Indoor optical wireless communication system with continuous and simultaneous positioning

Ke Wang; Tingting Song; Sithamparanathan Kandeepan; Hongtao Li; Kamal Alameh

doi:10.1364/OE.409395

1. Introduction

The optical wireless communication (OWC) technology has attracted intensive interests during the past years as a promising candidate to provide high-speed wireless connections in indoor environments [1–4]. Compared with traditional radio-frequency (RF) wireless communications, the OWC uses the license-free optical spectrum and has broad bandwidth available. Therefore, the OWC technology has been considered as one possible solution for the emerging beyond-5G (B5G) communications to realize ultra-high data rate, ultra-low latency and ultra-large capacity [5,6]. In addition, the OWC system is also free of conventional electromagnetic interference (EMI) and has high communication security [7].

In indoor OWC systems, including both visible light communication (VLC) systems and infrared OWC systems, both line-of-sight (LOS) and non-line-of-sight (NLOS) links can be explored for wireless connections [1–4,8]. Compared with the NLOS link, the LOS links are normally preferred due to the high energy efficiency and the minimal multipath dispersion. However, good alignment between the transmitter and the receiver is required. In addition, the transmission power of indoor OWC systems can be limited due to safety regulations, especially for infrared OWC systems. Therefore, knowing the location of users and providing LOS OWC connections correspondingly are highly desirable [4,9–12]. The indoor user positioning is also highly demanded for other applications in the anticipated B5G systems, such as realizing intelligent services and providing location-based e-health or advertisement [5,6].

The GPS technology has been widely used in outdoor positioning applications. However, it is not suitable for indoor positioning, and a number of alternative technologies have been studied to satisfy the indoor positioning demand. RF solutions, such as the WiFi-based and Zigbee-based, have been proposed and demonstrated [13,14], with indoor positioning obtained using the received signal strength (RSS), time-of-arrival (TOA) or angle-of-arrival (AOA) information. However, RF-based indoor positioning systems typically suffer from EMI and high level of signal diffractions and reflections, and hence, they have limited accuracy.

In addition to RF-based solutions, the OWC technology has also been widely studied for the indoor positioning function, and both the visible and near-infrared wavelengths have been investigated [9–12]. RSS, TOA and AOA information have been explored in OWC indoor positioning systems, and centimeter-scale positioning accuracy has been demonstrated [15]. Therefore, both high-speed wireless connections and accurate positioning can be achieved using the same system architecture [10]. Compared with TOA- and AOA-based methods, the RSS-based indoor OWC positioning method is normally simpler to implement, since it does not require accurate transceiver synchronization or a receiver capable of incidence angle detection [16]. Therefore, a large portion of previous studies have focused on the RSS-based solution [17–21].

In typical RSS-based indoor OWC positioning systems, multiple transmitters are used and the positioning is achieved by monitoring the light signal from each transmitter with known location and using the optical path-loss model. In order to distinguish the signals from different transmitters, a widely used method is time-division-multiplexing (TDM), where each transmitter sends out the positioning signal individually within the dedicated positioning time slot in order to avoid RSS interference [21]. However, the wireless communication needs to be paused for sending dedicated positioning signals, and this reduces the effective OWC data rate and limits the overall communication capacity, especially when the number of transmitters is large. In addition, since both positioning and wireless communication functions share the same system, the user positioning can only be obtained at discrete positioning time slots and continuous positioning cannot be provided.

To solve these issues, a number of advanced techniques have been proposed. The use of RF carriers has been studied [17], where each transmitter is allocated a dedicated RF carrier and the positioning signals from different transmitters can be separated via the filtering of different RF carrier frequencies. The concept has been extended with the use of orthogonal-frequency-division-multiplexing (OFDM) for optical wireless communication, where a dedicated OFDM sub-carrier is allocated to each transmitter for the positioning purpose [18]. However, this method requires dedicated RF carriers or subcarriers to carry the positioning signals, and this reduces the useable bandwidth of each transmitter, and hence, affects the communication data rate and system capacity. The use of the optical code-division-multiplexing-access (OCDMA) principle in indoor OWC positioning systems has also been investigated, such as the use of optical orthogonal codes [19]. In this type of systems, an optical orthogonal code with length n and weight w is allocated to each transmitter, and the positioning signals from different transmitters can be separated using correlation-type detectors. The Hadamard matrix has also been explored, where each transmitter is allocated a row or a column of the Hadamard code [20]. However, normally, relatively long codewords are required, which reduce the effective data rate and throughput of the wireless connections in OWC systems.

In this paper, we propose and demonstrate a novel filter-enhanced solution to provide communication and continuous user positioning simultaneously in indoor OWC systems. By dedicating a spatial filter to each transmitter, the RSS information from multiple transmitters can be obtained from the received wireless communication signal directly and simultaneously. Therefore, the dedicated positioning time slots, RF carriers or codewords are no longer needed, and the wireless communication data rate and throughput are not affected. In addition, the positioning information can be obtained continuously from the wireless communication signal to enable real-time user tracking. The proposed user positioning method is investigated and experimentally demonstrated in a near-infrared OWC system setting with laser-based transmitters. Results show that an average positioning accuracy of 5.41 cm can be achieved together with simultaneous 2.5 Gb/s wireless communication. The proposed filter-enhanced indoor positioning method can also be applied to VLC systems with LEDs and we demonstrate the feasibility via simulations. Results show that similar positioning accuracy can be achieved without interruption to the wireless data communication.

