In this paper we experimentally investigate a gigabit indoor optical wireless communication system with single channel imaging receiver. It is shown that the use of single channel imaging receiver rejects most of the background light. This single channel imaging receiver is composed of an imaging lens and a small photo-sensitive area photodiode attached on a 2-axis actuator. The actuator and photodiode are placed on the focal plane of the lens to search for the focused light spot. The actuator is voice-coil based and it is low cost and commercially available. With this single channel imaging receiver, bit rate as high as 12.5 Gbps has been successfully demonstrated and the maximum error-free (BER<10−9) beam footprint is even larger than 1 m. Compared with our previous experimental results with a single wide field-of-view non-imaging receiver, an improvement in error-free beam footprint of >20% has been achieved. When this system is integrated with our recently proposed optical wireless based indoor localization system, both high speed wireless communication and mobility can be provided to users over the entire room. Furthermore, theoretical analysis has been carried out and the simulation results agree well with the experiments. In addition, since the rough location information of the user is available in our proposed system, instead of searching for the focused light spot over a large area on the focal plane of the lens, only a small possible area needs to be scanned. By further pre-setting a proper comparison threshold when searching for the focused light spot, the time needed for searching can be further reduced.
©2012 Optical Society of America
CorrectionsKe Wang, Ampalavanapillai Nirmalathas, Christina Lim, and Efstratios Skafidas, "High-speed indoor optical wireless communication system with single channel imaging receiver: erratum," Opt. Express 20, 25356-25356 (2012)
While optical access networks have been widely deployed to provide broadband connectivity effectively to the doorstep of the customer premises with a possibility of data rates up to 1 Gb/s [1–3], providing the high bandwidth connectivity around the personal work spaces remains a challenge. In addition, new services and consumer electronic platforms begin to demand more bandwidth to deliver entertainment and business such as the high-definition video streaming, the exchange of large data files between the users and high-performance servers and the 3D television. Consequently even higher speed is in high demand for the in-building or indoor personal area network (PAN). For the wired PAN, the use of multi-mode fiber (MMF) and plastic optical fiber (POF) has been intensely studied [4–6]. Bit rate as high as 7.6 Gbps has been achieved over 10 m 1 mm diameter POF with discrete multi-tone modulation. With 50 μm MM-fiber, more than 40 Gbps transmission has been demonstrated over 100 m with adaptive bit and power allocation. On the other hand, the wireless in-building or indoor PAN is even more attractive due to the added advantage of supporting mobility. Among multiple choices, millimeter wave (mm-wave) system and optical wireless technique have been considered as the most promising candidates since both of them are capable of bit rate higher than 1 Gbps [7–14].
Mm-wave based systems targeting at the 60 GHz frequency band have been advanced to the point of realizing full-duplex transceiver integrated on a single CMOS chip, capable of achieving 5Gbps wireless communication over a range of several meters without any interference with only ~0 dBm transmission power [7–9]. However, to achieve multi-gigabit-per-second operation in the mm-wave systems, advanced modulation formats and complex signal processing are required, leading to added complexity and cost . Furthermore, the free space propagation loss in this frequency region is high so the communication distance is still limited . In addition, mm-wave signals do not propagate well through obstructions such as doors and walls limiting their application to single room deployments. Radio-over-Fiber (ROF) is a potential solution, however, high-speed opto-electronic devices are required and the fiber dispersion limits system performance [15–18]. Finally, the license-free bandwidth available is only 7 GHz in the 60GHz region and this will ultimately limit the highest possible bit rate.
On the contrary, optical wireless technique has also been considered for high speed in-building or indoor PAN [11–14]. This is due to the huge unregulated bandwidth resource available in the optical region. With the use of digital baseband modulation in conjunction with direct detection can be used to transmit information wirelessly. Therefore, compared with the mm-wave system, the signal distribution in optical wireless system is easier and the existing optical fiber distribution network can be utilized. In addition, optical wireless system is also immune to the electro-magnetic interference, so it can be used in the RF-hostile environments such as inside hospitals. However, there are also several drawbacks associated with the optical wireless system, such as the limited transmission power due to laser eye and skin safety regulations  and the possible interruption during communication due to the physical shadowing problem .
