Abstract 1. Introduction - URSI · A typical LCX-based intruder detection system is composed of a pair of LCX cables and an impulse response measuring equipment. The wideband RF (Radio

A wireless sensing technique based on channel estimation in leaky coaxial cable antenna system

Takeshi Higashino, Minoru Okada

Graduate School of Information Science, Nara Institute of Science and Technology

8916-5 Takayama cho, Ikoma city Nara prefecture 630-0192, Japan {higa,mokada}@is.naist.jp

Abstract This paper describes on wireless sensing techniques on the LCX (leaky coaxial cable) antenna system. Recently, LCX antenna has been developed for the use of GHz such as 2.4 GHz ISM band [1], and its diameter of dielectric is reduced to about 5mm. By virtue of the capability of flexible arrangement and of wide coverage of these slim cable antenna, they can be applied to wireless sensing infrastructure. This paper introduces various sensing applications using LCX antenna such as intruder sensor, linear cell MIMO system and position location.

1. Introduction

Wireless sensing technology has much attension because not only human to human communication, but also machine to machine communications are expected to be essentially required in mobile network. Figure 1 shows the typical configuration of wireless sensing application using LCX antenna. The system is composed of a pair of LCX cables and an impulse response measuring equipment. Infrared sensors, cameras, and microwave radar are widely used sensor devices for sensing devices and systems, however, these sensors are not suitable for wide-area surveillance because limitation in the sensing range. The LCX-based sensing can cover the blind area and can realize reliable surveillance systems on wide area. In particulariy, detection reliability is deteriorated by disturbance due to the cable vibration that is installed at the outside. Moreover, high computational cost is required to detect objest moving direction, position, and mobility in wide range. The [2] proposed that the compressed sensing algorithm is applied to reduce computational cost and to

enhance the realiability for sensing function. On the other hand, LCX can feed multiple-input multiple-output (MIMO) RF signal, because LCX has two RF ports naturally. The channel capacity peformance of MIMO system on LCX has reported recently[1]. The [12] has proposed the positioning method using OFDM channel estimation function and reported its precision.

2. Intruder sensor with compressed sensing A typical LCX-based intruder detection system is composed of a pair of LCX cables and an impulse response measuring equipment. The wideband RF (Radio Frequency) signal is transmitted through either one of the LCX cables. The signal is received by the other one of the cables. The signal is propagated among a pair of LCX cables. The impulse response between LCX is estimated, where the cross-correlation between the transmitted and received signals. When an intruder enters the region between cables, the impulse response must have a certain level of variation. Fig. 2 shows the signal processing for intruer sensing. Since the channel estimation is continuously carried out, the variation of impulse response can be detected. First, 2 dimensional impulse response is obtained, then waveform of impulse response is extracted at the certain location point. Doppler spectrum can be calculated by the FFT operation. This 2 dimensional operation can decriminate intruder moving component and disturbance. However, doppler spectrum

Fig. 1: Application to wide area wirelss sensing using LCX

978-1-4673-5225-3/14/$31.00 ©2014 IEEE

has a lot of other components, because conventional method emproys the Least Square (LS) algorithm, and additive noise has considerable effect for the estimation and these components induce false alarm. To overcome this problem, the Compressed Sensing (CS) is introduced.

Fig. 2. Principle of operation of 2 dimensional intruder sensing

The CS method is capable of estimating the impulse response vector accurately in noisy environment, only it the vector in sparse[2]. However, CS method requires huge computational cost. In order to reduce the computational cost, we propose a modified matching pursuit algorithm. Since the motion of the targets are relatively slow in compared with the measuring interval, the difference in target positions between the current and next estimation period is small. The proposed scheme uses this property. Firstly, it estimates the largest positions using the conventional CS algorithm. At the next step, the search for matching pursuit is carried out only around the vicinity of the target positions estimated at the previous estimation. The search is iteratively performed at the each estimation period. In this algorithm, we can reduce the computational cost thanks to the limited search range. 2.1 Performance

Figure 3 shows an example of detection. In the 2d LS method, it is found that intruder and disturbanse components are not clearly detevted due to the noise. On the other hand, intruder and disturbance components are clealy divided. The computatinal cost of the 2d LS and 2d CS method is O(N^2M+NM logM) and O(N^3M^3), where N and M are the number of position points and maximum observing time, respectively. The proposed method based on matching pursuit successly reduce to 0(Np log N), where Np is the number of nonzero elements.

3. LCX linear cell MIMO system

In Japan, the data traffic for mobile internet access grows up by more than twice per year. Especially, low connectivity at densely populated area such as large station, large shopping mall, and underground city becomes serious problem. To easily achieve high spectrum efficiency, a new configuration of wireless cell using LCX to establish broadband wireless hot spots is proposed[12]. For this purpose, multiple-input multiple-output (MIMO) technique is applied to LCX. The proposed cell configuration using the combination of LCX linear cell and MIMO. Position location is one of required function to perform seamless handover among cells. The terminal position information in linear cell is required.

