Optical Access Network

Introduction:
The access network is the last leg of the telecommunication network that runs from the service providers facility to the home or business. With fiber now directly available to many office buildings in metropolitan areas, network based on SONET/SDH (Synchronous optical networking and Synchronous Digital Hierarchy, are two closely related multiplexing protocols for transferring multiple digital bit streams using lasers or light-emitting diodes over the same optical fiber) or Ethernet based technologies are being used to provide high speed access to large business users. While this is happening, the telephone and cable companies are also placing a significant emphasis on the development of networks that will allow them to provide a variety of services to individual homes and small to medium business. According to Ramaswami and Sivranjan (2004) early efforts on developing high capacity access network were devoted to developing networks that would accommodate various forms of video such as video on demand and high definition televisions. However the range of services that users are expected to demand in the future is vast and unpredictable. Today, end users are interested in both Internet access and high speed data access services, for such application as telecommuting, distance learning, video conferencing etc. Future, unforeseen applications are also sure to arise and make ever increasing demands on the bandwidth available in the last mile. At a broad level these services can be classified based on three major criteria, the first is the bandwidth requirement. The second is whether this requirement is symmetric (two way), for example, videoconferencing or asymmetric (one way), for example, broadcast video. The last criterion is whether the service is inherently broadcast, where every user gets the same information, for example, broadcast video, or whether the service is switched, where different users get different information, for example, internet access.

Overview of PON: Passive Optical Network

With the expansion of services offered over the Internet, the last mile bottleneck problems continue to worsen. A Passive Optical Network (PON) is a technology viewed by many as an attractive solution to this problem. A PON which has an excellent performance-to-cost ratio the architecture uses the time-division multiplexing (TDM) approach to deliver data encapsulated in Ethernet packets from a collection of Optical Network Units (ONUs) to a central Optical Line Terminal (OLT) over the PON access network. The OLT, in turn, is connected to the rest of the Internet. Passive Optical Networks (PON) is point-to-multipoint optical networks with no active elements in the signals path from source to destination. The only interior elements used in such networks are passive combiners, couplers, and splitters to enable a single optical fiber to serve multiple premises, typically 32-128. In 1999 Lung referred as a PON configuration reduces the amount of fiber and central office equipment required compared with point to point architectures. Downstream signals are broadcast to each premise sharing a fiber. Upstream signals are combined using a multiple access protocol, invariably time division multiple access (TDMA).

Page | 1

Figure 1: Passive Optical network. According to the research of Pesavento and Kelsey (1999) PON technology is getting more and more attention by the telecommunication industry as the last mile solution. Advantages of using a PON for local access networks are numerous: A PON allows for longer distances between central offices and customer premises. While with the Digital Subscriber Line (DSL) the maximum distance between the central office and the customer is only 18000 feet (approximately 5.5 km), a PON local loop can operate at distances of over 20 km. A PON minimizes fiber deployment in both the local exchange and the local loop. A PON provides higher bandwidth due to deeper fiber penetration. While the fiber-to the-building (FTTB), fiber-to-the-home (FTTH), or even fiber-to-the-PC (FTTPC) solutions have the ultimate goal of fiber reaching all the way to customer premises, fiber-to-the-curb (FTTC) may be the most economical deployment today. As a point-to-multipoint network, a PON allows for downstream video broadcasting. A PON eliminates the necessity of installing multiplexers and demultiplexers in the splitting locations, thus relieving network operators from the dreadful task of maintaining them and providing power to them. Instead of active devices in these locations, a PON has passive components that can be buried into the ground at the time of deployment. A PON allows easy upgrades to higher bit rates or additional wavelengths.

The clear advantages of using PON technology in access networks dictate that we make important design decisions in implementing it. Because an access network aggregates traffic from a relatively small number of subscribers (compared to metro or regional networks), it is very cost sensitive. Therefore, a PON design should not require over-provisioning and should allow for incremental deployment.