In general, there are two major kinds of indoor optical wireless communication systems, namely the conventional line-of-sight (LOS) system and the diffused beam based system . In the first kind of system, a narrow laser beam is used to establish a point-to-point link between the transmitter and receiver. Since there is almost no multipath effect and the propagation loss is low, extreme high speed communication can be realized. However, to satisfy the strict alignment requirement, both transceivers need to be fixed so no mobility can be provided to users. In addition, the communication link may be frequently interrupted due to physical shadowing . Compared with the conventional LOS system, the diffused beam based system is capable of providing full mobility to end users and it is more robust to the shadowing problem since it uses the signal reflected back from walls, floor and ceiling. Nevertheless, multipath dispersion is severe and it results in inter-symbol interference (ISI) which limits the bit rate. Furthermore, to provide mobility over the entire room high transmission power (typically ~0.5-1 W) is needed . Although the signal is diffusive in nature and the power collected by human eye and skin may not so high, there is still some potential safety issues.
In previous studies, we proposed a novel indoor optical wireless system which combines the advantages of both direct LOS and diffused beam based systems [20–23]. In our proposed system, optical wireless technique is incorporated with the localization function and gigabit-per-second error-free transmission has been experimentally demonstrated with a reasonable beam footprint. Here we define the error-free as a bit-error-rate (BER) smaller than 10−9. Furthermore, in previous studies we just use the simplest single wide field-of-view (FOV) non-imaging receiver. However, this kind of receiver induces strong background light into the photodiode so the system performance is degraded .
To overcome this problem, we designed a novel single channel imaging receiver which consists of a simple lens, a small photo-sensitive area photodiode and a 2-axis actuator . As shown later, such a receiver is also simple in structure and low-cost. In this paper, we experimentally demonstrate that much better performance can be achieved with this kind of receiver. The results show that even when the bit rate is as high as 12.5 Gbps, the maximum error-free beam footprint is larger than 1 m. Compared with the previous results with non-imaging receiver , a remarkable improvement in error-free beam footprint of >20% has been achieved. When this optical wireless communication system with single channel imaging receiver is incorporated with the localization system we proposed recently [26, 27], error-free high speed mobile communication can be provided to the end user over the entire room. In addition, a theoretical model that allows the system performance to be analyzed has been proposed and the simulation results agree well with the experiments.
This paper is organized as follows: in Section 2 the basic architecture of the proposed system is briefly introduced; in Section 3 the structure of novel single channel imaging receiver is described and the system performance is theoretically studied and simulated; in Section 4 proof-of-concept experimental setup is described and experimental results are presented and discussed; and finally the conclusion is given in Section 5.
2. System architecture
The architecture of our proposed system is shown in Fig. 1 . The central office (CO)/access points (APs) architecture is used, which is widely used in the ROF systems . Such architecture can be easily integrated with the existing backbone networks. Furthermore, several APs can be served by the same CO via an optical fiber distribution network and we propose that all the complex control functions and expensive hardware are located inside the CO. Therefore, the cost can be shared by multiple rooms and multiple users .
The room considered here is a real office scenario. It is a 6 m × 4 m × 3 m room with two equal sized rectangular cubicles. We assume that all the cubicle partitions are opaque so the incident signal will be either absorbed or blocked. In addition, the office is well illuminated with four 100 W tungsten lamps.
Inside the room, optical wireless technique is incorporated with the localization function. The localization system is also based on optical wireless so a separate system (e.g. the WiFi based system or the Zigbee based system) is not needed. The details of the localization system are described in . Furthermore, instead of using one laser inside each room , the ceiling mounted fiber based device serves as the transmitter in our proposed system. The fiber transmitter is simple structured and it is just installed above the intersection of the two adjacent cubicles to overcome the possible physical shadowing problem. The ceiling mounted fiber transmitter is composed of a fiber end connected to the CO, a lens attached in front of the fiber end to increase the divergence of the signal beam for limited mobility purposes, and a steering mirror to change the orientation of the signal beam. With the rough localization information of the user, a comparatively large beam is used to cover the user’s position as well as the surrounding areas. When the user moves inside the area covered by the signal beam, high speed direct LOS link is always available. In case of the user moves out of the area, which can be identified by the localization system, the mirror in the fiber transmitter will then be steered adaptively to the new position. Through this way, high speed wireless communication as well as mobility can be provided over the entire room .
