Matching State Estimation Scheme for Content-Based Sensor Search in the Web of Things

Abstract

More recently, an increasing number of object-attached sensors are publishing their real-time state on the Internet by using stateof-the-art Web technologies, which make the sensor search service extremely important for the Web of Things (WoT). However, the existing issues that the sensor search service is facing bring huge challenges to the design of matching state estimation scheme. In this paper, an architecture of high-efficiency content-based sensor search system is depicted to provide a prototype system for sensor search. And then a matching state estimation scheme is proposed in detail, including a sensor state prediction approach to accurately estimate future sensor readings and a match estimating and verifying approach to effectively classify and verify candidate sensors, in order to enhance the performance of our search system. Simulation results show that our matching state estimation scheme dramatically reduces the communication overhead of search system and achieves excellent performance in terms of recall ratio and precision ratio.

SDU: State-Based Dual-Mode Sensor Search Mechanism Toward Internet of Things

“…The prediction performances of proposed HCPM are verified here by being compared with the typical MPM [23] and MSE [21] methods which are based on the shallow learning theory. Simulation results are shown below.…”

“…In this subsection, we validate the proposed CSSM method by comparing with MPM-based search method and MSE-based search method. As mentioned above and in literature [21]- [23], all the three search methods contain verification process, thus the search systems always return verified sensors that meet the search requirements. Then we can deduce that given the number of returned sensors R q , the efficiency of search mechanism is directly related to the number of sensors that need to be accessed for verification.…”

“…For realizing this vision, CSS was designed in literature [20], based on the Fuzzy Set theory where a lightweight sensor state prediction model was constructed by CSS to estimate the matching probability between the sensor current state and the search demand, and then the search results were returned in descending order of matching probabilities. In our previous work [21], based on the Grey System model, we proposed the sensor quantitative state prediction method, MSE, towards quantitative state based sensor search in IoT. Then in literature [22], [23] we further presented high-accuracy sensor quantitative state prediction method, MPM, which explored the change trend of sensor quantitative state by mining the time correlations between historical quantitative state data, thereby accurately predicting the sensor current quantitative state.…”

An Efficient Preference-Based Sensor Selection Method in Internet of Things

“…Truong et al [17] proposed a lightweight prediction model based on fuzzy logic algorithm to estimate the probability of sensor option and realize content-based sensor search in the Internet of things with low communication overhead and computational efficiency. Puning et al [18] proposed a sensor state prediction method to estimate the short-term state of the sensor, which could make better use of the time correlation between sensor data and accurately perceive the change trend of sensor readings in the future. Vasilev et al [19] proposed a scalable model based on hypergraph representation to evaluate the cooperative relationship between sensor nodes and achieve reliable node, which well covered the dynamic characteristics of complex sensor networks.…”

“…Zhang et al [32] established a multidimensional feature selection model and a dynamic sensor data prediction model by using the complex time series constituted by various sensors, which improved the accuracy and stability of long-term prediction results of iot sensor data. Zhang et al [18] described an efficient MSE structure for content-based sensor search, where the shortterm state of sensors are estimated in order to find candidate ones.…”