2. Filter-enhanced indoor OWC system

2.1 RSS-based indoor OWC positioning system

The general architecture of the indoor OWC system with both data communication and user positioning functions is shown in Fig. 1(a). The centralized architecture is considered, where a central unit (e.g., a home gateway) is connected to multiple rooms via the in-building fiber distribution network [16]. Inside each room, here we consider the case of near-infrared indoor OWC system with ${M_{tx}}$ transmitters and ${M_{rx}}$ receivers. As shown later in Section 4, the proposed method can also be applied to VLC systems. We assume that the OWC transmitters are laser-based and the receivers are photodiode (PD)-based. Due to the laser safety regulation, we assume only a limited area is covered by each signal beam and the transmitters are capable of the beam steering function [10]. We also assume that repetition coding (RC) is applied to the transmitters for wireless communications, since compared with other transmitter diversity scheme such as space-time-block coding (STBC), the RC scheme has shown to be highly effective in indoor OWC systems in providing better robustness, increasing the data rate, or extending the signal coverage area [22].

Fig. 1. (a) Architecture of indoor OWC system with RC; (b) transmitter architecture of the proposed filter-enhanced indoor OWC positioning system; and (c) receiver architecture of the proposed filter-enhanced indoor OWC positioning system.

Download Full Size | PDF

Assuming that the optical signal to be transmitted by OWC transmitters is ${\boldsymbol {TX}}$, which is a matrix with ${M_{tx}} \times 1$ dimension, and the channel gain is ${\boldsymbol H}$, which has a dimension of ${M_{tx}} \times {M_{rx}}$, then the detected electrical signal ${\boldsymbol {RX}}$ can be expressed as:

(1)$${\boldsymbol {RX}} = {\boldsymbol R\; } \odot {\boldsymbol \; }{({{\boldsymbol T}{{\boldsymbol X}^T}{\boldsymbol H}} )^T} + {\boldsymbol N}$$

where ${\odot} $ represents the pointwise multiplication, ${\boldsymbol R},$ with a dimension of ${M_{rx}} \times 1,$ represents the responsivity of the PDs, ${\boldsymbol T}{{\boldsymbol X}^T}$ represents the transpose of ${\boldsymbol {TX}}$, and ${\boldsymbol N},$ with a dimension of ${M_{rx}} \times 1,$ represents the noise of the received signal, which mainly consists of the background light and pre-amplifier-induced noise sources [23,24]. The $j$th element of matrix ${\boldsymbol {RX}}$ can be written as:

(2)$$r{x_j} = \mathop \sum \nolimits_{i = 1}^{{M_{tx}}} {r_j} \cdot {h_{ji}} \cdot t{x_i} + {n_j}$$

where $t{x_i}$ is the $i$th element of matrix ${\boldsymbol {TX}}$ that represents the signal transmitted by the transmitter i, ${r_j}$ is the responsivity of the $j$th receiver, ${h_{ji}}$ is the channel gain from transmitter i to receiver j, and ${n_j}$ is the noise of the $j$th receiver. Since here we consider the near-infrared indoor OWC system with laser transmitters and LOS links, the optical signal intensities at the receiver side can be modelled as approximate Gaussian distribution. Therefore, the received signal power from the transmitter i ($i = 1,\; 2,\; \ldots ,\; {M_{tx}}$) to the receiver j ($j = 1,\; 2,\; \ldots ,\; {M_{rx}}$) can be expressed as:

(3)$${P_{r,ji}} = \frac{{2{P_{t,i}} \cdot {A_{r,j}}}}{{\pi \omega _i^2}}exp\left( { - \frac{{2d_{ji}^2}}{{\omega_i^2}}} \right)$$

where ${P_{t,i}}$ is the signal power radiated by transmitter i, ${A_{r,j}}$ is the aperture size of the receiver j, ${\omega _i}$ is the beam footprint of the optical signal from transmitter i, and ${d_{ji}}$ is the distance from the $j$th receiver to the $i$th beam center. We assume that the OWC transmitters are located at the ceiling with a height of ${z_t}$ and pre-known locations of $({{x_{t,i}},{y_{t,i}}} )$, where $i = 1,\; 2,\; \ldots ,\; {M_{tx}}$, and the users (i.e., receivers) are located on a communication floor with a height of ${z_c}$ and unknown locations $({{x_{r,j}},{y_{r,j}}} )$. We also assume that the center of signal beam radiated by the transmitter i at the communication floor is $({{x_{b,i}},{y_{b,i}}} )$. Since the beam steering angles are controlled by the transmitters, the beam center locations are known. Given these conditions, the distance ${d_{ji}}$ from the $j$th receiver to the $i$th beam center can be expressed as:

(4)$${d_{ji}} = \sqrt {{{({{x_{r,j}} - {x_{b,i}}} )}^2} + {{({{y_{r,j}} - {y_{b,i}}} )}^2}} $$

Based on the received signal power, the OWC channel gain from transmitter i to receiver j can be calculated as:

(5)$${h_{ji}} = \frac{{2{A_{r,j}}}}{{\pi \omega _i^2}}exp\left( { - \frac{{2{{({{x_{r,j}} - {x_{b,i}}} )}^2} + {{({{y_{r,j}} - {y_{b,i}}} )}^2}}}{{\omega_i^2}}} \right)$$

Since the aperture size and the optical beam footprint are pre-known, when there are more than two transmitters available, the positioning of user can be obtained by monitoring the RSS and estimating the corresponding channel gain information from each transmitter. However, as described by Eq. (2), the signal detected by each receiver can be considered as the weighted sum of all transmitted signals and hence, the individual channel gains cannot be recovered directly. To obtain the RSS and channel gain from each transmitter, the TDM method can be used, which has been demonstrated in near-infrared OWC systems [10]. In VLC systems, TDM, RF carriers or sub-carriers, and CDMA methods have been proposed and investigated [15,17–20], as discussed in Section 1. However, due to the dedicated positioning time slots, RF carriers or relatively long codewords, the OWC data rate and system capacity are reduced. In addition, it is challenging to apply these positioning schemes together with RC, as this affects the OWC robustness and coverage area.