Page | 2

Standards:
ITU-T G.983 APON (ATM Passive Optical Network). This was the first Passive optical network standard. It was used primarily for business applications, and was based on ATM. BPON (Broadband PON) is a standard based on APON. It adds support for WDM, dynamic and higher upstream bandwidth allocation, and survivability. IEEE 802.3ah EPON or GEPON (Ethernet PON) is an IEEE/EFM standard for using Ethernet for packet data. 802.3ah is now part of the IEEE 802.3 standard. There are currently over 15 million installed EPON ports. With China's 2008 EPON deployments total installed base is expected to reach nearly 20 million subscribers by year end 2008. ITU-T G.984 GPON (Gigabit PON) is an evolution of the BPON standard. It supports higher rates, enhanced security, and choice of Layer 2 protocol (ATM, GEM, and Ethernet). In early 2008, Verizon began installing GPON equipment, having installed over 800 thousand lines by midyear. British Telecom and AT&T are in advanced trials. IEEE P802.3av 10G-EPON (10 Gigabit Ethernet PON) is an IEEE Task Force for 10Gbit/s, backward compatible with 802.3ah EPON. 10GigEPON will use separate wavelengths for 10G and 1G downstream. 802.3av will continue to use a single wavelength for both 10G and 1G upstream with TDMA separation. It will also be WDM-PON compatible (depending on the definition of WDM-PON). It is capable of using multiple wavelengths in both directions. SCTE IPS910 RFoG (RF over Glass) is an SCTE Interface Practices Subcommittee standard in development for Point to Multipoint (P2MP) operations that MAY have a wavelength plan compatible with data PON solutions such as EPON,GEPON or 10GigEPON.

EPON: Ethernet Passive Optical Network

Kramer, Mukherjee and Pesavento (2002) stated Ethernet for subscriber access networks also referred to as Ethernet in the First Mile (EFM), combines a minimal set of extensions to the IEEE 802.3 Media Access Control (MAC) and MAC Control sub-layers with a family of Physical (PHY) layers. EFM also introduces the concept of Ethernet Passive Optical Networks (EPONs), in which a point to multipoint (P2MP) network topology is implemented with passive optical splitters, and optical fiber PMDs, that support this topology. EPON is based upon a mechanism named MPCP (Multi-Point Control Protocol), which uses messages, state machines, and timers, to control access to a P2MP topology. At the basis of the EPON/MPCP protocol lies the P2P Emulation Sublayer, which makes an underlying P2MP network appear as a collection of point to point links to the higher protocol layers (at and above the MAC Client). It achieves this by attaching a Logical Link Identification (LLID) to the beginning of each packet, replacing two octets of the preamble. In addition, a mechanism for network Operations, Administration and Maintenance (OAM) is included to facilitate network operation and troubleshooting. EPON supports a broad range of EFM applications: Fiber to the Home (FTTH), Fiber to the Building (FTTB), Fiber to the MDU, MTU, Fiber to the Curb (FTTC).

Page | 3

GPON: Gigabit PON

According to Ramaswami et al. (2004) the case for Gigabit Passive Optical Network (GPON) is a simple one: in order to remain competitive, service providers need more capacity to carry bandwidth-intensive applications; fiber is the ultimate carrier of bandwidth, and GPON is one of the most cost-effective ways for a provider to deploy fiber. GPON operate at bitrates of above 1 Gb/s. Digital Subscriber Line (DSL) is currently the most well-known technology for access networks, providing data rates of up to 50 Mbit/s. DSL uses the telephone networks existing copper twisted pairs. However, future services such as video on-demand, video conferences, triple play, Voice over IP, digital TV and HDTV require access networks with higher data rates. Todays GPON access networks use data rates of 2.5 Gbit/s from network to subscriber (downstream) and 1.25 Gbit/s from subscriber to network (upstream), this bandwidth is shared by 32 users. A central device allows subscribers within a radius up to 20 kilometres to be connected. In order to reach subscribers living further away, an additional system is needed.

EPON vs. GPON: A Practical Comparison

Increasingly competing with copper as the infrastructure for access networks, fiber is making rapid headway in the world's leading technology-adopter markets. With PON technology gaining popularity, two point-to-multipoint standards EPON and ATM-based broadband passive optical networking (BPON) are both in active deployment. Meanwhile, industry observers view ATM-based GPON as the eventual successor to BPON. GPON defines a completely new protocol designed to support multiple services in their native formats. As the debate over EPON and GPON runs deep, this section will provide a practical comparison of the two technologies. Haran (2005) found some of the key differences between the two technologies and examines the strengths of each protocol.
BPON Standards Downstream Speeds Upstream Speeds Downstream Wavelength Upstream Wavelength Layer 2 Support Voice Support Video Support QoS Max. PON Splits Distance Avg. Bandwidth per user ITU G.983 155 Mbps, 622 Mbps (now), 1.2Gbps (upcoming) 155 Mbps or 622 Mbps 1480 -1500 nm 1260 1360 nm ATM TDM over ATM RF overlay (over 1550 nm) and/or IP video Ethernet (VLAN) 32 <20km (~12 miles) 19 Mbps/5Mbps EPON IEEE 802.3ah 1.25 Gbps 1.25 Gbps 1500 nm 1310 nm Ethernet TDM over Packet IP video Ethernet (VLAN) 16 <20km (~12 miles) 24 Mbps/18Mbps GPON ITU G.984 1.24 Gbps, 2.5 Gbps 155 Mbps, 622 Mbps, 1.2Gbps, 2.5 Gbps 1480 -1500 nm 1260 1360 nm Ethernet over GEM and/or ATM TDM over ATM/IP or native TDM RF overlay (over 1550 nm) and/or IP video GEM (Ethernet/IP, TDM) 64 <60km (~37 miles) 73 Mbps/35Mbps