At the receiver end, in the previous studies we just used the simplest single wide FOV non-imaging receiver . This receiver is based on a compound parabolic concentrator (CPC) with a FOV of 45°. With such a receiver, all background light incident onto the CPC with an angle smaller than the FOV will be collected and detected by a large sensitive-area photodiode. Therefore the background light induced shot noise is relatively high and this will result in the degradation of receiver sensitivity. We have shown both with theoretical analysis and experiments that the resulted power penalty is larger than 3.5 dB for the 1 Gbps system . Even for the 12.5 Gbps system, the induced power penalty is still about 0.5dB. To reduce the collected background light power, we have designed a novel single channel imaging receiver  and it will be discussed in detail in the next section.
3. Single channel imaging receiver and system simulation
3.1 Single channel imaging receiver structure and searching algorithm
A simplified schematic of our proposed single channel imaging receiver is depicted in the inset of Fig. 1. This receiver is composed of an imaging lens, a small photo-sensitive area photodiode and a 2-axis actuator. The imaging lens is used as an optical antenna to collect stronger signal light. Since we only use the signal coming from the direct LOS channel and the beam footprint at the receiver end is much larger than the aperture of the lens, the signal light incident onto the lens has almost the same incident angle and all the signal light can be focused onto a small spot on the focal plane of the lens. To detect the concentrated signal light, a small photo-sensitive area photodiode placed at the focal plane of the lens is utilized. The photodiode is attached on a two-axis actuator to search for the focused spot. The actuator is realized by using the voice coil. By controlling the electrical current passing through the voice coil, the magnetic field generated will be changing and the actuator arm can move in or out. This kind of actuator (one-axis) is widely used in the computer hard disk and CD read-head. Consequently it is commercially available and potentially low-cost. Moreover, the voice coil based actuator is capable of both high-speed and precise tuning in the order of tens of nm. In addition, in our experiment the photodiode used has a dimension of 65 μm × 65 μm. Since the focused spot of the signal light is relatively small (smaller than several μm), we chose the scanning step to be 60μm to achieve high speed searching and avoid the possibility that only a fraction of the focused spot falls onto the photodiode.
The actuator in the single channel imaging receiver allows the photodiode to move on the focal plane of the lens and search for the focused signal spot. Since a small photo-sensitive area photodiode is utilized, only the background light with almost the same incident angle as the signal can be collected and detected. Therefore, the background light induced noise can be greatly reduced and better system performance should be expected.
To search for the focused spot, the optical power incident onto the photodiode should be monitored. Since the output photo-current is proportional to the received optical power and the photodiode used in the experiment is integrated with a trans-impendence amplifier (TIA), the output voltage of the TIA is monitored and compared to find the maximum. In our system, the rough location information of the user is available from the localization system, so the position of the focused spot on the focal plane can be estimated. In consequence, instead of moving the photodiode over a large area searching for the focused signal spot, only a small portion of the focal plane needs to be scanned. This can greatly reduce the time needed for comparing the output voltage and searching for the focused spot and the effective throughput of the system can be much higher.
In addition, in considerable cases when the received signal power is relatively high (the user is closer to the beam center on the reception plane), finding the maximum output voltage of the TIA may be not necessary. Therefore we can pre-define an output voltage threshold Vth which assures error-free operation and when the output voltage of the TIA exceeds this threshold, the comparing and searching process can be stopped. The value of Vth is chosen according to the parameters of the photodiode and TIA used and the received background light estimation in the room. First of all, estimate the maximum received background light power and add it to the receiver sensitivity to set a threshold of the input optical power of the photodiode. Then use the photodiode responsitivity and the trans-conductance gain of the TIA to calculate a proper Vth. In this way the searching time can be further reduced.