2.2 Filter-enhanced RSS-based indoor OWC positioning system

To solve these limitations, we propose a spatial-filter-enhanced indoor OWC positioning scheme, as illustrated in Fig. 1(b). After applying RC to the transmitted data and modulating them onto optical carriers, rather than sending the modulated optical signals out directly for data communication or positioning, a dedicated waveform shaping filter is allocated to each OWC transmitter. Assume ${s_i}$ and ${f_i}$ are the symbol to be transmitted and the response of filter applied to transmitter i, respectively, due to the use of RC, the symbols from all transmitters are the same, i.e., ${s_i} = s$. In this case, the signal transmitted by the transmitter i and the signal detected by the receiver j can be described as:

(6)$$t{x_i} = {s_i} \otimes {f_i} = s \otimes {f_i}$$

(7)$$r{x_j} = \mathop \sum \nolimits_{i = 1}^{{M_{tx}}} {r_j} \cdot {h_{ji}} \cdot ({s \otimes {f_i}} )+ {n_j}$$

where ${\otimes} $ denotes the convolution operation. As shown in Fig. 1(c), after being detected by the PD, the converted signal passes through matched filters before symbol detection. Assuming the response of the matched filter corresponding to transmitting waveform shaping filter ${f_i}$ is ${g_i}$, then the signal after the $k$th ($k = 1,\; 2,\; \ldots ,\; {M_{tx}}$) matched filter can be expressed as:

(8)$$rx_{jk}^{\prime} = \mathop \sum \nolimits_{i = 1}^{{M_{tx}}} {r_j} \cdot {h_{ji}} \cdot ({s \otimes {f_i}} )\otimes {g_k} + n_{jk}^{\prime}$$

where $n_{jk}^{\prime}$ is the noise after the matched filter. In order to obtain the channel gain from each transmitter simultaneously without requiring dedicated time slots, RF carriers or relatively long codewords, here we propose to apply orthogonal waveform shaping filters to the transmitters. The filters applied to the transmitters satisfy

(9)$${f_i} \otimes {g_k} = \left\{ {\begin{array}{{cc}} {C,\; \; when\; i = k}\\ {0,\; \; when\; i \ne k} \end{array}} \right.$$

where C is a constant. Given this property of the filters, the output of the matched filter k of receiver j shown in Eq. (8) can be re-written as:

(10)$$rx_{jk}^{\prime} = r_j^{\prime} \cdot {h_{jk}} \cdot s + n_{jk}^{\prime}$$

where $r_j^{\prime} = C{r_j}$ and it is known for the receiver j. Therefore, the channel gain from transmitter k to receiver j can be estimated as:

(11)$${\tilde{h}_{jk}} = \frac{{{P_{filter,jk}}}}{{{P_{t,k}}}}$$

where ${P_{filter,jk}}$ is the optical power of the signal output from matched filter k of receiver j. Assuming the output signal from matched filter j at time t is $rx_{jk}^{\prime}(t )$ and the unit time period is T, then ${P_{filter,jk}}$ can be calculated as:

(12)$${P_{filter,jk}} = \mathop \smallint \nolimits_{{T_0}}^{{T_0} + T} \frac{{rx_{jk}^{\prime}(t )}}{{r_j^{\prime}}}dt. $$

Therefore, by using matched filters in parallel, the channel gains from all transmitters to the receiver j can be estimated simultaneously, and the positioning of receiver j can be obtained using Eq. (5). From the channel gain estimation process shown above, it can be seen that the noise existing in the received optical signal affects the channel gain estimation accuracy, and hence, it limits the positioning accuracy that can be achieved. This limitation is common in RSS-based indoor OWC positioning systems and its impact has been analysed in a previous study [21], where the positioning accuracy can be improved by reducing the received background light power (noise) or increasing the optical signal power (signal).

Unlike previously studied methods, the proposed indoor OWC positioning system employs spatial waveform shaping filters, hence, the RSS information can be obtained directly from the wireless communication signals to realize the positioning function. Therefore, the wireless data transmissions are not interrupted, and the data rate and system capacity are not reduced. In addition, since the channel gain estimation ${\tilde{h}_{ji}}$ can be obtained using real-time wireless communication signals, continuous positioning function can be realized, thus enabling better tracking of users. Furthermore, the proposed method is compatible with the RC transmitter diversity scheme, and hence, obtaining the positioning information does not reduce the communication robustness or coverage area. The summary of comparison amongst different schemes to obtain RSS information for positioning is shown in Table 1.

Table 1. Comparison of methods for RSS information

View Table | View all tables in this article

The frequency-division-multiplexing (FDM) scheme can also be used to realize the positioning function from real-time wireless communication signals, where each transmitter occupies a sub-band in the frequency domain and the receivers monitor the received signal power levels in different sub-bands for collecting RSS information. However, the FDM-based method has the fundamental limitation of reduced data rate per transmitter compared with the single transmitter case, as only a fraction of available frequency can be used. On the other hand, the proposed filter-enhanced positioning system does not have this limitation. With the use of spatial waveform shaping filters, multiple transmitters can use the same full frequency band, and hence, the data rate per transmitter is the same as the single transmitter case. This advantage of proposed scheme becomes more significant when the number of transmitters increases, since the data rate per transmitter further reduces in the FDM-based scheme due to the sharing of the same frequency band by more transmitters.

