Project topics and materials
Get Complete Project Material File(s) Now! »
Indoor positioning: a critical need
Context
Personal Networks have been designed to meet the demands of users to interconnect their various personal devices at different places into a single network as shown in Figure 1-1.
This architecture will emerge new concepts and features for wireless data transmission and transponder systems. Some of the numerous possible application areas are: self-organizing sensor networks, ubiquitous computing, location sensitive billing, context dependent information services, tracking, and guiding.
So a well suited Personal Networks should include an accurate reliable and real-time indoor positioning protocols and services [2], [3], [4], especially for the future generation of communications networks in smart cities, where there is rapid development of integrated networks and services in PNs [5]. In addition, location information is one of the most important needs for several objectives since it helps to get better network planning [6], network adaptation [7], and load balancing [8] , etc.
Nowadays, IP Systems enable valuable position-based applications and services for users in Personal Networks such as homes, offices, sports centers, etc. For example, inside complex hospitals environments they provide guidance to the patients for efficient use of the limited medical resources system with achieving communication distance up to tens of meters away. Another example such as specifying a location of products stored in a warehouse may impact directly on storage costs. In addition, detecting firemen location in a building on fire, with maximum 3m accuracy and 95% accuracy, following up police dogs trained to find explosives in a building, and finding out tagged maintenance tools and equipment scattered all over a plant, remain relevant applications for IP. Another emerging field, requiring more precise positioning, deals with the Body Area Network [9]
General definitions
For positioning purpose, we need to distinguish between self-positioning, where the connected object or mobile unit (MU) itself determines its spatial coordinates relatively to fixed access or reference unit (RU), like for GPS, and remote positioning where the position of the MU is determined by a central point, like radar.
In some networks or for IoT purpose, one can also meet hybrid approaches leading to indirect remote positioning (IRP) or indirect self-positioning (ISP) system. For IRP system, each MU first performs a self-positioning and then transmits its position to a central point. For ISP the central point disseminates in the network the position of all objects. IRP and ISP are of great importance in updating the neighborhood tables highly required for routing algorithms.
Depending on different applications, IP Systems assign different types of location information mainly classified as [10]:
– Physical location which expresses a location in the form of coordinates;
– Symbolic location which expresses a location in a natural-language way such as : in the office, in the third-floor, in the bedroom, etc.;
– Absolute location which uses a shared reference grid for all located objects;
– Relative location which depends on its own frame of reference.
Categorizing indoor positioning Systems can be based on the technology options as well as on the positioning algorithms used for. From positioning algorithms point of view there are mainly three types [11]:
– Conventional triangulation,
– Scene analysis,
– Proximity positioning algorithms.
Based on these fundamental technologies and algorithms, research centers and universities try to find out new IP Systems by taking into account the advantages of one of the three positioning technologies or by combining, in a relevant way, some of them. The capability of positioning a device can be done for instance for wireless technology, through four steps, as shown in Figure 1-2.
– Detect the signals coming from fixed access points within particular vicinity,
– Calculate the propagation times,
– Estimate the position with respect to these access points,
– Use positioning information for every context aware application (shopping mall, hospital, marketing…etc.
Indoor Positioning Systems
In this section, we introduce a short review about a variety of IP Systems. These IP Systems will be explained according to the criteria and requirements specified in the previous section which focuses on the user needs in Personal Networks. Thus we can know the advantages and limitations of these IP Systems from the user point of view.
Infrared (IR) Positioning Systems
Infrared (IR) positioning systems [12], [13], [14] use IR technology to perform localization as shown in Figure 1-3. There are three main systems IR uses: Active Badge, Firefly, and OPTOTRAK PRO series.
The simplicity of the systems architecture, the accurate positioning and the ability to be carried by a person, are the common advantages of these systems. On the other hand, the main disadvantages are that IR positioning systems are limited within a room and need high directional line-of-sight communication between transmitters and receivers without interference from strong light sources. Some limitations for sensing the location in terms of security, privacy, cost, and finally the IR wave cannot penetrate opaque materials, and many tags has to be installed on the localized object, which adds more complexity for implementation.