Table: PON Standards Comparison

Page | 4

Perhaps the most dramatic distinction between the two protocols is a marked difference in architectural approach. GPON provides three Layer 2 networks: ATM for voice, Ethernet for data, and proprietary encapsulation for voice. EPON, on the other hand, employs a single Layer 2 network that uses IP to carry data, voice, and video. However, these aren't the only differences between the technologies. Designers will also find differences in terms of bandwidth, reach, efficiency, per-subscriber costs, and management. Let's look at each of these elements in more detail. 1. Usable Bandwidth Bandwidth guarantees vary between the two protocols: GPON promises 1.25-Gbit/s or 2.5Gbit/s downstream, and upstream bandwidths scalable from 155 Mbit/s to 2.5 Gbit/s. EPON delivers 1-Gbit/s symmetrical bandwidth. EPON's Gigabit Ethernet service actually constitutes 1 Gbit/s of bandwidth for data and 250 Mbit/s of bandwidth for encoding. 2. Reach With either protocol, the practical limitation to reach comes from the optical-link budget. With the reach of both protocols currently specified at approximately 20 kilometres, the difference in split rates the number of optical network units (ONUs) supported by one optical line terminal (OLT) is a point of differentiation. GPON promises to support up to 128 ONUs. With the EPON standard, there is no limit on the number of ONUs. Depending on the laser diode amplitude, when using low-cost optics, EPON can typically deliver 32 ONUs per OLT, or 64 with forward error correction (FEC). 3. Subscriber cost The use of EPON allows carriers to eliminate complex and expensive ATM and Sonet elements and to simplify their networks, thereby lowering costs to subscribers. Currently, EPON equipment costs are approximately 10 percent of the costs of GPON equipment, and EPON equipment is rapidly becoming cost-competitive with VDSL. 4. Efficiencies: With both PON protocols, a fixed overhead is added to convey user data in the form of a packet. In EPONs, data transmission occurs in variable-length packets of up to 1518 bytes. In ATM-based PONs, including GPONs, data transmission occurs in fixed-length 53-byte cells (with 48-byte payload and 5-byte overhead). This format makes it inefficient for GPONs to carry traffic formatted according to IP, which calls for data to be segmented into variablelength packets of up to 65,535 bytes. For GPONs to carry IP traffic, the packets must be broken into the requisite 48-byte segments with a 5-byte header for each. This process is time-consuming and complicated and adds cost to the central-office OLTs as well as the customer premise-based ONUs. Moreover, 5 bytes of bandwidth are wasted for every 48-byte segment. By contrast, using variable-length packets, Ethernet was made for carrying IP traffic and can significantly reduce the overhead relative to ATM. One study shows that when considering trimode packet size distribution, Ethernet packet encapsulation overhead was 7.42 percent, while ATM packet encapsulation overhead was 13.22 percent. In addition, since Ethernet frames contain a vastly higher ratio of data to overhead than GPON, that high utilization can be reached while using low-cost optics. The more precise
Page | 5

timing required with GPON results in more expensive optics. High-precision optics are mandatory as part of the GPON standard. 5. Management Systems: EPON requires a single management system, versus three management systems for the three Layer 2 protocols in GPON, which means EPON results in a significantly lower total cost of ownership. EPON also does not require multiprotocol conversions, and the result is a lower cost of silicon. GPON does not support multicast services, which makes support for IP video more bandwidth-consuming. 6. Support for CATV overlay: Both protocols support a cable television (CATV) overlay, which meets requirements for a high-speed downstream video service. EPON wavelengths are 1490 nanometres downstream and 1310 nanometres upstream, leaving the 1550-nanometer wavelength for a CATV overlay similar to the wavelengths for BPON and GPON. 7. Encryption: With GPON, encryption is part of the ITU standard. However, GPON encryption is downstream only. EPON, on the other hand, uses an AES-based mechanism, which is supported by multiple silicon vendors and deployed in the field. Furthermore, EPON encryption is both downstream and upstream. 8. Network Protection: Both protocols provide vendor-specific and carrier-specific protection. This includes support for vendor-specific and carrier-specific operations, administration and maintenance (OAM).

