BACKGROUND FIELD OF THE DISCLOSURE The present disclosure relates to communication devices and communication methods for RF-based communication and acquisition of information for use in position determination. DESCRIPTION OF RELATED ART The use of RF signals for indoor positioning is a possible future key technology. However, RF-based positioning often suffers from non-line-of-sight (NLOS) propagation, i.e. the receiver receives multiple time-delayed and attenuated copies of the transmit signal due to signal reflections and refractions but the direct propagation path is missing or at least strongly attenuated. Hence, there is a need for improved ways enabling indoor positioning using RF signals conventionally used for communication purposes. The "background" description provided herein is for the purpose of generally presenting the context of the disclosure. Work of the presently named inventor(s), to the extent it is described in this background section, as well as aspects of the description which may not otherwise qualify as prior art at the time of filing, are neither expressly or impliedly admitted as prior art against the present disclosure. SUMMARY It is an object to provide communication devices and methods that enable and improve indoor positioning using RF signals. It is a further object to provide a corresponding computer program and a non-transitory computer-readable recording medium for implementing for implementing said methods. According to an aspect there is provided a communication device for RF-based communication and position determination, said communication device comprising: an antenna unit configured to transmit and receive RF signals, a beamforming unit configured to perform beamforming and to control the antenna unit to transmit and/or receive RF signals using one or more selected beams, a control unit configured to control the beamforming unit in a training phase to perform beamforming for determining an initiator line of sight, LOS, beam to a re- sponder communication device, and a processing unit configured to determine the initiator LOS beam and/or initiator angular information of the initiator LOS beam and to determine the position of said communication device using the determined initiator LOS beam and/or initiator angular information in a measurement phase. [0006] According to a further aspect there is provided a communication method for use by an initiator communication device for RF-based communication with a responder communication device and for position determination of the initiator communication device, said communication method comprising: controlling the initiator communication device to perform beamforming in a training phase for determining an initiator line of sight, LOS, beam from the initiator communication device to the responder communication device, determining the initiator LOS beam, and/or determining initiator angular information of the initiator LOS beam determining the position of the initiator communication device using the determined initiator LOS beam and/or initiator angular information in a measurement phase. [0007] According to an aspect there is provided a communication device for RF-based communication and position determination, said communication device comprising: an antenna unit configured to transmit and receive RF signals, a beamforming unit configured to perform beamforming and to control the antenna unit to transmit and/or receive RF signals using one or more selected beams, a control unit configured to control the beamforming unit in a training phase to perform beamforming for determining a responder line of sight, LOS, beam to an initiator communication device, and a processing unit configured to determine the responder LOS beam and/or responder angular information of the responder LOS beam and to determine the position of the initiator communication device using the determined responder LOS beam and/or responder angular information in a measurement phase . [0008] According to a further aspect there is provided a communication method for use by a responder communication device for RF-based communication with an initiator communication device and for position determination of the initiator communication device, said communication method comprising: controlling the responder communication device to perform beamforming in a training phase for determining a responder line of sight, LOS, beam to the initiator communication device, determining the responder LOS beam and/or responder angular information of the responder LOS beam, and determine the position of the initiator communication device using the determined responder LOS beam and/or responder angular information in a measurement phase for determining the position of the initiator communication device. [0009] According to still further aspects a computer program comprising program means for causing a computer to carry out the steps of the methods disclosed herein, when said computer program is carried out on a computer, as well as a non-transitory computer- readable recording medium that stores therein a computer program product, which, when executed by a processor, causes the methods disclosed herein to be performed are provided. [0010] Embodiments are defined in the dependent claims. It shall be understood that the disclosed communication methods, the disclosed computer program and the disclosed computer-readable recording medium have similar and/or identical further embodiments as the claimed communication devices and as defined in the dependent claims and/or disclosed herein. [0011] One of the aspects of the disclosure is to apply beamforming for RF-based positioning. Beamforming provides means to resolve major issues in RF indoor localization such as NLOS signal propagation and LOS path retrieval, i.e. it is easier or even possible at all to differentiate between LOS and NLOS components in an impulse response, also because the LOS path in impulse response can be strongly attenuated and because angular information of beamforming sectors can be wrong as a consequence. Beamforming particularly provides means to increase detection probability of LOS path in impulse response. Further, while conventional beamforming is used to