Search

For a millimeter wave (mmWave) intelligent re-configurable surface (IRS)-MIMO system, if it can correctly detect the interference occurrence and their locations, the patterns of interference signal can be collected and learned using machine learning for the prediction of interference arrival. With the information of interference location and activity pattern, the capacity of the system can be largely improved using many techniques such as beamforming, interference cancellation, and transmission scheduling. This paper aims to detect interference occurrence using a low-complexity MUSIC (MUSIC: multiple signal classification) spectrum-based method, and then localize their sources for mmWave IRS-MIMO system. The MUSIC spectrum of wireless system can be regarded as somehow the ‘signature’ related to the signals transmitted from different users or interference. We utilize such property to detect the occurrence of interference, and then localize their sources in a low-complexity way. Finally, the pattern of interference occurrence can be learned to predict the interference arrival from the collected data. This paper also proposed an efficient probabilistic neural network (PNN)-based predictor for the interference arrival prediction and showed its prediction accuracy. From simulated results, our proposed method can achieve the correct results with the accuracy near to 100% when the fingerprint samples is over 10. In addition, the localization error can be within 1 m with more than 65% and 43% for Y-axis and X-axis, respectively. Finally, based on the results of the interference occurrence, the proposed PNN-based predictor for the interference arrival prediction can capture correctly the similar distribution function of the coming continuous idle status.

The rapid growth in the IoT network comes with a huge security threat. Network scanning is considered necessary to identify vulnerable IoT devices connected to IP networks. However, most existing network scanning tools or system do not consider the burden of scan packet traffic on the network, especially in the IoT network where resources are limited. It is necessary to know the status of the communication environment and the reason why network scanning failed. Therefore, this paper proposes a multimodel-based approach which can be utilized to estimate the cause of failure/delay of network scanning over wireless networks where a scan packet or its response may sometimes be dropped or delayed. Specifically, the factors that cause network scanning failure/delay were identified and categorized. Then, using a machine learning algorithm, we introduced a multimodel linear discriminant analysis (MM-LDA) to estimate the cause of scan failure/delay based on the results of network scanning. In addition, a one-to-many model and a training data filtering technique were adopted to ensure that the estimation error was drastically reduced. The goal of our proposed method was to correctly estimate the causes of scan failure/delay in IP-connected devices. The performance of the proposed method was evaluated using computer simulation assuming a cellular (LTE) network as the targeted IoT wireless network and using LTE-connected devices as the targeted IoT devices. The proposed MM-LDA correctly estimates the cause of failure/delay of the network scan at an average probability of 98% in various scenarios. In comparison to other conventional machine learning classifiers, the proposed MM-LDA outperforms various classification methods in the estimation of the cause of scan failure/delay.

Using the real wireless spectrum occupancy status in 2.4 and 5 GHz bands collected at a railway station as representative of a heavy wireless LAN (WLAN) traffic environment, this paper studies the modeling of durations of busy/idle (B/I) status and its predictability based on predictability theory. We first measure and model the channel status in the heavy traffic environment over almost all of the WLAN channels at 2.4 GHz and 5 GHz bands in a busy (rush hour) period and non-busy period. Then, using two selected channels at 2.4 GHz and 5 GHz bands, we analyze the upper bound (UB) and lower bound (LB) of predictability of the busy/idle durations based on predictability theory. The analysis shows that the LB predictability of durations can be easily increased by changing their probability distribution. Based on this property, we introduce the data categorization (DC) method. By categorizing the busy/idle durations into different streams, the proposed data categorization can improve the prediction performance of some streams with large LB predictability, even if it employs a simple low-complexity auto-regressive (AR) predictor.