Ultra-sound Positioning Systems
Another way to perform object positioning is to use ultrasound signal. With this kind of inexpensive positioning solutions, ultrasonic technology and triangulation technique are used to estimate the location of a target installed on a person. Usually, a combination of the ultrasound signals and Radiofrequency signals are used to perform synchronization and coordination in the system [15], [16]. This increases the system coverage area.
There are three Ultrasound positioning systems: Active Bat [17], Cricke [18] and, Sonitor [19]. All of them suffer from reflected ultrasound signals, noise, and have lower measurement accuracy (several centimeters) than IR-based systems (several millimeters).
Radio Frequency (RF) Positioning Systems
Today the main solutions for localization use RF wireless techniques that are reported in a huge amount of publications and which can be summarized in a comprehensive survey assessed in 2001, by Hightower & Borriello [20]. As mentioned previously, the taxonomy of localization mechanisms shows that there are mainly four ways to perform localization. The first one includes active localization where the beacon sends signals to localize target and acts as RP or radar. The second one deal with cooperative localization, i.e. the target cooperates with the system, and acts as RP or SP. The third way concerns passive localization where the system deduces location from observation of opportunistic signals, acting as SP, and finally the last way is blind localization where the system deduces location of object without a priori knowledge of its characteristics, and hence acts as RP.
Radiofrequency (RF) technologies are used in IP Systems to provide larger coverage area. In addition, they need less hardware comparing to other systems because of their capability to travel through walls and human bodies.
The main techniques used by RF-based positioning systems are triangulation and fingerprinting techniques. Fingerprinting gives a good estimation performance in complicated indoor environments, where it depends on location related characteristics to calculate the location of a user or a device. Here are some introductions to different types of RF positioning system.
Radio Frequency Identification (RFID)
The radio frequency identification (RFID) is commonly used in complex indoor environments such as office, hospital, etc. it can be used to stores and retrieves data through electromagnetic transmission, also it enables flexible and cheap identification of an individual person or device. There are two kinds of RFID technologies, passive RFID and active RFID [21]. We could distinguish between the two kinds by specifying where the target is, if it is at the receiver side then the technology is passive RFID, otherwise it is active RFID. The targets with passive RFID are small and inexpensive but their coverage range is short. Conversely, in Active RFID, cost of targets is higher and their coverage area is larger. As an example, RadarGolf sells a golf ball that helps golf player to locate his golf ball quickly over a range of some 10 to 30 meters. The system uses received signal strength or imaging techniques to locate the ball.
Wireless local area network (WLAN)
Many of the public areas such as train stations, universities, and many else have used WLAN technology to implement their networks. So the existing WLAN infrastructures in indoor environments have been reused by WLAN-based positioning systems, which reduce drastically the cost of indoor positioning. The WLAN-based positioning systems can also reuse PADs, laptops, and mobile phones as tracked targets to locate persons, hence this WLAN technology is already integrated into these wireless devices.
Companies such as AeroScout, Ekahau, PanGo, and WherNet provide Wi-Fi tags able to track locations of notebook PC and persons. However, there are many problems, dealing with the channel effects, that affect the accuracy of location estimation.
Bluetooth, the IEEE 802.15.1 standard
Bluetooth replaces the IR ports mounted on mobile devices because it enables a longer range of few tens of meters (Bluetooth 2.0 Standards). Bluetooth chipsets are low cost, it results in low price tracked targets which are used in the positioning systems. In general, the infrastructures in Bluetooth-based positioning systems [22] consist of various Bluetooth clusters. In each cluster, the other mobile terminals locate, using fingerprinting technique, the position of a Bluetooth mobile device. However, accuracy from 2m to 3m and delay of about 20s is only what can Bluetooth-based positioning system provides, as it suffers from the drawbacks of RF positioning technique in the complex and changing indoor situations.