maximize channel capacity, it is used according to the present disclosure to increase detection probability of the LOS path, e.g. by minimizing time of arrival and by applying beamforming to suppress NLOS components coming from other directions than the beam main lobe. Thus, the present disclosure enables a more reliable use of sector angular information for positioning since LOS path detection probability is increased. For the subsequent exchange of RF signals between the communication devices in a measurement phase, in which the acquired information can be used for position determination of the initiator communication device, either absolutely or with respect to the responder communication device, a communication channel is used. The communication channel consists preferably of a LOS path, but may still have NLOS components. However, due to the use of beamforming as disclosed, NLOS components are attenuated compared to the LOS component so that the communication channel substantially corresponds to the LOS path. The disclosed way of acquiring information for use in position determination may be applied with Wireless LAN (WLAN) in the 60GHz frequency band (as e.g. described in IEEE 802.1 1 ad). Further, changes to the existing positioning protocol FTM (Fine Time Measurement) and to 60GHz WLAN standards (IEEE 802.1 1 ad/ay) may be made as disclosed herein, which enable enhanced positioning employing beamforming. Particularly in communication systems that feature beamforming inherently (such as mm-wave systems operating at 60GHz for example) the disclosed devices and methods may favorably be applied. However, the disclosure is not limited to mm-wave systems. The foregoing paragraphs have been provided by way of general introduction, and are not intended to limit the scope of the following claims. The described embodiments, together with further advantages, will be best understood by reference to the following detailed description taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWING A more complete appreciation of the disclosure and many of the attendant advantages thereof will be readily obtained as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings, wherein: shows a diagram generally illustrating positioning using ToF and angul information; shows diagrams generally illustrating a LOS and a NLOS scenario where beamforming is performed which maximizes channel capacity; shows a diagram illustrating the difference between beamforming maximizing channel capacity and beamforming for LOS detection according to the present disclosure; shows a schematic diagram of an embodiment of an analog beamforming arrangement as it is typically applied in mm-wave systems; shows a schematic diagram of an embodiment of a digital beamforming arrangement as typically applied in Ml MO communication systems operating well below 60GHz; shows diagrams illustrating angular definitions for LOS detection and device orientation estimation; shows a schematic diagram of a first embodiment of an initiator communication device and a responder communication device; shows a schematic diagram of a second embodiment of an initiator communication device and a responder communication device; shows a schematic diagram of a third embodiment of an initiator communication device and a responder communication device; shows a diagram illustrating a conventional fine time measurement procedure; shows a diagram illustrating a first embodiment of a fine time measurement procedure according to the present disclosure; shows a diagram illustrating a second embodiment of a fine time measurement procedure according to the present disclosure; shows a diagram illustrating a method to define device orientation in WLAN systems; shows a diagram illustrating the training phase of a method according to the present disclosure; shows diagrams illustrating a conventional SSW feedback field format; shows a diagram illustrating a known SSW feedback field format and an embodiment of a SSW feedback field format according to the present disclosure; shows a diagram illustrating a BRP Request field format; shows diagrams illustrating a DMG STA Capability Information field format; shows diagrams illustrating a Beacon Interval Control field format; shows a flowchart of a method according to the present disclosure; shows a simplified diagram illustrating the elementary FTM message exchange according to the first or second embodiment of the fine time measurement procedure shown in Fig. 1 1 or Fig. 12, respectively; Fig. 22 shows a simplified diagram illustrating the elementary FTM message exchange according to a third embodiment of a fine time measurement procedure according to the present disclosure; Fig. 23 shows a diagram illustrating beam tracking of receive beams; Fig. 24 shows a diagram illustrating beam tracking of transmit beams; Fig. 25 shows another diagram illustrating beam tracking of receive beams; and Fig. 26 shows another diagram illustrating beam tracking of transmit beams. DETAILED DESCRIPTION OF THE EMBODIMENTS [0016] In the past few years, data transmission in the 60 GHz frequency band (mm-wave) has attracted attention since it is a promising alternative to the congested frequency bands in the unlicensed spectrum at 2.4 and 5 GHz. Furthermore, high bandwidths (>2 GHz) are available which allow for high data rate in excess of several Gbps. However, free space path loss is significantly higher compared to the 2.4 and 5 GHz band due to the high frequency and oxygen absorption in the 60 GHz frequency band. In order to anyway achieve reasonable link lengths, directional communications by beamforming is applied. Thereby, transmitter and/or receiver feature steerable antenna arrays, which can form a directive beam towards the remote station. The employed beams feature a large gain (to overcome high path loss) but have a very narrow spatial shape. Thus, beam alignment is crucial for good system performance. [0017] Besides, 60 GHz communication systems have also good properties for positioning. The high sampling frequency which is required to resolve large bandwidths allows for highly accurate timing measurements between transmitter and receiver without applying rather complex super resolution