This paper presents a new pattern recognition approach for enhancing the selectivity of gas sensor arrays for clustering intelligent odor detection. The aim of this approach was to accurately classify an odor using pattern recognition in order to enhance the selectivity of gas sensor arrays. This was achieved using an odor monitoring system with a newly developed neural-genetic classification algorithm (NGCA). The system shows the enhancement in the sensitivity of the detected gas. Experiments showed that the proposed NGCA delivered better performance than the previous genetic algorithm (GA) and artificial neural networks (ANN) methods. We also used PCA for data visualization. Our proposed system can enhance the reproducibility, reliability, and selectivity of odor sensor output, so it is expected to be applicable to diverse environmental problems including air pollution, and monitor the air quality of clean-air required buildings such as a kindergartens and hospitals.

The ephemeris data format of legacy GPS receivers is improper for positioning stationary pseudolites on the ground. Therefore, to utilize pseudolites for navigation, GPS receivers must be modified so that they can handle the modified data formats of the pseudolites. Because of this problem, the practical use of pseudolites has so far been limited. This paper proposes a pseudolite-based positioning system that can be used with unmodified legacy GPS receivers. In the proposed system, pseudolites transmit simulated GPS signals. The signals use standard GPS ephemeris data format and contain ephemeris data of simulated GPS satellites, not those of pseudolites. The use of the standard format enables the GPS receiver to process pseudolite signals without any modification. However, the position output of the GPS receiver is not the correct position in this system, because there are additional signal delays from each pseudolite to the receiver. A post-calculation process was added to obtain the correct receiver position using GPS receiver output. This re-estimation is possible because it is based on known information about the simulated signals, pseudolites, and positioning process of the GPS receiver. Simulations using generated data and live GPS data are conducted for various geometries to verify the proposed system. The test results show that the proposed system provides the desired user position using pseudolite signals without requiring any modifications to the legacy GPS receiver. In this initial study, a pseudolite-only indoor system was assumed. However, it can be expanded to a GPS-pseudolite system outdoors.

Current navigation systems rely heavily on global navigation satellite systems (GNSSs), which provide stable positioning results in open-sky environments but exhibit severe performance degradation in areas where the signal is compromised, such as dense urban environments, underground parking lots, and tunnels. Current navigation systems use an initial entry velocity to estimate a vehicle's position in a tunnel. Therefore, when the speed of a vehicle changes, a corresponding error occurs with respect to the determined position of the vehicle. This study proposes a novel localization technology to estimate a vehicle's position using long-term evolution (LTE) signals measured using a smartphone in a tunnel. Many antennas are installed along the length of the tunnel to increase LTE coverage. Thus, the LTE received signal strength indicator (RSSI) forms a unique pattern with several peaks in the tunnel. The accumulated LTE RSSI pattern was used to estimate the vehicle position in the tunnel. After the LTE fingerprint inside the tunnel was constructed, the position with the highest correlation with the LTE RSSI sequence of the user buffer was determined as the current vehicle location. To demonstrate the feasibility of the proposed system, we conducted an extensive field test in an actual tunnel using various off-the-shelf smartphones. Experimental results show that the proposed technology provides stable performance throughout the tunnel space.

This study proposes a 2D surface correlation-based indoor localization technology using LTE fingerprinting with an accuracy of several meters. The most important problem with RF fingerprinting is that the location discernment of signal strength becomes exceedingly low as the distance from the RF signal source increases. Instantaneous RSS measurement based conventional fingerprinting involves the installation of several signal sources to improve location discernment. However, additional installations of LTE base stations (BSs) are impossible. To improve location discernment, the proposed technology utilizes a spatial RSS pattern extracted using the Pedestrian-Dead Reckoning during user movement. The use of the proposed technology greatly improves the accuracy and availability of LTE signals using the pattern. Additionally, the following two points should be considered. First, the spatially accumulated pattern contains location errors that can cause pattern distortion. The proposed technology performs pattern correction through feature matching using RSS mark and crossroad locations. Second, the accuracy of pattern matching may be decreased prior to sufficient pattern accumulation. For the rapid convergence of the pattern matching, the proposed technology performs correlation pattern analysis. This approach detects the point in which the discernment is increased by pattern accumulation and limits the search range around the matching point. To verify the performance, we conducted tests in a shopping mall where only one LTE BS ID is available. Consequently, the convergence distance of pattern matching was improved by 69% after pattern analysis. Furthermore, it was confirmed that the localization error after convergence improved from 4.16 m to 2.82 m.

Stable estimation of a user’s position in a wide range of indoor spaces is difficult. In an indoor localization field, a fingerprint method that compares the received signal strength (RSS) vector with the previously collected signal strength vector in a radio map is mostly applied. However, this method is dependent on the infrastructure around it and does not always provide stable performance owing to the severity of the RSS noise in an indoor environment, which limits the application of this system. In this paper, we propose an indoor positioning system using an accumulated RSS for mobile users. A user spatial RSS pattern is generated using the estimated user trajectory and the accumulated RSS. The user’s trajectory is estimated using pedestrian dead reckoning (PDR) technology. In particular, we proposed a novel concept, namely a heading flag, to estimate a user’s relative heading change in a certain period. The user’s spatial RSS pattern can be generated using this flag even when the user performs various motions while walking. We developed our own Bluetooth low energy (BLE) beacons and installed few beacons on a testbed for real-time positioning. To demonstrate the performance of the proposed system, we conducted a real field test in an office building. An average position estimation result of less than 2 m was obtained using the proposed system. It is expected that a commercial level location-based service (LBS) will be possible using the proposed system.

Among various sensor measurements that are obtained through smartphones, the received signal strength indicator (RSSI) of a radio frequency (RF) signal is widely used to estimate a user’s location. In this study, we propose a novel RF-based simultaneous localization and mapping (SLAM) technology for robust map construction. Typically, building an indoor map using the SLAM requires accurate loop closure detection. However, it is generally difficult to accurately estimate the loop closure using RSSI owing to the large variance of RF RSSI in indoor environments. To address this, we utilize the user trajectory and RSSI sequence to identify the correct loop closure. The RSSI sequence accumulated in consecutive poses generates a unique signal pattern. Therefore, it is possible to accurately estimate the revisiting of a pedestrian to a specific position. To evaluate the performance of the proposed method, we performed a field test in a laboratory building. Moreover, SLAM was performed using only 25% of the received RF measurements to validate that the proposed method can perform accurate SLAM despite a poor RF infrastructure environment. The obtained experimental results verify that a mean error of less than 3 m can be achieved in various test scenarios.