Sensor Networks
Here, the sensors simply detect any specific changes in a physical or an environmental condition, for example, sound, pressure, temperature, light, etc., and produce relative outputs.
Sensor based IP Systems use a number of known sensors in a fixed position and locates the position of an object from the measurements taken from these sensors. Due to the decreasing price and size of sensors, IP Systems sensor based [23], [24] provides a cost-effective and convenient way of locating persons. Compared with the mobile phone, cheap and small sensors have limited processing capability and low battery power. The drawbacks could be summarized as less accurate, low autonomy, low computational ability…. So, more efforts are needed to offer precise and flexible indoor positioning services.
Table of contents :
General Introduction
Chapter 1: Context and State of the art of Indoor Positioning
1.1 Introduction
1.2 Indoor positioning: a critical need
1.2.1 Context
1.2.2 General definitions
1.3 Indoor Positioning Systems
1.3.1 Infrared (IR) Positioning Systems
1.3.2 Ultra-sound Positioning Systems
1.3.3 Radio Frequency (RF) Positioning Systems
1.3.4 Alternative systems
1.4 Measuring Principles
1.4.1 RF Metrics for Wireless Localization
1.4.2 Scene Analysis
1.4.3 Proximity
1.4.4 Conclusion
1.5 Objectives
1.6 Conclusion
Bibliography
Chapter 2: Multi carrier communication signals
2.1 Introduction
2.2 General principle of MC based TDOA estimation
2.3 Multi carrier based positioning systems and OFDM solutions
2.3.1 Blind solution
2.3.2 Training solution
2.3.3 Alternative solution
2.3.4 OFDM based positioning
2.3.5 Conclusion
2.4 Data modulation
2.4.1 QPSK Based Communication System
2.4.2 SNR calculation
2.4.3 Simulation results
2.5 OFDM communication system
2.5.1 Transmitter/Receiver module
2.5.2 Guard Interval
2.5.3 Guard Band and roll of factor
2.6 MATLAB implementation
2.6.1 Transmission part
2.6.2 Reception part
2.7 Channel Estimation
2.7.1 Pilot block
2.7.2 Mathematical derivation
2.7.3 Channel estimation testing
2.8 Conclusion
Bibliography
Chapter 3: OFDM based TDOA estimation
3.1 Introduction
3.2 Algorithms for TDOA-based positioning
3.3 Definition of the direct model
3.3.1 Frequency limitation
3.3.2 Signal model
3.3.3 Energy based approach
3.3.4 Channel based approach
3.4 Inverse problem: TDOA extraction
3.4.1 Large TDOA
3.4.2 Small TDOA
3.4.3 Very small TDOA
3.4.4 Cramer Rao Bound Limit
3.5 Communication parameters effect
3.5.1 Estimation of the coefficients ,
3.5.2 Number of pilots
3.6 Communication environment effect
3.6.1 Multipath modeling
3.6.2 Emulating Multipath
3.6.3 Multipath effect reduction
3.7 Conclusion
Bibliography
Chapter 4: Experimental setup and results
4.1 Presentation of the environment
4.1.1 The controlled electromagnetic room
4.1.2 Radiating devices
4.1.1 Amplifier
4.1.2 Arbitrary waveform generator
4.1.3 Digital storage oscilloscope
4.1.4 Conclusion
4.2 SISO communication system setup
4.2.1 The transmitter
4.2.2 The receiver
4.2.3 Signal acquisition and I-Q constellation
4.2.4 OFDM communication performances
4.2.5 Channel estimation
4.3 Direct and Inverse models validation
4.3.1 MISO configuration
4.3.2 Baseline calculation
4.3.1 Calibration MISO system
4.3.2 MISO configuration for TDOA estimation
4.3.3 Direct model validation
4.3.4 Inverse model validation
4.3.5 Multipath effects
4.4 Conclusion
Conclusions and perspectives