Fig. 3. Detected doppler frequency (a) 2d LS, (b) 2d CS

Figure 4 shows an LCX linear cell MIMO system. An LCX plays role of antenna and feeder, and it makes radio coverage at area along cable. This type of cell is called the Liner cell. Base station equips modulator and demodulator for the radio air interface, and LCX is fed. In a), LCX typically makes LOS (Line os Sight) path in wireless channel, and spatial streams go through path with difference length. Technical solutions are required to combat the LOS and correlated MIMO channel. In b), new design is required in order to increase the number of spatial streams more than two, because the number of RF ports of typical LCX . In c), position location is required to perform handover seamlessly [4-7]. The d) means the feasibility inspection for more than two spatial streams multiplexing transmission.

3.1 Position Location in Linear cell

MIMO system Figure 5 shows the principle of operation of positioning. OFDM signal with pilot symbol is transmitted from MT. Delay time of each received signal at R1 and R2 are estimated by using channel estimatoion function. The estimated position at x-axis can be calculated from TDOA[8,9] as, 2x/v=t1-t2, where v is group velocity of radio wave in LCX. Channel response is easily estimated by using pilot sequence. For instance in WiFi system, Long Training Sequence (LTF) is used. Figure 5 also shows configuration of receiver. Received two RF signals are independently processed. After the symbol synchronization, FFT is carried out. Frequency domain equalization is performed is employed in many wireless systems using OFDM signal. After the channel estimation, delay time is estimated using phase response, because the slope of phase response is equivalent of delay time of its channel. This method can estimate delay time and their difference

3.2 Performance

The impulse responses are measured in anechoic chamber. Figure 6 shows relationship between MT position and estimated position. Error between ideally estimated position and estimated position using measured response is less than 1[m]. Since this evaluation doesn’t include the additive noise, 1[m] precision can be considered as the best case evaluation.

Fig. 4. Research issues in LCX linear cell MIMO system

Fig. 5. Configuration and princple of operation

Fig. 6. Error performance

4. Conclusion

This paper described various wireless sensing applications using LCX system. Since the 5-D LCX has been developed, varoius sensing applications on LCX antenna infrastrucure are considered and proposed. Not only LCX configuration, but also signal processing technique has an impotarant role for enhancement of accurate detection.

5. Acknowledgments

This paper was partly supported by MIC SCOPE (Strategic Information and Communications R&D Promotion Programme) Grant number 135007001.

6. References

1.F. Suzuki, A. Niwa, K. Takano, “Thin Leaky Coaxial Cable LCX- 5D,” IEICE General Conference 2012, B-1-156, March 2012. 2. Tomonori Sato, Ziji Ma, Takeshi Higashino and Minoru Okada, “Low complexity algorithm for leaky coaxial cable based intruder detection system,” Proc. of ICESIT2013, vol.1, no.1, pp75-79, 14 Jan. 2013. 3. J. Medbo, A. Nilsson, “Leaky Coaxial Cable MIMO Performance in an Indoor Office Environment,” IEEE 23rd International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC 2012), pp. 2061–2066, Sep. 2012. 4. S. Tsukamoto, S. Ano, T. Maeda, S. Sonobe, H. Ban, K. Kobayashi, “A Feasible Study of 2×2 MIMO System with Single Leaky Coaxial Cable,” IEICE General Conference 2013, B-1-220, March 2013. 5. J. H. Wang, K. K. Mei, “Theory and analysis of leaky coaxial cables with periodic slots,” IEEE Transactions on Antennas and Propaga- tion, vol.49, no.12, pp.1723–1732, Dec 2001. 6. Samer S. Saab, Zahi S. Nakad,“ A Standalone RFID Indoor Positioning System Using Passive Tags, ” IEEE TRANSACTIONS ON INDUS- TRIAL ELECTRONICS, vol.58, no.5, pp.1961-1970, May. 2011. 7. Ali Asghar Nazari Shirehjini, Abdulsalam Yassine, Shervin Shirmoham- madi,“ An RFID-Based Position and Orientation Measurement System for Mobile Objects in Intelligent Environments, ” IEEE TRANSAC- TION ON INSTRUMENTATION AND MEASUREMENT, vol.61, no.6, pp.1664-1675, Jun. 2012. 8. Angus F. C. Errington, Brian L. F. Daku, Arnfinn F. Prugger,“Initial Position Estimation Using RFID Tags: A Least-Squares Approach, ”IEEE TRANSACTION ON INSTRUMENTATION AND MEASUREMENT, vol.59, no.11, pp.2863-2869, Nov. 2010. 9. Hao Ni, Guangliang Ren, Yilin Chang,“ A TDOA Location Scheme in OFDM Based WMANs, ”IEEE TRANSACTIONS ON CONSUMER ELECTRONICS, vol.54, no.3, pp.1017-1021, Aug. 2008. 10. K. Nishikawa, T. Higashino, K. Tsukamoto, S. Komaki,“Two dimentional position detection method using bi-directional leaky coaxial cable based on TDOA, ”Proc. of 2009 IEEE 20th International Symposium on Personal, Indoor and Mobile Radio Communications, pp.2167-2170. 11. H. Shiranaga, S. Ogawa, “ Positioning Algorithm, ” IEICE Technical Report, ITS2008-2, Japan, vol.108, no.42, pp.9-14, May 2008. 12. S. Tsukamoto, T. Maeda, S. Ano, Y. Hou, H. Ban, M. Ariyoshi and K. Kobayashi, “Performance Evaluation of 2x2 MIMO System with Single Leaky Coaxial Cable,” Proc. of Vietnam-Japan International Symposium on Antennas and Propagation, 2014.