The entire process of the comparing and searching algorithm is as follows:
- 1. With the rough localization information of the user, estimate the possible position of the focused spot of signal light on the focal plane. Then move the photodiode to one corner of the possible area;
- 2. Monitor the output voltage from the TIA and compare it to the pre-defined threshold Vth. If the output voltage is larger than Vth, which means a satisfying performance can be achieved, stop the compare process and go to step 5. On the contrary, if the output is smaller than Vth, then store the voltage from TIA and the position on the focal plane;
- 3. Move the photodiode to the next possible position and the scanning step is 60μm as mentioned before. Repeat step 2 and compare the output voltage with Vth. If the output is smaller than Vth, then compare it with the stored previous TIA output value and store the larger one. The position associated with the larger voltage should also be stored;
- 4. Repeat steps 3 and 2 and scan the entire possible area on the focal plane if needed. Then move the photodiode to the position where the maximum TIA output voltage is obtained;
- 5. Start high speed optical wireless communication and continuously monitor the output voltage from the TIA. If a sharp drop in the output is observed, which means the focused spot of signal light is no longer captured by the photodiode, start the searching process again.
With this comparing and searching process, the position of focused spot of signal light can always be found and high speed optical wireless mobile communication can be achieved.
3.2 Theoretical analysis and system simulation
To study the performance of the system, we considered a real office scenario and the configuration of the room is shown in Fig. 2 . It is a 6 m × 4 m × 3 m room with two 3 m × 3 m × 3 m rectangular cubicles. The fiber transmitter is located just above the intersection of two cubicles and in the coordinates shown in the figure, its position is (xt, yt, zt) = (3, 1, 3). To create a well illuminated environment, four strong background lamps are considered and they can be modeled as Lambertain sources with a radiant intensity (W/Sr) given by28].
In the system with single channel imaging receiver, only the background light that is also focused onto the small photo-sensitive area photodiode will introduce additional noise. Suppose the position of the user is (x, y, z), the distance df between the focused spot of signal light and the center of the focal plane isFig. 3(a) . The results when the non-imaging receiver is used are also shown in Fig. 3(b). Here we suppose that both kinds of receivers have the same entrance aperture. From the figures it is clear that for the system with single channel imaging receiver, the received background power is always lower than 3 μW and for most parts of the room is it lower than 1 μW. On the other hand, for the system with non-imaging receiver, the received background light power is higher than 4 μW in most of the areas and even as large as ~8 μW in the positions directly under lamps. Therefore, the received background light power can be greatly reduced in the system with single channel imaging receiver. Furthermore, for those positions directly under lamps, the background power is higher than the surrounding area for both kinds of systems.
The field distribution of the signal light exiting the ceiling mounted fiber end is approximately Gaussian and the beam waist is ~5.6 μm . After passing through the attached lens and free space propagation, although the divergence of the beam has increased, the field distribution remains Gaussian. In addition, because the beam footprint at the reception plane is much larger than the aperture of the lens, the optical field can be seen as evenly distributed over the entire aperture.
Signal-to-noise ratio (SNR) and BER are significant parameters for system performance indication. In our proposed system, the simplest and most mature on-off-keying (OOK) modulation format is used, so they are defined as 12, 29]. Therefore σ02 and σ12 are given by
The preamplifier used in our system is a p-i-n FET transimpedance receiver proposed in . The principle noise sources in this preamplifier are Johnson noise associated with the FET channel conductance, Johnson noise from the load or feedback resistor, shot noise arising from gate leakage current and 1/f noise. The preamplifier shot noise variance is given by
Furthermore, the background light induced shot noise can be calculated byFig. 3.
Using Eqs. (1)–(8), we simulated the SNR and BER performances over the entire room and the results are shown in Fig. 4 . In the simulation we fix the transmission power at 3 mW at the output plane of the fiber transmitter, the beam footprint at 2 m at the user end and the transmission bit rate is 1 Gbps. In order to achieve error-free operation over the entire room, the FOV of the imaging lens is chosen to be 60°. The results shown that over the entire room a SNR>14 dB or a BER<1.2 × 10−8 can always be achieved and in most parts of the room except those directly under lamps and at room corners, a SNR>16 dB can be achieved. This is consistent with the noise analysis. Compared with the performances of our previously proposed system with non-imaging receiver , much better performances have been achieved. In addition, to achieve error-free operation (BER<10−9) over the entire room, we have also undertaken simulations to monitor the worst BER inside the room for different transmission powers when the bit rate is fixed at 1 Gbps. The result is shown in Fig. 5 and it is clear that a transmission power of only 3.24 mW is sufficient.