3. Experiments and results

3.1 Experimental setup

To demonstrate the proposed filter-enhanced indoor OWC positioning system, we considered the simple case with two near-infrared transmitters. This can be easily extended to the system with more transmitters following the principle discussed in Section 2.2. There are multiple ways to design the waveform shaping filters applied to the two transmitters to satisfy the condition described by Eq. (9), and here we followed the method applied in our previously studied RC OWC system [25]. The responses of waveform shaping filters can be described as:

(13)$${f_1}(t )= a(t )\cdot \cos ({2\pi {f_c}t} )$$

(14)$${f_2}(t )= a(t )\cdot \sin ({2\pi {f_c}t} )$$

where $a(t )$ is the filter base waveform which is selected as the square root raised cosine function, and ${f_c}$ is the carrier central frequency. By applying matched filters at the receiver side and monitoring the output signal from the two receiving filters, the channel gains from two transmitters can be obtained and the positioning can be estimated (detailed in Section 3.2).

Proof-of-concept experiments were carried out to demonstrate the proposed filter-enhanced indoor OWC positioning system, and the experimental setup is shown in Fig. 2. The RC signals were generated using an arbitrary waveform generator (AWG) with 2²³–1 PRBS data, and they were modulated onto the optical carriers generated by lasers using two Mach-Zehnder modulators (MZMs). The waveform shaping filters described by Eq. (13) and Eq. (14) were also implemented with the AWG. The impulse responses of the two orthogonal waveform shaping filters are shown in the inset of Fig. 2. Both filters were implemented as finite response filters (FIR). The optical signals were then radiated to the free-space via two OWC transmitters, and each OWC transmitter consisted of a lens to control the beam divergence (i.e., ${\omega _i}$ at the receiver side) and a steering mirror to control the beam direction (i.e., $({{x_{b,i}},{y_{b,i}}} )$ at the receiver). After free-space transmission, the optical signals were collected by an optical concentrator and then detected by a PD. Here we only considered the case with one receiver, whilst multiple users can be supported simultaneously, as the positioning function is realized by each receiver. The electrical signal converted by PD then passed through two matched filters, and the outputs from matched filters were monitored for the RSS information and decoded for the wireless data communication.

Fig. 2. Experimental setup for demonstrating the proposed indoor OWC positioning system.

Download Full Size | PDF

In the experiment, the light wavelength of both transmitters was 1550.12 nm and the radiation power from each transmitter was 7 dBm, within the maximum power level allowed by the laser safety regulation. The MZM and the PD had about 10 GHz and 2.5 GHz electrical bandwidth, respectively. The wireless communication data rate was 2.5 Gb/s with the simple on-off-keying (OOK) modulation format. The two transmitters were repetition coded (i.e., same data for both transmitters). Here the data rate was mainly limited by the PD bandwidth, and much higher data rate can be achieved [2]. The optical signal free-space transmission distance in the z-direction (as shown in Fig. 2) was 2 m and background light from illumination lamps were incorporated in the measurement. We assume the center of transmitters was located at $({x,y} )= ({0,0} )$.

3.2 Demonstration of simultaneous positioning and wireless communication

To demonstrate the proposed filter-enhanced indoor OWC positioning scheme, we first adjusted the lenses in two transmitters so that the beam footprints at the receiver side were the same, where ${\omega _1} = {\omega _2} = 60$ cm. The receiver location was changed along the y-direction ($x = 10\; cm$) as shown in Fig. 2 and the channel gains were estimated using Eq. (11). Based on the RSS, the location of receiver was then estimated with Eq. (5). The two beam centers were selected at $({{x_{b,1}},{y_{b,1}}} )= ({10,\; 10} )$ cm and $({{x_{b,2}},{y_{b,2}}} )= ({ - 10,\; - 10} )$ cm, respectively. Since there are two possible receiver locations satisfying Eq. (5) with the use of two transmitters, in the initialization stage, one of the two transmitters was steered (transmitter 2 changed in the experiment) to work as a third ‘virtual’ transmitter and generate a third ‘virtual’ beam center, where $({{x_{b,3}},{y_{b,3}}} )= ({5,\; - 5} )$ cm. It is important to note that in practical scenarios such additional optical signal adjustment is only needed in the initialization stage when two transmitters are used. This is because that the proposed method has continuous positioning capability and the user’s trajectory is continuous as well. Therefore, only one of the two subsequent receiver locations satisfying Eq. (5) is possible and the subsequent positioning function can be achieved without additional transmitter adjustment. In addition, when there are more than two transmitters available, the user positioning can be obtained without any transmitter adjustment process, including the initialization stage.

The positioning results are shown in Fig. 3(a). At each location, the estimation process was conducted for three times, and the average results are shown in the figure (no major difference observed amongst the estimation results). It can be seen that the positioning of receiver inside the area covered by both transmitters can always be obtained, and the maximum positioning error is about 7.02 cm. The positioning accuracy degrades when the receiver moves away from the signal beam centers. This is mainly because that the background light is almost evenly distributed in the setup, whilst the signal power decreases when the receiver approaches the beam edges. Therefore, the impact of noises becomes larger and the positioning accuracy becomes worse. As discussed in Section 2.2, one important advantage of the proposed filter-enhanced scheme is the capability of realizing the positioning function from the wireless communication signals. This was demonstrated in the experiment as well, where the positioning was estimated from the 2.5 Gb/s RC wireless communication signals. The received wireless communication signal waveforms after the two matched filters at locations of (10, 0) cm, (10, 10) cm and (10, 20) cm are shown in Fig. 3(b) – Fig. 3(d), respectively. It can be seen that at different locations, the strengths of communication signals from the two transmitters are different, which represent the RSS information for positioning. In addition, the repetition coded communication signals from both transmitters can be separated at all locations. Therefore, both wireless data transmission and user positioning functions can be achieved simultaneously. The wireless communication performance was also experimentally characterized, and the bit-error-rate (BER) performance is shown in Fig. 3(e), revealing that the BER performance follows the similar trend as the positioning accuracy, where both BER and positioning accuracy become worse when the receiver moves toward the beam edges. The worse BER performance was also mainly caused by the lower signal power.

