert’s Law to better model the rate-capacity effect of the batteries.

Multichannel communication has great potential in environments with unknown interference patterns. However, existing link quality metrics and models are generally established and verified under a single-channel scenario, which does not consider the impacts of radio interference and channel change. Therefore, it is hard to directly judge whether these metrics and models are still valid under a multichannel scenario. This paper empirically analyzes the applicability of popular link quality metrics and models in multiple channels with different interference levels. Results show that the link quality estimation (LQE) capability of traditional metrics will be affected by the interference level of the channel, which makes the conclusions obtained under a single-channel scenario no longer valid. Meanwhile, traditional LQE models are basically not adaptive to radio interference and channel change. They are only valid for channels with similar interference under which they are modeled. If these models are directly used under a multichannel scenario, the link quality will be overestimated inevitably. In other words, traditional LQE metrics and models cannot be directly used in the multichannel scenario. It is necessary to deeply analyze the statistical characteristics of popular link quality metrics in multiple typical channels and design channel and interference adaptive metrics and models to support effective multichannel communication.

With the increase of the capacity of photovoltaic (PV) systems, how to alleviate the problem caused by the random output power of PV system becomes significant. This paper uses Markov model to correct prediction results of power generation for grid-connected PV system based on gray neural network model. Using the hourly data of PV system output power under similar climate conditions, gray model of the output power is established in real-time. Then, the output of the gray model, the temperature, irradiance and measured values are used to build the prediction model with neural network, and the residual is corrected by Markov chain. The accuracy of model is researched under three typical weather conditions. The results show that this model gains the high precision, and the efficiency is comparatively better in sunny and cloudy condition than low irradiance. The proposed method reflects the actual trend of the PV generations, which can be successfully applied to engineering and scientific research.

Due to its low cost, primary battery such as alkaline battery and carbon battery is still one of the main options for powering low power devices, such as wireless sensor nodes. In order to prolong the survival time, it is necessary to accurately estimate the State of Charge (SOC) of the battery and then dynamically adjust the working modes of device. Among the existing SOC estimation methods, Electromotive Force (EMF) is generally recognized as an accurate indicator. Taking popular alkaline and carbon batteries as examples, this paper studies the relationship of EMF and SOC through intermittent discharge experiments with both constant current and constant resistance loads. Experimental results show that the change patterns of SOC with EMF are very similar for batteries with the same chemical material. Meanwhile, the discharge mode (constant current or constant resistance) and load have very little effect on the relationship of EMF and SOC. Then, the SOC estimation models based on EMF are established using different fitting methods. Verification results show that, for the target primary batteries, the piecewise linear fitting method is the most accurate, and its estimation error is no more than 1.4%.

The size of time windows used by low power wireless protocols to compute packet reception ratio (PRR) directly affects the accuracy and agility of link quality prediction. Using small time windows would make the prediction more agile, but its accuracy decreases seriously. To improve the accuracy while maintaining agility, many existing methods generally predict the physical layer parameters within small time windows first and then calculate PRR using mapping models of such parameters and PRR. However, due to the propagation characteristics of wireless signal, these mapping models usually introduce large errors. The mapping error will superimpose on the prediction error of physical layer parameters, which would inevitably degrade the accuracy and reliability of PRR prediction. This paper proposes to eliminate the step of mapping from physical layer parameters to PRR and directly use the historical series of PRR computed within small time windows as input parameters to build prediction model. Then, the PRR within large time windows could be predicted. This effectively avoids the errors introduced by using mapping models while not sacrificing the agility. To verify the proposed method, several machine learning algorithms were chosen to implement the prediction model. Compared with similar methods based on mapping models, the proposed one could achieve higher accuracy under different size of time windows. Specifically, under large time windows which are commonly used to describe PRR in practice, its accuracy is much higher. More importantly, agility of the proposed method is basically equivalent or even superior to existing ones.