4. Experiments and discussions
In addition to theoretical analysis and simulations, we have also carried out proof-of-concept experiment and the experimental setup is shown in Fig. 6 . An optical carrier is generated and modulated in the CO. The modulated signal is transmitted to the room via fiber distribution network and in the experiment it is emulated by the 5.6 km standard single-mode fiber. Then inside the room the signal light exiting the fiber goes through a lens (10.0 mm in the focal length and 6.0 mm in the diameter, coated in the 1550 nm band) to increase its divergence before propagating in the free space and at the receiver side it is captured by an imaging lens. The photodiode used in the experiment is integrated with TIA on a printed circuit board (PCB). The PCB is placed at the focal plane of the lens and it is mounted on two 1-axis motion controllers (Newport MM2500). The two 1-axis motion controllers are used to emulate the 2-axis actuator proposed in Fig. 1 due to device limitations in our lab. To search for the focused signal spot, we use a FPGA to monitor the output voltage of the TIA and implement the proposed searching algorithm. Since the speed of built-in analog-to-digital converter (ADC) is much lower than the communication bit rate, the average received optical power is monitored. The FPGA also compares the output voltages of the TIA at different positions and stores the maximum value together with the associated position of the photodiode. If further scanning is required or the focused spot has been found, then the FPGA communicates with a computer via the serial port (RS-232). The LabView software is used to capture the output from FPGA and control the motion controllers. Ideally the computer with LabView software is not needed. However, the motion controller used here can only be controlled by LabView so an extra computer is used. Here the output of TIA, the FPGA, the computer with LabView and the motion controllers comprise a close-loop control system. When the focused light spot has been found, high speed OW communication is now ready to start. The detected signal is then amplified and characterized with a bit-error-rate tester (BERT) and a wideband digital communication analyzer (DCA).
In the experiment the bit rate is 10 Gbps and the transmission power at the output plane of the fiber is fixed at 7 mW, the maximum possible power due to laser eye and skin safety regulations . Furthermore, we fix the beam footprint to be 1.2 m at the receiver end and strong background light from direct overhead lamps is also incorporated. Although in real applications both the background lamps and fiber transmitter are installed in the ceiling, for proof-of-concept purpose the setup in the experiment is still feasible. Furthermore, in real situations, when the user moves inside the room, there is a varying distance from beam center at the reception plane. Therefore, in the experiment we systematically characterize the system performance as a function of the distance from beam center and this is emulated by titling the transmitter end in the experiment and the results are shown in Fig. 7 . When the BER is smaller than 10−12, the exact value is not measured since it can be seen as error-free. In the figure, the green line shows the results when the entire possible area on the focal plane is scanned for finding the maximum output voltage of the TIA (complete searching) and the red line shows the results when the searching algorithm proposed in section 3 is used. The photodiode used has a dimension of 65 μm × 65 μm and we chose a moving step of 60 μm to avoid the situation that the focused signal spot falls onto the edge of the photodiode and only a fraction of the signal can be detected. In addition, the average received background light power is measured to be ~-29.7 dBm and from the theoretical analysis stated in the previous section, we choose the threshold input optical power to be −17dBm which is enough for error-free operation.
From the figure it is clear that when the distance from beam center increases, the BER increases as well. This is simply due to the fact the received signal power is lower when the distance is larger while the noise almost remains constant. Furthermore, the red line indicates that our proposed searching algorithm is feasible and the focused signal spot can always be found. In addition, compared with the results obtained from complete searching, at the some distances from beam center the BER from the proposed searching algorithm is even better. This is because there is some overlapping area between two adjacent searching positions of the photodiode on the focal plane and when the focused signal spot falls into this overlapping area, the searching algorithm may choose any one of the possible positions. In some cases the position with smaller collected optical power is chosen, which means less background light power is detected while the signal power remains almost the same. Therefore a better BER can be achieved. However, when the distance from beam center is larger than 50cm, the two searching methods show the same BER performance. This is because in these cases, the collected optical powers are always lower than the −17 dBm threshold and the proposed algorithm is equivalent to the complete searching method.