Fig. 3. (a) Estimated positioning results when the receiver moves along $x = 10\; cm$; (b)–(d) received OWC signal waveforms after matched filters at (10, 0) cm, (10, 10) cm and (10, 20) cm, respectively; and (e) measured BER of the wireless communication function.

Download Full Size | PDF

More measurements were carried out when the receiver moved along the $x = 15\; cm$ and $x = 20\; cm$ lines. The positioning and wireless communication BER results are shown in Fig. 4. Similar with the results shown in Fig. 3, better positioning accuracy and BER performance were achieved when the receiver is closer to the beam centers, which is due to the stronger received signal strength. In addition, when the receiver moved along the $x = 10\; cm$, $x = 15\; cm$ and $x = 20\; cm$ lines, the average positioning accuracy is 4.86 cm, 5.95 cm and 6.37 cm, respectively. Compared with the case when the receiver moved along the $x = 10\; cm$ and $x = 15\; cm$ lines, both the positioning accuracy and the BER were worse when the receiver moved along the $x = 20\; cm$ line. This is mainly because that the locations of two beam centers were further away from the $x = 20\; cm$ line, leading to larger impact of noises.

Fig. 4. (a) Positioning error when the receiver moves along $x = 15\; cm$ and $x = 20\; cm$; and (b) the BER performance of the wireless communication function.

Download Full Size | PDF

In previous characterizations, the footprints of the two signal beams were selected to be the same. We also demonstrated the proposed filter-enhanced indoor OWC positioning system with different beam footprints, where ${\omega _1} = 50$ cm and ${\omega _2} = 60$ cm. The beam centers were kept unchanged at $({{x_{b,1}},{y_{b,1}}} )= ({10,\; 10} )$ cm and $({{x_{b,2}},{y_{b,2}}} )= ({ - 10,\; - 10} )$ cm, respectively. The positioning and wireless communication performances are shown in Fig. 5. The average positioning accuracy was 4.52 cm, 5.11 cm and 5.64 cm along the $x = 10\; cm$, $x = 15\; cm$ and $x = 20\; cm$ lines, respectively. The average positioning accuracy was slightly better than that shown in Fig. 3 and Fig. 4, and it was mainly due to the smaller positioning errors achieved when the receiver was close to the signal beam centers. Due to the use of a smaller beam footprint for the signal beam 1, a larger signal power was collected and better positioning accuracy was achieved when the user was close to the beam center $({{x_{b,1}},{y_{b,1}}} )$. On the other hand, when the receiver was moving toward the edge of signal beam 1 (e.g., $y ={-} 30\; cm$), the positioning accuracy degraded rapidly and became worse than that achieved with a larger beam footprint. This is mainly caused by the smaller received signal power from the signal beam 1. As shown by Eq. (3), due to the approximate Gaussian distribution of the optical power at the receiver side, increasing the distance from the center of the signal beam decreases the received signal power, especially when the signal beam footprint is relatively small. Hence, the impact of noise becomes larger and the positioning accuracy decreases. This trend is also reflected in the measured BER performance, where compared with the results shown in Fig. 3 and Fig. 4, the BER was better at locations close to the signal beam centers and worse at locations close to the beam edges. In addition, it can be seen from the results that the positioning and wireless communication functions can always be achieved simultaneously, demonstrating the capability of proposed filter-enhanced indoor OWC positioning system.

Fig. 5. (a) Positioning error under different beam footprints; and (b) the BER performance under different beam footprints.

Download Full Size | PDF

4. Filter-enhanced indoor positioning in VLC systems

4.1 Filter-enhanced RSS-based indoor VLC positioning system

In the previous analysis and demonstrations, the near-infrared indoor OWC system using the near-infrared wavelength and laser transmitters was considered. In addition to lasers, the LEDs have also been widely investigated for indoor wireless connections [26]. In this section, we demonstrate the feasibility of proposed filter-enhanced indoor positioning scheme in VLC systems via numerical simulations.

Here, we consider an indoor environment with ${M_{tx}}$ LED transmitters. For each LED, it can be modelled as the Lambertain source due to its beam divergence. Considering the LOS link, the wireless channel gain from LED transmitter i to the receiver j can be modelled as [20]:

(15)$${h_{ji}} = \left\{ {\begin{array}{lc} {\frac{{({{m_i} + 1} ){A_{r,j}}}}{{2\pi L_{ji}^2}}{{\cos }^{{m_i}}}({{\varphi_{ji}}} )},&{\psi_{ji}} \le {\psi_c}\\ {0},&{\psi_{ji}} > {\psi_c} \end{array}} \right.$$

where ${A_{r,j}}$ is the effective size of the receiver, ${\varphi _{ji}}$ and ${\psi _{ji}}$ are the angle of irradiance and the angle of incidence between LED transmitter i and receiver j, respectively, ${\psi _c}$ is the field-of-view (FOV) of the receiver j, ${m_i}$ is the order of Lambertain emission of LED transmitter i, and ${L_{ji}}$ is the free-space signal propagation distance, which can be described as:

(16)$${L_{ji}} = \sqrt {{{({{x_{t,i}} - {x_{r,j}}} )}^2} + {{({{y_{t,i}} - {y_{r,j}}} )}^2} + {{({{z_t} - {z_c}} )}^2}} $$

In order to enable simultaneous wireless communication and positioning functions, as described in Section 2, a waveform shaping filter is applied to each LED transmitter and the RC is applied to multiple transmitters for data transmission. As with the near-infrared OWC system, matched filters are used at the receiver and signals from different transmitters can be separated due to the orthogonality of waveform shaping filters, as shown by Eq. (9) and Eq. (10). Therefore, the RSS from each LED transmitter can be obtained and the corresponding distance between the transmitter and receiver can be estimated. When more than three transmitters are available, the positioning function can be realized via trilateration.

