**2.2. Simulated scenario**

4 Will-be-set-by-IN-TECH

constitutes a small-baseline bi-static radar, which somehow resembles the sensing capability of a bat and allows the estimation of both object range and direction (by using time difference of arrival, TDoA). We will also refer to these nodes as "bat-type sensors". Their advantage is that mutual coherent processing is restricted to one platform. So, if several of those nodes are cooperating, they can directly exchange range and DoA (direction of arrival) information, which we call "non-coherent cooperation" since no exact phase synchronization between nodes is required. There is, of course, also a mixed approach which may consist of non-synchronized illuminators and locally coherent (differential) observers. So, with regard to the capabilities of sensor nodes involved and to their mutual cooperation, we distinguish

• The multi-static approach, which assumes full coherent cooperation. The position of those

• The "bat-type" approach consisting of self-contained sensor nodes which are able to detect and recognize characteristic features of the propagation environment on their own. This allows the building up of partial maps of the environment, the investigation of unknown

• The mixed approach is characterized by a cooperation of all types of sensors. For example, the multi-static sensors will support the localization of the scout sensors, and these will deliver additional reference information that relates the local sensor coordinate system to

The basic architecture of a bat-type sensor is described in Fig. 2, see [47], [29], and see Chapter 14. The transmit signal is a periodic M-sequence spread spectrum signal which is generated by a high speed 12-stage digital shift register. It is clocked by a stable RF-clock generator - a dielectric resonance oscillator operating at 7 GHz which allows a usable frequency band up to 3.5 GHz (baseband). The sequence contains 4095 chips corresponding to 585 ns duration and about 175 m of usable distance. In the passband mode the signal is up-/down-converted by the same frequency which results in usable frequency band from 3.5 GHz to 10.5 GHz, which corresponds to the well-known FCC-specification. The receiver

between three basic structures of sensor networks:

the structural details of the environment.

**Figure 2.** Basic architecture of UWB sensors node.

nodes is estimated and tracked w.r.t. the anchor nodes.

objects in more detail, the identification of object features, etc.

As we did not have everything ready for the demonstration scenario, e.g. the robot was not available, a scenario was simulated using the ray tracing tool 3, in order to develop and test the algorithms for map building, localization and imaging independently.

For the final demonstration, the scenario described in Section 2.3 was used.

The simulated scenario is shown in Fig. 3. It consists of a room of the size 9x8x4 m with a pillar in its center. There are six different objects (shown in Fig 3 on the right side) distributed in the room. Their positions are marked in the figure. Metal is chosen as material for the walls and for the six objects.

The bat-type sensor follows the track indicated by the dotted line. The sensor is equipped with one transmit and two receive antennas. The transmit antenna is located at the point (0,0,1) of the local Cartesian coordinate system of the robot (with x being the moving direction, y, and z the height) and the receive antennas at (0,0.5,1) and (0,-0.5,1). At 78 positions, indicated by the red circles on the track, the robot stops and takes measurements (the channel is simulated accordingly). Therefore, the transceiver array rotates in the azimuth plane (x-y-plane) with an angle resolution of 3◦, so that at each position 120 simulations are performed. A horn antenna with a 3 dB beam-width of around 60◦ at 9 GHz was used as the radiation pattern for the transmit and receive antennas. The channel was simulated from 4.5 to 13.4 GHz in steps of 6.25 MHz in order to describe the frequency selective behavior realistically.

The inclusion of high-frequency aspects in terms of the transmission channel is particularly important for the accuracy of the simulation results. Otherwise, unrealistic conclusions will result from the simulation. In order to achieve a realistic evaluation of the efficiency of sensing and imaging systems, the channel model must describe the multi-path propagation realistically. A description of the used channel simulator as well as some verification measurements are given in Section 3.