The BER results with the single wide FOV non-imaging receiver are also shown in Fig. 7 (black line, with the same photodiode as in the single channel imaging receiver) . It is obvious that much better performance can be achieved with the single channel imaging receiver proposed here. The reason for this is that most of the background light can be rejected with the imaging receiver (for the system with non-imaging receiver, the collected background light power is ~-25.4 dBm).
In addition to the 10 Gps system, we have also undertaken the experiment when the bit rate is increased to 12.5 Gbps. The transmission power is still 7 mW due to safety issues but the beam footprint at the receiver end is decreased to 1 m. The results are shown in Fig. 8 and the BER of the system with non-imaging receiver is also shown. The BER also increases with increasing distance from the beam center, similar to that shown in Fig. 7. Furthermore, the previously proposed searching algorithm is still feasible as well and the system with single channel imaging receiver has much better performance than that with non-imaging receiver. In addition, it is clear that even at beam boundaries, for the system with single channel imaging receiver, error-free operation (BER<10−9) can still be achieved. When it is incorporated with the indoor localization system proposed before , error-free high speed mobile communication can be achieved over the entire room.
To find out the maximum error-free beam footprint, we have also carried out experiment to monitor the BER at beam boundaries for different beam footprints. We only measure the BER there because it is the highest over the entire beam footprint. The results for the 10 Gbps and 12.5 Gbps systems with both single channel imaging receiver and non-imaging receiver are shown in Fig. 9 . It is clear that the maximum error-free beam footprints for the 10 Gbps and 12.5 Gbps systems are ~111.6 cm and ~101 cm respectively when the single channel imaging receiver is utilized. For the systems with non-imaging receiver, the maximum error-free beam footprints are ~89 cm and ~84 cm, respectively . Therefore an improvement in beam footprint of larger than 20% has been achieved.
We have also undertaken experiments to study the maximum error-free beam footprint when the transmission bit rates are 1 Gbps, 2.5 Gbps and 5 Gbps, respectively. The results are summarized in Fig. 10(a) . In some cases, error-correction code (ECC) is used so the error-free threshold can be set at BER < 10−6. The maximum beam footprints when ECC is used are also shown in the figure. It can be seen that when the bit rate is 1 Gbps, a beam footprint larger than 2.8 m is feasible even without ECC. Furthermore, compared with the non-imaging receiver, the system with the single channel imaging receiver proposed in this paper can achieve considerable improvement in error-free beam footprint and the improvement in percentage is summarized in Fig. 10(b). It is clear that the improvement decreases with the increasing bit rate. This is because for higher speed system, the impact of background light induced noise is not so obvious  and the pre-amplifier induced noise is the dominant one.
Finally, to validate the theoretical model proposed in Section 3.2, we have carried out experiment to measure the BER at beam boundaries for different transmission powers when the bit rate is fixed at 1 Gbps and the beam footprint is fixed at 2 m. The results are shown in Fig. 11 . It can be seen that the experimental results agree well with the simulation results shown in Fig. 5. Furthermore, the minimum transmission power to achieve error-free operation in the experiment is ~3.57 mW, ~0.4 mW higher than the simulated result. This is mainly due to the unavoidable coupling losses in the receiver end in the experiment.
In this paper we proposed the use of a novel single channel imaging receiver in indoor optical wireless communication system. This single channel imaging receiver comprises of a small photo-sensitive area photodiode, a 2-axis voice coil based actuator and an imaging lens. The photodiode is placed on the focal plane of the lens and attached on the actuator. The actuator allows the photodiode to move on the focal plane and search for the focused signal spot. Since the photodiode is small, only the background light with almost the same incident direction will be detected. Therefore the impact of background light can be greatly reduced and better performance can be achieved.
We have carried out theoretical analysis and simulation and a searching algorithm has been proposed. It is shown that the received background light power is much lower than that with the previously used single wide FOV non-imaging receiver. Furthermore, simulation results show that when the bit rate is 1 Gbps and the beam footprint is 2 m, only 3.16 mW transmission power is needed for error-free operation.