4.2 Simulations and results

We demonstrated the feasibility of proposed filter-enhanced VLC positioning system via simulations. In the simulations, we considered a room with the size of 4 m × 4 m × 3m, three LED transmitters and one receiver. The transmitter locations were selected at (1 m, 1 m, 3m), (3 m, 1 m, 3 m) and (2 m, 3 m, 3 m), and the receiver was located on a communication floor with a height of 1 m. We assumed that RC was applied to the transmitters (i.e., the same data sent by all transmitters), and the LEDs were modulated using the 4-pulse-amplitude-modulation (4-PAM) format with a symbol rate of 1 MBaud/s. More advanced modulation formats and much higher symbols rates can be used as demonstrated in previous studies [26]. In this proof-of-concept work, we just considered simple data modulation and relatively low symbol rates, since here we do not focus on the data rate or spectral efficiency. Key parameters used in the simulations are summarized in Table 2, where we mostly followed the setting in [27].

Table 2. Simulation parameters

View Table | View all tables in this article

To implement the proposed filter-enhanced positioning scheme, waveform shaping filters are required. In the simulation, we realized these filters in the frequency domain as FIR filters using the minimax algorithm [28], and one filter was applied to each LED transmitter. The detected wireless communication signal passed through the corresponding matched filters to separate signals from different LEDs for RSS and channel gain calculation. The positioning information was then estimated based on all RSS information and known locations of LED transmitters. In addition, we assumed that the major noise sources are the shot noise and the thermal noise, and we followed the noise model and parameters in [29]. The wireless communication BER was then calculated based on the simulated signal-to-noise-ratio (SNR) and the 4-PAM format.

The simulated positioning results are shown in Fig. 6(a), where both the room center and room corners were considered. The simulated received signal waveforms after the matched filters from the different transmitters at selected locations of (0, 0), (200 cm, 200 cm) and (350 cm, 350 cm) are shown in Fig. 6(b) – Fig. 6(d). The simulated BER performance of the wireless communication system is shown in Fig. 6(e). For comparison, we also simulated the positioning performance using the previously studied TDM method. In the simulations, four dedicated time slots were used, and each transmitter was allocated one time slot. The LEDs were then turned on alternatively during the allocated time slot to send the positioning signal. The receiver then monitored the corresponding RSS information for positioning information extraction. It can be seen that the average positioning accuracy achieved in the simulation using the proposed filter-enhanced method is about 6.15 cm, and better positioning accuracy is achieved when the receiver is closer to the room center, where a better BER performance is also achieved, and this is mainly due to the larger signal power detected. Compared with the TDM-based positioning method, almost identical positioning accuracy was achieved. However, no dedicated positioning signal was needed in the proposed filter-enhanced scheme. Note that, the wireless communication signals from different LED transmitters can always be separated using the proposed filter-enhanced scheme regardless of the location for positioning directly. Therefore, both wireless communication and positioning functions can be realized simultaneously.

Fig. 6. (a) Simulated positioning error with filter-enhanced VLC positioning scheme; (b)–(d) received signal waveforms after matched filters at (0, 0), (200 cm, 200 cm) and (350 cm, 350 cm), respectively; and (e) measured BER of the wireless communication function.

Download Full Size | PDF

In both the LED-based VLC positioning system simulated above and the laser-based OWC positioning system investigated in Section 3, the RC was applied in the wireless communication part, where the same data was sent by all transmitters. We used RC as it has been shown to be effective in OWC systems, especially in improving the signal coverage area [22]. The proposed filter-enhanced positioning scheme is also feasible when multiple transmitters send different data streams. With the spatial filter-enhanced transmitters, when different data streams are transmitted, due to the orthogonality of these spatial filters as described by Eq. (9), data streams from different transmitters can be separated at the receiver side after matched filters for separate detection. Therefore, simultaneous wireless communications can be realized. With the separated communication data streams from different transmitters, the corresponding average signal power and channel gain can be directly obtained following Eq. (11) and Eq. (12). Hence, the positioning function can still be achieved by extracting the RSS information.

In addition, centimeter-level positioning accuracy has been realized in this paper, which is consistent with other studies on indoor OWC positioning systems [15]. It is noted that instead of improving the positioning accuracy, this paper mainly focuses on exploring the wireless communication signals directly for indoor positioning. With the proposed filter-enhanced scheme, signals sent from different transmitters can simultaneously be separated at the receiver side after the matched filers. Hence, the RSS information from multiple transmitters can be obtained directly from the communication signals without the need to interrupt the wireless communication for positioning. Therefore, the system data rate or capacity is not affected.