Recent Research:
A recent research of Siemens (2007) as part of an innovation project sponsored by the European Union, a research team from Siemens Networks was able for the first time to transmit error-free at a data transmission rate of 10 gigabits per second via a passive optical access network. This is four times faster than the rate previously possible. At the same time, a 100 kilometre distance between hub and subscribers, the so-called last mile, could be bridged for the first time. Today only subscribers within a 20 km radius can be connected. These research results bring within reach the next generation of subscriber line technology that is making optical networks more powerful and cost-efficient. Siemens Networks just recently announced a world record in connection with optical networks: During an experimental setup, 107 Gbit/s were transmitted in a carriers optical core network. Data rates up to 40 Gbit/s are currently possible in this network area for commercial operation. Now the research team from Siemens has reached a new highest record this time in the optical access network. To further optimize PON systems, Siemens Networks is taking part in the European Unionsponsored innovation program Multi Service Access Everywhere. Siemens developed the prototype for a new PON system as part of the subproject Extended reach PON systems. Using this system, it was possible to increase data transmission rates to 10 Gbit/s downstream and 2.5 Gbit/s upstream. At the same time, the range between hub and subscribers was extended to 100 kilometres and the number of subscriber lines per splitter was expanded to 512.
Page | 6

With the new PON system, network carriers can provide their customers at each hub greater bandwidth as well as offer cost-efficient lines to subscribers in more sparsely populated areas. In addition, the higher split factor in urban areas allows considerably more subscribers to be connected to a system. At the same time, the PON system eliminates the need for the aggregation devices that are now commonly used to collect and distribute data traffic locally. In the future, it will be possible to connect subscribers directly to the core network. Schulz (2008) in her recent research report of AMCC describes that GPON offers a higher split ratio, longer reach, and higher downstream bandwidth, compared to EPON. GPON also offers an advantage over EPON as a result of its high-efficiency mapping of Ethernet packets using a Generic Framing Procedure (GFP) -like scheme. While there are still provisions to transport ATM, almost all carriers have decided to deploy GPON in its Ethernet-only version, called GPON-lite. This is where GPON beats EPON on its home turf. In the downstream direction EPON efficiency is around 73 percent, which results in 875 Mbits/sec of billable bandwidth. In the upstream direction, the bandwidth efficiency is even worse and can be as low as 61 percent. In comparison, GPON bandwidth efficiency downstream is approximately 94 percent, thereby providing 2,250 Mbits/sec of billable bandwidth. Upstream efficiency is about 93 percent. A hot topic today is which form the next generation of PON will take. On an article of Engebretson (2008) found many vendors and service providers are gravitating toward an emerging PON architecture known as Wave Division Multiplexing PON (WDM-PON), which increases bandwidth by dividing a single fiber into multiple wavelengths, each capable of carrying the same bandwidth that previously required an entire fiber. Both the IEEE and ITU are looking ahead to 10-Gbit/sec and WDM-based PON. WDM is looking quite attractive, since it provides a relatively straightforward route for merging and scaling of existing EPON, BPON and GPON networks.

OFDMA-PON:
In OFDMA PON, all traffic flows are transmitted in parallel on a number of sub-carriers at different wavelengths; several sub-carriers can be further grouped into one sub-channel that may have different modulation/coding formats and data rates. The multiple-access in OFDMA PON is usually achieved by assigning subsets of sub-carriers to individual optical network unit (ONU) or a group of ONUs. Each ONU in one group can share the bandwidth resource dynamically via TDMA or hybrid OFDMA/TDMA. The OLT node structure has one fixed transmitter for the downstream usage, and one special parallel signal receiver which can receive multiple signals at different wavelengths simultaneously for the upstream usage. By using an appropriate signalling (e.g., request/grant) scheme, an adaptive bandwidth allocation scenario between OLT and ONU can be achieved in medium access control (MAC) layer, which also allows simultaneous data transmission in upstream/downstream.

