A Review of Dynamic Bandwidth Allocation Algorithm for GPON Networks

Over the years, a lot of research works have been carried out to study the Dynamic Bandwidth Allocation (DBA) algorithm of GPON network structure and different type of FTTx model. Some of them are discussed below.

Matthew Andrews et al. (1998) propose several algorithms for scheduling the efficient transmission of data to multiple users [12]. They also described that a new feature, the various schemes exploit knowledge of the buffer contents to achieve high throughput, while maintaining fairness by providing quality of service (QoS) to individual users. The proposed algorithms are backward compatible with existing cellular and personal communications services (PCS) standards such as IS-136. They provided a powerful approach to improving spectrum efficiency in forthcoming high-speed data cellular services. The extensive simulation experiments were demonstrated that the algorithms significantly outperform conventional schemes. In this paper they did not provided any idea about dynamic bandwidth allocation for GPON network.

Rudy Hoebeke et al. (1999) have described about Dynamic bandwidth allocation on SuperPON [13]. The SuperPON system exploits all possible upgrades of an FSAN APON system. Due to the significant changes in system parameters, it becomes of the utmost importance to implement a dynamic Medium Access Control (MAC) protocol on the SuperPON, since the upstream bit rate is rather limited for such a large number of connected Optical Network Units. Main challenges specific to the SuperPON are the impact of long range and high number of users on the efficiency of the MAC protocol. In this research work they also missed the point for how bandwidth will be dynamically allocated for the users. It focused only dynamic medium access control.

M.C. Yuang et al. (2000) theoretically described that a new contention access protocol (CAP) augmented with a dynamic bandwidth allocator (DBA) for wireless ATM networks supporting ABR and signalling control (SCR) traffic [14]. CAP incorporates a dynamic tree-splitting collision resolution algorithm parameterized by an optimal splitting depth (SD). DBA performs estimation and on-line prediction of ABR self-similar traffic characteristics. It in turn determines the optimal SD per frame, satisfying ABR throughput and SCR blocking probability requirements while retaining maximal aggregate throughput. Simulation results postulate the optimal SDs under various ABR and SCR traffic conditions. This research work also unable to meet the Dynamic Bandwidth Allocation criteria for GPON system.

Joanna Ozimkiewicz et al. (2001) have simulated GPON System with Two Dynamic Bandwidth Allocation algorithms used for coordination of the available bandwidth between end users in a GPON network have been simulated using OPNET to determine and compare the performance, scalability and efficiency of status reporting and non-status reporting dynamic bandwidth allocation [15]. They described that the status reporting is more efficiently using the bandwidth while non-status reporting provides better QoS for real time services. Their solution however requires more bandwidth due to the over allocated bandwidth. The choice of the algorithm should be made depending on the QoS SLAs agreed upon with particular customers and the dominant traffic type in the network.

Saki Hatta et al. (2001) have addressed on low latency dynamic bandwidth allocation (DBA) method with high bandwidth efficiency that is intended for use in campus area networks and mobile fronthaul based on TDM-PON (time division multiplexing passive optical network) [16].  Their proposed network systems require low latency of under 100 μs and high bandwidth efficiency. Their method involves only three steps for allocation and employs an adaptive DBA cycle depending on the traffic load. The DBA cycle length, which is proportional to the latency, can be minimized because the simple allocation steps are appropriate for hardware implementation. Their DBA method automatically optimizes the cycle length to reduce the latency and improve bandwidth efficiency. They also implemented it on a 10-gigabit Ethernet passive optical network (10G-EPON) media access control system-on-a-chip and evaluated the allocation results and the latency on the 10G-EPON system. Our DBA achieved a minimum latency of 60 μs with priority control and high bandwidth efficiency, depending on traffic. They demonstrated that their method automatically adjusts the DBA cycle length and the bandwidth efficiency depending on traffic load. These results show that our DBA can be used in various networks employing TDM-PON, such as future campus-LANs and 5G MFH.

S. Yoshmara et al. (2002) have proposed a new DBA algorithm for broadband passive optical network (B-PON) that completely satisfies the requirements specified in ITU-T G983.4 [17]. They presented an algorithm that consist of three processes. The first process predicted the requested bandwidth and ensures quick response to ONT input data. The second process determine the number of cell slots (CS) of each bandwidth type in each bandwidth assignment period (BAP) and ensures complete assignment of guaranteed bandwidth, and fair assignment for non-assured bandwidth among transmission containers (T-CONT). The third process determine the CS locations of each bandwidth type in each BAP according to the number of CS obtained from the second process so as to minimize the cell delay variation (CDV). They also evaluated the proposed algorithm using a prototype system. But they proposed it only for B-PON system, there is no proposal for GPON networks.