5. Conclusions

In this paper, we have proposed a filter-enhanced indoor OWC system that is capable of simultaneous positioning and wireless communication functions. Spatial waveform shaping filters have been applied to multiple repetition-coded transmitters and the separation of signals from different transmitters simultaneously has been enabled. Based on the outputs of matched filters at the receiver side, the RSS information has been obtained and the positioning function has been realized. Results have shown that the proposed scheme eliminates the need of dedicated positioning time slots, RF carriers or codewords, and hence, the wireless communication data rate and throughput is not affected. In addition, since the positioning is estimated using the wireless communication signals directly, continuous positioning becomes possible for real-time user tracking.

The proposed filter-enhanced scheme for continuous and simultaneous positioning and wireless communication has been experimentally demonstrated in a near-infrared indoor OWC system with laser transmitters. An average positioning accuracy of 5.41 cm has been achieved, and 2.5 Gb/s wireless data communication has been realized simultaneously. The feasibility of proposed filter-enhanced scheme in VLC systems with LED transmitters has also been studied and demonstrated via simulations. Simultaneous wireless communication with 4-PAM format and positioning has been realized with three LED transmitters. Similar positioning accuracy has been achieved. Therefore, the proposed filter-enhanced scheme provides a promising positioning method in indoor OWC systems without affecting the wireless data communication.

Funding

Australian Research Council (DP170100268).

Disclosures

The authors declare no conflicts of interest.

References

1. F. Feng, P. Sangwongngam, G. Faulkner, and D. O’Brien, “Wide filed-of-view optical broadcasting for bi-directional indoor optical wireless communications employing PAM-4 modulation,” Opt. Lett. 44(24), 6009–6012 (2019). [CrossRef]

2. X. Zhang, C. Li, Y. Jiao, E. Tangdiongga, Y. Liu, Z. Cao, and T. Koonen, “Crosstalk-mitigated AWGR-based two-dimensional IR beam-steered indoor optical wireless communication system with a high spatial resolution,” J. Lightwave Technol. 37(15), 3713–3720 (2019). [CrossRef]

3. Y. Zhou, Y. Wei, F. Hu, J. Hu, Y. Zhao, J. Zhang, F. Jiang, and N. Chi, “Comparison of nonlinear equalizers for high-speed visible light communication utilizing silicon substrate phosphorescent white LED,” Opt. Express 28(2), 2302–2316 (2020). [CrossRef]

4. G. Cossu, M. Presi, R. Corsini, P. Choudhury, A. M. Khalid, and E. Ciaramella, “A visible light localization aided optical wireless system,” in Proc. IEEE Globecom Workshops (2011), pp. 802–807.

5. S. Dang, O. Amin, B. Shihada, and M.-S. Alouini, “What should 6G be?” Nat. Electron. 3(1), 20–29 (2020). [CrossRef]

6. E. C. Strinati, S. Barbarossa, J. L. G.- Jimenez, D. Ktenas, N. Cassiau, L. Maret, and C. Dehos, “6G: the next frontier: from holographic messaging to artificial intelligence using subterahertz and visible light communication,” IEEE Veh. Technol. Mag. 14(3), 42–50 (2019). [CrossRef]

7. T. Liang, K. Wang, C. Lim, E. Wong, T. Song, and A. Nirmalathas, “Secure multiple access for indoor optical wireless communications with time-slot coding and chaotic phase,” Opt. Express 25(18), 22046–22054 (2017). [CrossRef]

8. Z. Cao, X. Zhang, G. Osnabrugge, J. Li, I. M. Vellekoop, and A. M. J. Koonen, “Reconfigurable beam system for non-line-of-sight free-space optical communication,” Light: Sci. Appl. 8(1), 69 (2019). [CrossRef]

9. A. Gomez, K. Shi, C. Quintana, G. Faulkner, B. C. Thomsen, and D. O’Brien, “A 50 Gb/s transparent indoor optical wireless communication link with an integrated localization and tracking system,” J. Lightwave Technol. 34(10), 2510–2517 (2016). [CrossRef]

10. K. Wang, A. Nirmalathas, C. Lim, and E. Skafidas, “Experimental demonstration of a novel indoor optical wireless localization system for high-speed personal area networks,” Opt. Lett. 40(7), 1246–1249 (2015). [CrossRef]

11. K. Qiu, F. Zhang, and M. Liu, “Let the light guide us: VLC-based localization,” IEEE Robot. Autom. Mag. 23(4), 174–183 (2016). [CrossRef]

12. Y. Zhuang, L. Hua, L. Qi, J. Yang, P. Cao, Y. Cao, Y. Wu, J. Thompson, and H. Hass, “A survey of positioning system using visible LED lights,” IEEE Commun. Surv. Tutorials 20(3), 1963–1988 (2018). [CrossRef]

13. J. K. Part, J. Kim, and S. J. Kang, “A situation-aware indoor localization (SAIL) system using a LF and RF hybrid approach,” Sensors 18(11), 3864 (2018). [CrossRef]

14. M. I. AlHarjri, N. T. Ali, and R. M. Shubair, “Indoor localization for IoT using adaptive feature selection: a cascaded machine learning approach,” Antennas Wirel. Propag. Lett. 18(11), 2306–2310 (2019). [CrossRef]

15. J. Luo, L. Fan, and H. Li, “Indoor positioning systems based on visible light communication: state of the art,” IEEE Commun. Surv. Tutorials 19(4), 2871–2893 (2017). [CrossRef]

16. K. Wang, A. Nirmalathas, C. Lim, K. Alameh, and E. Skafidas, “Optical wireless-based indoor localization system employing a single-channel imaging receiver,” J. Lightwave Technol. 34(4), 1141–1149 (2016). [CrossRef]

17. H.-S. Kim, D.-R. Kim, S.-H. Yang, Y.-H. Son, and S.-K. Han, “An indoor visible light communication positioning system using a RF carrier allocation technique,” J. Lightwave Technol. 31(1), 134–144 (2013). [CrossRef]

18. B. Lin, X. Tang, Z. Ghassemlooy, C. Lin, and Y. Li, “Experimental demonstration of an indoor VLC positioning system based on OFDMA,” IEEE Photonics J. 9(2), 1–9 (2017). [CrossRef]

19. S. Yamaguchi, V. V. Mai, T. C. Thang, and A. T. Pham, “Design and performance evaluation of VLC indoor positioning system using optical orthogonal codes,” in Proc. IEEE Int. Conf. Commun., 54–59 (2014).