In addition, proof-of-concept experiment has also been undertaken. The 2-axis voice coil based actuator in single channel imaging receiver is emulated by using two 1-axis motion controllers which are controlled by LabView. To monitor the output voltage from TIA and search for the focused signal spot, a FPGA with built-in ADC is used. Furthermore, the experimental results show a considerable improvement in BER performance when compared with the system with non-imaging receiver and up to 12.5 Gbps error-free operation has been demonstrated. In addition, we have examined the maximum error-free beam footprint for different bit rates when the transmission power is fixed at 7 mW. It is shown that an improvement of larger than 20% can be achieved. When this system is incorporated with our previously proposed indoor localization system, high speed error-free mobile communication can be provided over the entire room.
This work was supported in part by NICTA. NICTA is funded by the Australian Government as represented by the Department of Broadband, Communications and the Digital Economy and the Australian Research Council through the ICT centre of Excellence program.
References and links
1. P. Heron, “Next generation optical access networks,” in Proceedings of Advanced Photonics Congress – Access Networks and In-House Communications (Toronto, Canada, 2011), AMA2.
2. M. Daneshmand, C. Wang, and W. Wei, “Advances in passive optical networks,” IEEE Commun. Mag. 49(2), s12–s14 (2011). [CrossRef]
4. D. Visani, C. Okonkwo, S. Loquai, H. Yang, S. Yan, H. van den Boom, T. Ditewig, G. Tartarini, J. Lee, T. Koonen, and E. Tangdiongga, “Beyond 1 Gbit/s transmission over 1mm diameter plastic optical fiber employing DMT for in-house communication systems,” J. Lightwave Technol. 29, 622–628 (2011).
5. R. E. Freund, C. A. Bunge, N. N. Ledentsov, D. Molin, and C. Caspar, “High-speed transmission in multi-mode fibers,” J. Lightwave Technol. 28(4), 569–586 (2010). [CrossRef]
6. I. Mollers, D. Jager, R. Gaudino, A. Nocivelli, H. Kragl, O. Ziemann, N. Weber, T. Koonen, C. Lezzi, A. Bluschke, and S. Randel, “Plastic optical fiber technology for reliable home networking: overview and results of the EU project pof-all,” IEEE Commun. Mag. 47(8), 58–68 (2009). [CrossRef]
7. C. Liu, E. Skafidas, and R. Evans, “Capacity and data rate for millimeter wavelength systems in a short range package radio transceiver,” IEEE Trans. Wirel. Comm. 9(3), 903–906 (2010). [CrossRef]
8. E. Skafidas, F. Zhang, B. Yang, B. N. Wicks, Z. Liu, C. M. Ta, Y. Mo, K. Wang, G. Felic, P. Nadagouda, and T. Walsh, W. Shiehm I. Mareels, and R. Evans, “A 60GHz transceiver on CMOS,” in Proceedings of International Topical Meeting on Microwave Photonics (Sydney, Australia, 2008), 306–309.
9. B. Yang, Y. Mo, K. Wang, Y. Feng, B. Wicks, C. M. Ta, F. Zhang, Z. Liu, C. Liu, G. Felic, P. Nadagouda, T. Walsh, and E. Skafidas, “Implementation of a gigabit per second millimeter wave transceiver on CMOS,” in Proceedings of 2nd International Conference on Wireless Broadband and Ultra Wideband Communications (Sydney, Australia, 2007), 71–74.