C.M. Assi et al. (2003) have proposed to use the multipoint control protocol defined by the IEEE 802.3ah task force to arbitrate the transmission of different users, and they presented different dynamic bandwidth allocation (DBA) algorithms to allocate bandwidths effectively and fairly between end users. [18]. These DBA algorithms are also augmented to support differentiated services, a crucial requirement for a converged broadband access network with heterogeneous traffic. They simulated that queueing delays under strict bandwidth allocation algorithms result in an unexpected behavior for certain traffic classes, and they suggested the use of DBA with appropriate local queue management to alleviate this inappropriate behavior.

Michael P. McGarry et al. (2004) have introduced a framework for classifying dynamic bandwidth allocation schemes and provide a comprehensive Survey of the dynamic bandwidth allocation methods proposed to date. [19]. They presented a side by side comparison of the schemes based on their most prominent characteristics, and outline future developments of dynamic bandwidth allocation schemes.

Chengwei Xiao et al. (2004) have addressed an algorithm based on traffic prediction is proposed, in bid to predict instantaneous traffic in ONU, guide higher-priority bandwidth allocation, and avoid “T+2” queuing delay [20]. They simulated that the Dynamic Bandwidth Allocation (Pre_DBA) for EPONs, which also includes a prediction corrected mechanism, to maximize bandwidth utilization. Their proposed algorithm was ensuring effectiveness and fairness of the premise, was quite able to alleviate delay and jitter in higher-priority services as well but this proposal was only for EPONs.

A. Shami et al. (2005) have stated a new dynamic bandwidth allocation (DBA) algorithm that effectively allocate bandwidths between end users [21]. They presented DBA algorithm for EPON Network which can support differentiated services in a network with heterogeneous traffic.

Yao Liang et al. (2006) have focused on prediction-based dynamic bandwidth allocation and examined the multiresolution learning neural-network-based traffic predictor [22]. They discussed a well-known-heuristic based approach RED-VBR scheme was used as a baseline for performance evaluation and a new concept of renegotiation control was introduced and a novel renegotiation control algorithm based on binary exponential backoff (BEB) was proposed to efficiently reduce renegotiation frequency. Dynamic bandwidth allocation attempts to adaptively allocate resources to capture the burstiness of VBR video traffic, and therefore could potentially increase network utilization substantially while still satisfying the desired QoS requirements.

J. Elias et al. (2006) have introduced a new service model that provides quantitative per-flow bandwidth guarantees, where users subscribe for a guaranteed rate; moreover, the network periodically individuates unused bandwidth and proposes short-term contracts where extra-bandwidth was allocated and guaranteed exclusively to users who can exploit it to transmit at a rate higher than their subscribed rate [23]. They developed an efficient bandwidth allocation algorithm that takes explicitly into account traffic statistics to increase the users’ benefit and the network revenue simultaneously. They demonstrated through simulation in realistic network scenarios that the proposed dynamic provisioning model was superior to static provisioning in providing resource allocation both in terms of total accepted load and network revenue.

Si Yin et al. (2007) have established a system model to analyse the stability of the predictor-based dynamic bandwidth allocation (PDBA) scheme over Ethernet passive optical networks (EPONs) [24]. They proved that an EPON system with PDBA was stable by proper pole placement as the traffic changes dynamically. Their suggested a straightforward framework for designing the DBA algorithm that enables EPONs with stability.

Onn Haran et al. (2008) have described Bandwidth Allocation in the PON network and highlights the advantages of Dynamic Bandwidth Allocation (DBA). A well-defined DBA algorithm can significantly improve network performance, provide a means of flexibly tailoring network responsiveness and enable a service provider to generate more revenue from their FTTH networks without boosting raw bandwidth by increasing the percentage of acceptable oversubscription [25]. They compared currently available Bandwidth Allocation methods and their impact on the service provided to the subscriber. By simulating a real-life TCP download scenario, it shows that Status Reporting (SR) DBA is a superior tool to provide a high level of quality of service.

Ching-Hung Chang (2008) has addressed towards improving the performance of passive optical networks (PONs) through the development of advanced dynamic bandwidth allocation (DBA) protocols. [26]. His main aim was to enhance the quality of service (QoS) offered by standard PONs by means of providing subscribers with service level agreement (SLA) to fulfil applications and associated bandwidth requirements on demand.