20. J. K. Park, T.-G. Woo, M. Kim, and J. T. Kim, “Hadamard matrix design for a low-cost indoor positioning system in visible light communication,” IEEE Photonics J. 9(2), 1–10 (2017). [CrossRef]

21. K. Wang, A. Nirmalathas, C. Lim, K. Alameh, H. Li, and E. Skafidas, “Indoor infrared optical wireless localization system with background light power estimation capability,” Opt. Express 25(19), 22923–22931 (2017). [CrossRef]

22. T. Song, A. Nirmalathas, C. Lim, E. Wong, K.-L. Lee, Y. Hong, K. Alameh, and K. Wang, “Performance analysis of repetition-coding and space-time-block-coding as transmitter diversity schemes for indoor optical wireless communications,” J. Lightwave Technol. 37(20), 5170–5177 (2019). [CrossRef]

23. J. M. Kahn and J. R. Barry, “Wireless infrared communications,” Proc. IEEE 85(2), 265–298 (1997). [CrossRef]

24. K. Wang, A. Nirmalathas, C. Lim, and E. Skafidas, “Impact of background light induced shot noise in high-speed full-duplex indoor optical wireless communication systems,” Opt. Express 19(22), 21321–21332 (2011). [CrossRef]

25. K. Wang, “Digital orthogonal-filters enhanced spatial modulation for high-speed indoor optical wireless communications,” J. Lightwave Technol. 37(24), 5988–5995 (2019). [CrossRef]

26. L. E. M. Matheus, A. B. Vieira, L. F. M. Vieira, M. A. M. Vieira, and O. Gnawali, “Visible light communication: concepts, applications and challenges,” IEEE Commun. Surv. Tutorials 21(4), 3204–3237 (2019). [CrossRef]

27. T. Komine, J. H. Lee, S. Haruyama, and M. Nakagawa, “Adaptive equalization system for visible light wireless communication utilizing multiple white LED lighting equipment,” IEEE Trans. Wireless Commun. 8(6), 2892–2900 (2009). [CrossRef]

28. G. Stepniak, “Comparison of efficiency of N-dimensional CAP modulations,” J. Lightwave Technol. 32(14), 2516–2523 (2014). [CrossRef]

29. Z. Xueli, D. Jingyuan, F. Yuegang, and S. Ancun, “Theoretical accuracy analysis of indoor visible light communication positioning system based on received signal strength indicator,” J. Lightwave Technol. 32(21), 4180–4186 (2014). [CrossRef]

Method	Advantages	Disadvantages
TDM [10,21]	Simple implementation	• Dedicated positioning time slots required • Wireless communications interrupted
RF carrier or subcarrier [17,18]	No complete interruption to wireless communications by using one RF carrier or subcarrier per transmitter	• Dedicated frequency resource for positioning required • Effective data rate and system capacity reduced
Optical orthogonal code [19,20]	Minimal interruption to wireless communications via dedicated codes for transmitters	• Long codes required • Effective data rate and system capacity reduced
Filter-enhanced (this work)	• No dedicated positioning resource required (directly explores wireless communication signals) • Data rate or system capacity not affected • Possible to achieve continuous positioning	• Spatial waveform shaping filters needed (added to transmitters)

Parameter	Value
LED radiation power	450 mW (4 LEDs)
Semi-angle at half power	60°
Receiver effective size	1 cm²
Field-of-view	90°
Responsivity	0.4 A/W
Concentrator gain	1

Method	Advantages	Disadvantages
TDM [10,21]	Simple implementation	• Dedicated positioning time slots required • Wireless communications interrupted
RF carrier or subcarrier [17,18]	No complete interruption to wireless communications by using one RF carrier or subcarrier per transmitter	• Dedicated frequency resource for positioning required • Effective data rate and system capacity reduced
Optical orthogonal code [19,20]	Minimal interruption to wireless communications via dedicated codes for transmitters	• Long codes required • Effective data rate and system capacity reduced
Filter-enhanced (this work)	• No dedicated positioning resource required (directly explores wireless communication signals) • Data rate or system capacity not affected • Possible to achieve continuous positioning	• Spatial waveform shaping filters needed (added to transmitters)

Parameter	Value
LED radiation power	450 mW (4 LEDs)
Semi-angle at half power	60°
Receiver effective size	1 cm²
Field-of-view	90°
Responsivity	0.4 A/W
Concentrator gain	1

Indoor optical wireless communication system with continuous and simultaneous positioning

Abstract

1. Introduction

2. Filter-enhanced indoor OWC system

2.1 RSS-based indoor OWC positioning system

2.2 Filter-enhanced RSS-based indoor OWC positioning system

3. Experiments and results

3.1 Experimental setup

3.2 Demonstration of simultaneous positioning and wireless communication

4. Filter-enhanced indoor positioning in VLC systems

4.1 Filter-enhanced RSS-based indoor VLC positioning system

4.2 Simulations and results

5. Conclusions

Funding

Disclosures

References

Cited By

Figures (6)

Tables (2)

Equations (16)

Optics Express