Limitation of Current Optical Access Network:

According to Qian, Hu, Ji, Wang and Cvijetic (2008), next generation optical access networks are required to delivery various services for multiple customers simultaneously. Those services include legacy T1/E1, signals from cellular base stations, layer-2 VPN
Page | 7

services, security channels for storage network, etc. Current PON architectures (EPON, GPON) require complex scheduling algorithms and framing technology to support a variety of services. The performance of these TDM based PONs is highly sensitive to the latency of the packets transmission and can be easily affected by other data traffic flowing through the same link. On the other hand, WDM-PON can transparently delivery multiple services since each service can use a dedicated wavelength. However, multiple wavelengths would also require multiple transceivers and arrayed-waveguide grating (AWG) or optical filters to distribute wavelengths to the correct receivers, which will increase both system cost and complexity. In addition, WDM-PON also lacks the flexibility to dynamically allocate the resource among different services. From another research of Wei, Wang, Qian and Hu (2008) mention some other limitation of current PON includes physical impairments, craving for higher bandwidth, quality of services etc.

How to Overcome these Limitations:

Xu, Qian, Hu, Wei, and Wang (2008) found Orthogonal Frequency Division Multiplexing (OFDM) has been widely recognized as a promising digital multicarrier modulation scheme for high-speed transmissions in wireless and wired systems, e.g., WiMAX and xDSL. According to Djordjevi et al (2006), OFDM has also been investigated in the area of optical fiber communications; it was shown that great benefits (e.g., the mitigation of physical impairments) could be achieved when employing OFDM techniques in long haul high-speed optical transmission systems (e.g., 100Gb/s). Furthermore, the OFDMA, a multi-user version of OFDM, is capable of assigning subsets of sub-carriers to different users dynamically. Thus, the total system bandwidth resources can be shared by all users. Qian et al. (2008) in their research propose a novel system architecture based on OFDMAPON which transparently support various services and enable dynamic bandwidth allocation among these services. In addition, a single optical receiver can receive multiple services at different wavelengths simultaneously, while neither the AWG nor optical filters are required. They experimentally demonstrated WiMAX radio frequency (RF) signal (3.5MHz bandwidth, 3.4GHz carrier frequency) transmission through a 40MHz pipe at 3.4GHz in a 5GHz bandwidth OFDMA-PON. As the modulated data streams are orthogonal to each other in frequency domain, meaning that cross-talk between the sub-channels is eliminated. By removing some sub-carriers, transparent pipes can be placed within the overall OFDMA signal bandwidth, the pipes could be used for relaying various signals or traffics from difference services or systems without involving any layers other than PHY. They also experimentally demonstrated a heterogeneous 10-Gb/s OFDMA-PON architecture by transmission of standard WiMAX over 20km fiber, together with two PON channels. The uplink WiMAX throughput remained after 20km fiber transmission through a 40MHz pipe at 3.4GHz which was embedded inside 5GHz ONU-s broadband signal. The experiment also proved that the WiMAX signal transmission did not affect regular OFDMA-PON traffics.

Page | 8

Wei et al. (2008) studied the application of OFDMA technologies in optical access networks. They presented a novel flexible, high bandwidth, and cost-effective OFDMA-PON, which combines the advantages of PON with OFDMA technologies, and transparently supports various types of services and enables dynamic sub-carriers allocation to provide bandwidth sharing across multiple optical network units (ONUs) and applications. Their simulation mainly focuses on the QoS performance of packet loss ratio (PLR) and packet average delay of traffic flows. They conclude DCT (Discrete Cosine Transform) is more suitable than FBT (Fixed Burst Time) in supporting voice/PWE/TDM traffic flows, whereas FBT is more suitable than the DCT in supporting data/video traffic flows. The unique feature of DCT is a simple signalling and can guarantee the QoS performance of non-bursty traffic flows well; however it may be inefficient when supporting highly dynamic traffic flows since DCT could not estimate the bandwidth accurately for sub-carriers allocations. Another research of Wei, Wang and Qiao (2008) refer how to support the slicing of bandwidth resources as well as the virtualization mechanism of control in next generation optical access networks. In their paper they studied one virtualization mechanism for a novel PON based on OFDMA technologies. By taking advantage of the transparent parallel data transmissions, high bandwidth, and the advanced digital signal processing (e.g., QAM and FFT/IFFT modulation), they present a design concept of virtual PON (VPON) in a physical OFDMA-based PON architecture. Each VPON may be Ethernet PON (EPON), Gigabit PON (GPON), or WiMAX based access, which are usually different in terms of frame formats, multiplexing methods, MAC protocols, and scheduling algorithms. The concept of VPON makes more efficient use of the wavelength bandwidth resource and, unlike SONET ring or WDM-PON, OFDMA-based VPON provides the transport for different types of traffic in their native formats, giving each a separate sub-channel that might be operating at a different MAC from other sub-channels. Their simulation investigates the impacts of different resource provisioning models (dedicated and shared) with different traffic profiles (high burst ratio and low burst ratio) on the VPON performance. From simulation results they observe that the shared model outperforms the dedicated model (an equivalent view is that given the same number of VPONs, the shared model can achieve a lower PLR). They also observe that the average packet delay (including transmission delay plus queuing delay) increases dramatically as the number of VPONs increases, but the shared model always outperforms the dedicated model. The dedicated model has a better performance than the shared model which is also attributed to the dynamic allocation property of the shared model. Conclusion: In summary, optical fiber represents the optimal waveguide for delivering digital information to business and residential subscribers. Arguably, PON technologies are the optimal form of FTTP, and GPON is the optimal form of PON. With the economic tipping point having been reached, telecommunications and cable service providers alike are increasingly electing to employ GPON in all builds. Increasingly, first with telecoms, but eventually with cable operators as well, we expect to see all service providers moving toward an end-to-end fiber
Page | 9