Liu Yang et al. (2009) have analysed and proposed GPON protocols and network configuration, the information about dynamic bandwidth allocation (DBA) in GPON protocols, a new DBA algorithm based on the all the transmission containers (TCONT) [27]. They elaborated the specific algorithm for improved scheme, and builds the GPON model for this algorithm and test its performance. Their proposed algorithm allocated the bandwidth of the upstream channel for the GPON and based on the all the T-CONT, and using the polling algorithm to solution the increase the channel utilization.

Bjorn Skubic et al. (2009) have compared the typical characteristics of DBA, such as bandwidth utilization, delay, and jitter at different traffic loads, within the two major standards for PONs, Ethernet PON and gigabit PON [28]. They illustrated some differences between EPON and GPON by means of simulations for the two standards. Moreover, they considered the evolution of both standards to their next-generation counterparts with the bit rate of 10 Gb/s and the implications to the DBA. A new simple GPON DBA algorithm was used to illustrate GPON performance. It was shown that the length of the polling cycle plays a crucial but different role for the operation of the DBA within the two standards. Moreover, only minor differences regarding DBA for current and next-generation PONs were found.

Liu Yang et al. (2010) have analysed the ITU-T GPON standard G.984.1 to G.984.6 and GPON network configuration, the information about dynamic bandwidth allocation (DBA) in GPON standard [29]. They build the GPON model for this algorithm and tested its performance. They found that the average delay in different frame cycle can satisfy the Quality of Service (QoS) requirements of GPON.

Joanna Ozimkiewicz et al. (2010) have proposed two approaches for Dynamic Bandwidth Allocation in GPON networks and validated through simulations. One approach addresses a Status Reporting scheme, where the bandwidth allocation originates from the client request. The second use a centralized Non-Status Reporting scheme. Furthermore, parameters to cope with variances in the traffic pattern was quantified [30]. Their proposed SR algorithm utilized bandwidth more efficiently than NSR. In the low loaded networks where a lot of bandwidth was available, SR algorithm allocates around 5% more bandwidth than T-CONTs source load. Efficiency of NSR algorithm depends highly on the expansion factor used, and the assured bandwidth parameter of the T-CONT. If high assured bandwidth is granted, with expansion factor of 1.25, T-CONTs obtain bandwidth grants for up to 129% of the source bandwidth.

Mustafa Vahabzadeh Dolama et al. (2010) have proposed a new Dynamic Bandwidth Allocation (BDA) scheme called Multi Group DBA (MG-DBA) to improve utilization and to decrease packet delay [31]. They simulated that in MG-DBA, ONUs are classified into several groups. Therefore, when one group is sending its data to the OLT, the bandwidth for another group is simultaneously calculated. In addition, the OLT uses the Smallest Available Report First (SARF) for bandwidth calculation for each group to minimize packet delay.

I-Shyan Hwang et al. (2011) have introduced an adaptive priority scheduling integrated with Bi-partition Dynamic Bandwidth Allocation (APS-BDBA) mechanism to enhance the differentiated services over GPON and in the Inter-ONU scheduling, the B-DBA mechanism to divide the cycle time by partitioning the ONUs into two groups with some timing overlap to execute interleaved bandwidth allocation, and adjusts the bandwidth dynamically between the first group for high priorities T-CONT 1-2 and the second group for low priorities T-CONT 3-4 to eliminates the idle period problem and reduce bandwidth waste [32]. In the Intra-ONU scheduling, they designed an adaptive priority scheduling embedded in the ONUs, which will meet the minimum delay of highest priority traffic without sacrificing the low priority traffic bandwidth requirements. The scheduling is constructed by one System Queue (SQ), for T-CONT 1, and multi-Custom Queue (CQ), for T-CONT 2-4, when the congestion occurs, the CQ can be maintained in the application of the bandwidth in accordance with the ratio corresponding to DiffServ to improve CoS. The system performance of APS-BDBA is compared with the Jiang’s protocol and DWFQ-MDBA in terms of the throughput, end-to-end

delay and jitter in different proportion of traffic scenarios.

Zhang Qi-yu et al. (2012) have proposed a novel dynamic bandwidth allocation (DBA) scheme called P-DBA based on traffic prediction for GPON in order to improve the performance of Gigabit Passive Optical Network (GPON) systems, which can better meet the optical network’s technical requirements of high bandwidth, low delay and low packet loss rate [33]. They stated that using the high-order moving average model, P-DBA scheme predicts the amount of high priority traffic arriving during the waiting period, as a basis for bandwidth allocation, so as to ensure the performance of high priority services.

Fahmida Rawshan et al. (2012) have proposed a Guaranteed Limited Dynamic Bandwidth Allocation (GLDBA) scheme for flexibly managing downstream bandwidth among multiple OLTs [34]. For upstream transmission in the proposed multi-OLT PON system, they modified Limited Dynamic Bandwidth Allocation (LDBA) scheme of the existing EPON. Performance analysis showed that the PON architecture can successfully provide services from multiple service providers to access area.