algorithms. Timing measurements are used to determine the time of flight (ToF) between two communication devices. For example, 2 GHz sampling frequency results in range measurements with a maximum error of ±7.5cm. Further, the direction of a beam holds angular information, which gives the direction of the device location. As shown in Fig. 1 , both timing and angular information can be readily used to determine the position of a communication device B relative to a communication device A. In two dimensions, timing information defines a circle whereas angular information gives the circle segment where device B is located. In contrast to other positioning techniques such as triangulation, which require several anchors with known location, this positioning technique requires only a single anchor (communication device A). In three dimensions, ToF and angular information define a spherical segment. Based on this information, a relative positioning estimate can be obtained, which may be already sufficient for some applications. If the absolute position is required for device B, then the absolute position of device A needs to be known. [0018] The optimum beam alignment between communication device A and B depends on the link application. If the link is applied for data communications (which is the case in the IEEE802.1 1 ad amendment), the best beam is defined by the beam(s) or sector(s) which maximize(s) channel capacity. In case of SISO systems, i.e. communication device A and B have only one transmit or receive antenna array, this is equal to maximizing received power or signal-to-noise ratio (SNR). However, this beam alignment strategy is not optimal for positioning and can lead to wrong positioning information. [0019] Fig. 2 shows two scenarios with optimal beam selection for data communications. In the situation shown in Fig. 2A, the line-of-sight (LOS) exists and the best sectors are aligned to the LOS path. In the situation shown in Fig. 2B, an obstacle 01 between device A and B attenuates the LOS path and a further obstacle 02 acts as a reflector. Given that the attenuation of the LOS component is stronger than the reflection loss, the best sectors are chosen such that both align towards the reflecting obstacle 02 yielding maximum SNR or received power. Thus, the weak LOS path due to partial blocking as shown in Fig. 2B leads to data communication via the reflector 02. [0020] From a positioning point of view, the scenario shown in Fig. 2A will provide correct information on ToF and sector angle, whereas the scenario shown in Fig. 2B will provide incorrect information. Consequently, the ToF is estimated too long, i.e. the distance between both communication devices A and B is estimated too large and the sector angle is wrong as well, as communication device B is expected to be above and on the left with respect to communication device A although it resides directly above. Thus, beam alignment for communication is not suitable for positioning. [0021] The channel conditions in the scenario shown in Fig. 2B are often referred to as a near NLOS (n-NLOS) condition, where the channel consists of a weak LOS and strong NLOS components. Power of the LOS part is still assumed to be greater than zero; otherwise the channel would be considered as NLOS. Further, the sectors are formed by very directive antenna characteristics and strongly suppress impulse response information which is not aligned to the main direction of the sector. Therefore, it is impossible or very complicated to reconstruct the LOS component out of baseband data generated by sectors favoring any NLOS part of the channel. [0022] The present disclosure makes use of beamforming for positioning applications in a wide sense, i.e. the disclosed methods and devices are not limited to 60GHz systems only, but may also be applied with regular multi-antenna communication schemes, where beam- forming is not necessarily required for data communications. Fig. 3 shows a diagram illustrating the difference between beamforming maximizing channel capacity and beam- forming for LOS detection according to the present disclosure. Briefly summarized, as shown in Fig. 3, the NLOS path P1 may be used for (regular) communication, i.e. data sector selection, and the LOS path P2 may be used for positioning, i.e. positioning sector selection. [0023] In general, the following embodiments of beamforming architectures shown in Figs. 4 and 5 may be applied for beamforming in the context of the present disclosure. Fig. 4 shows a typical mm-wave transmitter architecture 100, where beamforming is performed in analog domain. Input data is processed in baseband processing unit 101 (e.g. including an encoder, a modulator, etc.) and allocated to an RF chain (RFC). The RFC is divided in part A (RFC-A 102), which holds DAC, filtering, and frequency up-converter, and part B (RFC-B 103), which holds a power amplifier, respectively. The number of RFC-A elements 102 corresponds to the number of antenna arrays (one in Fig. 4). A controllable phase shifter 104 is located between RFC-A 102 and each RFC-B 103. The phase shifters 104 are controlled by a beam control unit 105 such that directive beams can be formed by the antenna array(s). The number of phase shifters 104 and RFC-B elements 103 corresponds to the number of antenna elements in the associated antenna array. In contrast, Fig. 5 shows a regular multi-antenna transmitter architecture 1 10, where digital beamforming is performed. Two baseband processing units 1 1 1 , 1 12 are provided, one before and one after the beamforming unit 1 13. Each transmit antenna features a full RFC 1 14. Beamforming is performed in digital domain and is done such that an ensemble of transmit antennas can form directive beams. The receivers have a similar setup as the transmitters but perform inverse operations. Beamforming at the receiver side is done in a similar way as at the transmitter so that several receive beams can be selected. To summarize, different beamforming architectures can generate several directive beams or sectors at transmitter side and/or receiver side, which can be controlled. One embodiment of this disclosure concerns a dedicated beam selection algorithm for positioning. Instead of