infrastructure. As this takes place, OFDMA-PON, with its superior bandwidth and operational flexibility, is emerging as the technology that will dominate all access networks in the future. References: 1. Onn Haran (Jan 25, 2005), EPON vs. GPON: A Practical Comparison, Passav Technologies, Comms Design. 2. Rajiv Ramaswami, Kumar N. Sivranjan (2004), Optical Networks A Practical Perspective (2nd edition), chapter 11, page 591-593. 3. Glen Kramer, Biswanath Mukherjee, Gerry Pesavento (Feb. 2002), Ethernet PON (EPON): Design and Analysis of an Optical Access Network, IEEE Communication Magazine, pp. 6673. 4. G. Pesavento and M. Kelsey (September 1999), PONs for the broadband local loop, Lightwave, PennWell, vol. 16, no. 10, pp. 68 74. 5. B. Lung (September 1999), PON architecture futureproofs FTTH, Lightwave, PennWell, vol. 16, no. 10, pp. 104 107. 6. IEEE 802.3 Ethernet in the First Mile Study Group [Online]. Available: http://www.ieee802.org/3/efm/public/index.html 7. ITU-T, Study Group 15 (2001), G.983: High Speed Optical Access Systems based on Passive Optical Network (PON) Techniques. 8. Full Services Access Networks. [Online]. Available :http://www.fsanet.net/ 9. Successful Siemens research: Passive optical access network with faster data transmission and extended reach, (February 19, 2007), Press Presse Prensa, Munich [Online]. Available: http://www.siemens.cz/ 10. Armin Schulz (22 December, 2008), PON opens the bandwidth door to the home, AMCC, Lightwave, PennWell. 11. Joan Engebretson (September, 2008), Get Ready for WDM-PON, article reprint Tellabs Emerge Magazine. 12. Djordjevic I.B, Vasic B. (Aug. 2006), 100-gb/s transmission using orthogonal frequencydivision multiplexing, Photonics Technology Letters, IEEE Volume 18, Issue 15, Page(s):1576 1578. 13. Lei Xu, Dayou Qian, Junqiang Hu, Wei Wei, Ting Wang (2008), OFDMA-based passive optical networks (PON), IEEE/LEOS Summer Topical Meetings, 2008 Digest 21-23 July, Page(s):159 160, 14. Wei Wei, Ting Wang, Chunming Qiao (2008), Resource Provisioning for Orthogonal Frequency Division Multiple Access (OFDMA)-based Virtual Passive Optical Networks (VPON), Optical Fiber communication/National Fiber Optic Engineers Conference, 2008. OFC/NFOEC 2008, Conference on 24-28 Feb. Page(s):1 3. 15. Wei Wei, Ting Wang, Dayou Qian, Junqiang Hu (2008), MAC Protocols for Optical Orthogonal Frequency Division Multiple Access (OFDMA)-based Passive Optical Networks, Optical Fiber communication/National Fiber Optic Engineers Conference, 2008. OFC/NFOEC 2008, Conference on 24-28 Feb. Page(s):1 3. 16. D. Qian, J. Hu, P. Ji, T. Wang, M. Cvijetic (2008), 10-Gb/s OFDMA-PON for Delivery of Heterogeneous Services, Optical Fiber Communication Conference and Exposition and The National Fiber Optic Engineers Conference, OSA Technical Digest (CD), Optical Society of America, 2008.

Page | 10