10. J. Lee, Y. Chen, and Y. Huang, “A low-power low-cost fully-integrated 60-GHz transceiver system with OOK modulation and on-board antenna assembly,” IEEE J. Solid-state Circuits 45(2), 264–275 (2010). [CrossRef]
11. F. R. Gfeller and U. Bapst, “Wireless in-house data communication via diffuse infrared radiation,” Proc. IEEE 67(11), 1474–1486 (1979). [CrossRef]
12. J. M. Kahn and J. R. Barry, “Wireless infrared communications,” Proc. IEEE 85(2), 265–298 (1997). [CrossRef]
13. J. R. Barry, J. M. Kahn, W. J. Krause, E. A. Lee, and D. G. Messerschmitt, “Simulation of multipath impulse response for indoor wireless optical channels,” IEEE J. Sel. Areas Comm. 11(3), 367–379 (1993). [CrossRef]
14. M. Alresheedi and J. M. H. Elmirghani, “Performance evaluation of 5 Gbit/s and 10 Gbit/s mobile optical wireless systems employing beam angle and power adaptation with diversity receivers,” IEEE J. Sel. Commun. 29(6), 1328–1340 (2011). [CrossRef]
15. A. Islam, M. Bakaul, A. Nirmalathas, and G. E. Town, “Millimeter-wave radio-over-fiber system based on heterodyned unlocked light sources and self-homodyne RF receiver,” IEEE Photon. Technol. Lett. 23(8), 459–461 (2011). [CrossRef]
16. C. Lim, A. Nirmalathas, M. Bakaul, P. Gamage, Y. Ka-Lun Lee, D. Yizhuo Yang, Novak, and R. Waterhouse, “Fiber-wireless networks and subsystem technologies,” J. Lightwave Technol. 28(4), 390–405 (2010). [CrossRef]
17. Y.-T. Hsueh, H.-C. Chien, A. Chowdhury, J. Yu, and G.-K. Chang, “Performance assessment of radio links using millimeter-wave over fiber technology with carrier suppression through modulation index enhancement,” J. Opt. Commun. Netw. 3(3), 254–258 (2011). [CrossRef]
18. C. H. Yeh and C. W. Chow, “Heterogeneous radio-over-fiber passive access network architecture to mitigate Rayleigh backscattering interferometric beat noise,” Opt. Express 19(7), 5735–5740 (2011). [CrossRef] [PubMed]
19. AS/NZS 2211.1:2004, Safety of Laser Products (Standards Australia International Ltd and Standards New Zealand, 2004).
20. K. Wang, A. Nirmalathas, C. Lim, and E. Skafidas, “High-speed duplex optical wireless communication system for indoor personal area networks,” Opt. Express 18(24), 25199–25216 (2010). [CrossRef] [PubMed]
21. K. Wang, A. Nirmalathas, C. Lim, and E. Skafidas, “Indoor gigabit optical wireless communication system for personal area networks”, in Proceedings of 23rd IEEE Photonics Society Annual Meeting (Denver, 2010), 224–225.
22. K. Wang, A. Nirmalathas, C. Lim, and E. Skafidas, “Gigabit optical wireless communication system for indoor applications”, in Proceedings of Asia Communications and Photonics Conference and Exhibition (Shanghai, China, 2010), 453–454.
23. K. Wang, A. Nirmalathas, C. Lim, and E. Skafidas, “High-speed optical wireless communication system for indoor applications,” IEEE Photon. Technol. Lett. 23(8), 519–521 (2011). [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] [PubMed]
25. K. Wang, A. Nirmalathas, C. Lim, and E. Skafidas, “12.5Gbps indoor optical wireless communication system with single channel imaging receiver,” in Proceedings of European Conference on Optical Communication (Geneva, Switzerland, 2011), Tu.3.C.6.
26. K. Wang, A. Nirmalathas, C. Lim, and E. Skafidas, “High-speed 4 × 12.5Gbps WDM optical wireless communication systems for indoor applications,” in Proceedings of Optical Fiber Communication Conference and Exposition and the National Fiber Optic Engineers Conference (Los Angeles, 2011), JWA081.
27. K. Wang, A. Nirmalathas, C. Lim, and E. Skafidas, “4 × 12.5Gbps WDM optical wireless communication system for indoor applications,” J. Lightwave Technol. 29(13), 1988–1996 (2011). [CrossRef]
28. J. B. Carruthers, “Multipath channels in wireless infrared communications: modeling, angle diversity and estimation,” Ph.D. dissertation, Univ. of California, Berkeley (1997).
29. B. Leskovar, “Optical receivers for wide band data transmission systems,” IEEE Trans. Nucl. Sci. 36(1), 787–793 (1989). [CrossRef]