selecting the sector which maximizes received power, the sector may be selected which minimizes time of arrival (ToA). For outlining the concept, the simplified (and non-limiting) scenario shown in Fig. 6A is considered, in which communication device A can form several directive beams, whereas communication device B has a (quasi) omni-directional pattern symbolized by the flat ellipsoid. This setup is often referred to as directional-to-omni (D20), whereas the setup of Fig. 2 shows a directional-to-directional (D2D) setup. Extension from D20 to D2D is straightforward and will be addressed later. In the following, it is assumed that device A is in transmit mode, whereas device B is in receiving mode. A 2D omnidirectional channel model between both devices is given by the complex-valued multi-dimensional function Η(φ, η). Thereby, ψ denotes the angle of departure (AoD) at device A and n denotes discrete time. During a training phase, the transmitter sweeps through all or a subset of sectors while the receiver is listening (quasi) omni-directional. The receiver is able to estimate the impulse response which is achieved for a certain sector. Directive transmission can be considered as spatial filtering of Η(φ, η) with respect to φ. Each transmit sector a can be represented by a function Ta which filters Η(φ, η) to get the impulse response ha(n) seen or measured by the receiver Thereby, Ta typically enhances spatial components in the main direction of sector characterized by the AoD φα whereas it attenuates components outside the main direc¬ tion. [0028] This concept can be generalized to a 3D environment supporting D2D. In this case, H{ h*kr)]} with hschr > 0 being a detection threshold depending on noise and/or training sequence parameters. A simple but effective threshold can be ft ter = ma_xj i3(ti) | which is equiva¬ lent to considering the maximum of \ hs(n) \. This selection is effective because the sectors perform spatial filtering, i.e. sectors s have directive characteristics which amplify multi- path components coming from angles equal or close to main direction and suppress components from outside main direction. For conventional transmitter architectures, which implement 020, this threshold selection has poor performance, because the weak LOS path is not detected since its magnitude is typically well below the maximum of | hjn) \ in n-NLOS. Thus, the directivity of the antenna sectors is gainfully used for positioning. [0032] Determination of t¾ requires timing synchronization between transmitter and receiver. This synchronization is required to be stable during the entire impulse response estimation process, i.e. timing synchronization must be either equal or appropriately corrected for all h (n). Given that impulse responses are successively estimated in a frame structure, the time base of each 3(n) has to be seen in reference to the first or previous received sector. An adequate positioning sector sj is given by that sector which achieves highest magnitude at index n0, i.e. j = arg maxrtJ | h£n0) L In contrast, beam selection for communications is typically done by sf = m a ^∑„=o i ^ (« ) i 2■ The related ToA value is deduced by t0 = nQTs— e with being the sampling interval and e being implemen- tation dependent processing delay. Further, the corresponding angles given by sector are defined as follows ψβ!φ^!θα!θύ = arg(sj). For some applications, ToA accuracy may be crucial and the quantization of t0 to multiples of Ts may not be sufficient. In this case super-resolution algorithms (e.g. MUSIC as described in X. Li and K. Pahlavan, "Super-Resolution TOA Estimation With Diversity for Indoor Geolocation," IEEE Transactions on Wireless Communications, vol. 3, no. 1 , pp. 224-234, 2004) are applicable. Thereby, the first arrival path deduced by MUSIC shall be considered for each hs(n). Minimization of the first arrival path with respect to all or a subset of sectors s yields n0. In contrast to the method above, n0 can now be a rational number and the equations from above change as follows p The MUSIC operator in the equation above retrieves from each impulse response the non-integer spaced impulse response taps. In a real system, it is desirable to first perform beamforming for bidirectional communications, i.e. best communication or data sectors for device A transmitting to device B are determined. As a second step, beamforming for positioning is performed which is only applied for channel sounding, and/or for positioning-related data communi- cations. Regular data transfer is still performed by using the best sectors for communications or data exchange. Thus, the extent of the sector search for positioning is flexible and in general independent of the communication or data sector search. The following search spaces or beam setups are reasonable: (a) D20, according to which communication device A or B applies a directive pattern, whereas communication device B or A features a (quasi) omni-directional receive characteristic; and (b) D2D, according to which both communication device A and B apply directive patterns for transmitting and receiving data, respectively. The options are sorted by complexity or training overhead in an increasing order. However, option (a) results in a worse performance compared to option (b) because spatial filtering is stronger in case (b) since two directive patterns Ta and Rb are applied to H. Since option (a) considers a (quasi) omni-directional receive or transmit pattern, this yields less attenuation of the NLOS components of the channel outside the main direction. This holds for both ToA and angular information. Furthermore, angular information φα,θα is available in case (a) and (b), whereas φ ¾,% is only available in case (b) or in bidirectional D20 in case of a reciprocal channel (see later). In fact, BJ¾are sufficient for positioning as shown in Fig. 1 , but <¾,,% can be used for detection of a NLOS scenario. Fig. 6B shows the basic method in 2D. In order to evaluate the angular information of device A and B, the orientation of the antenna array of communication device B relative to device A φάιΒ is required, as will be explained in more detail below for a hybrid application. The raw angular information