Antonios Sarigiannidis et al. (2013) have provided a comprehensive compilation of the latest architectures, integrated technologies, QoS features, and dynamic bandwidth allocation (DBA) schemes. In addition, new trends towards wireless-optical convergence were presented [35]. They demonstrated an up-to-date hybrid network standard remain under development, since there is not yet an integrated standard for approving hybrid network access platforms.

Chih-Ta Chiu et al. (2013) have proposed a request-based dynamic bandwidth allocation (DBA) of gigabit passive optical network (GPON), the optical line terminal (OLT) in GPON grants bandwidth to optical network units (ONUs) and ONUs report request bandwidth which depends on queue lengths of traffic containers (TCONTs) to the OLT [36]. They simulated that in the OLT, DBA of GPON supports a request-based polling order to allocate bandwidth and their request-based dynamic bandwidth allocation focuses on weight assignments in the request-based polling order. Weight assignments allocate bandwidth in proportion to guaranteed and request bandwidth.

Sandu Abeywickrama et al. (2014) have presented two novel single-receiver based dynamic bandwidth allocation algorithms to further optimize local storage aided video-on-demand over passive optical networks [37]. The quality-of-service and power performances of the algorithms are critically analyzed using packet level simulations and formulation of power consumption models. They showed that the energy-efficiency of a local storage-based video-on-demand scheme can be increased without compromising the quality-of-service by the use of single receiver algorithms. Further, they compared the two newly introduced algorithms against dual-receiver based and without local storage schemes to find the most appropriate bandwidth allocation algorithm for local storage-based video-on-demand delivery over passive optical networks.

Vipan Kumar et al. (2014) have introduced an efficient PDF based DBA algorithm which uses the last recent polling table for the current allocation of the bandwidth among the ONU’s dynamically [38]. Their proposed scheme can early allocate bandwidth to the ONU’s requesting bandwidth lower than minimum guaranteed bandwidth and ONU’s demanding bandwidth greater than minimum guaranteed whose PDF in the last polling table is greater than threshold value. It allocates excess bandwidth of lightly loaded ONU’s to the heavily loaded ONU’s. This algorithm uses the early assignment concept thus incorporates the idle period.

Ozgur Can Turna et al. (2015) have proposed a basic schemes of dynamic bandwidth allocation algorithms on EPON have been presented and a novel online-offline based method called hcDBA (Half Cycling Dynamic Bandwidth Allocation). In addition, hcDBA algorithm has been improved with early prediction mechanism (p-hcDBA) [39]. They investigated that hcDBA algorithm performs better than existing mechanism in terms of byte loss ratio and access delays criteria, p-hcDBA also improved the performance in terms of access delay.

Li Li et al. (2016) have made a detailed analysis and comparison of the existing bandwidth allocation algorithm, the deterioration of light load, low utilization rate of bandwidth, poor fairness and inability to guarantee QoS are pointed out and the original dual-clocking DBA algorithm was put forward. Meanwhile, OLT/ONU bidirectional authentication during registration was proposed and Diffie-Hellman key exchange protocol was introduced [40]. They experimented that the information that can perform key exchange was embedded into authentication information and passing another initial key was no longer needed for the following data encryptions, which strictly ensured system security in the authentication step.

N. Merayo et al. (2017) have introduced a new DBA scheme to allow a fair and efficient allocation of the upstream channel capacity, a FEx-DBA (fair excess-dynamic bandwidth allocation) algorithm using the NUM model [41]. They showed that FEx-DBA (i) provides bandwidth guarantees to the users according to the service level agreement (SLA) contracted and fairly distributes the excess bandwidths among them; (ii) has a stable response and fast convergence when traffic or SLAs change, avoiding the oscillations appearing in other proposals; (iii) improves average delay and jitter measures; and (iv) only depends on a reduced set of parameters, which can be easily tuned.

M K Multani et al. (2017) have reviewed the algorithms for Time Division Multiplexed (TDM) EPON and Hybrid TDM/WDM (wavelength division multiplexed) EPON. A Partially Online DBA (ParOnD) for Hybrid TDM/WDM EPON was proposed to improve bandwidth utilization in Hybrid EPONs [42]. They analyzed that ParOnD reduces both packet delay and the average queue depth of an Optical Network Unit (ONU) over low to medium loads and thus, ParOnD improves bandwidth utilization considerably.

Weibing Zhou et al. (2017) have proposed a novel Cross-layer Dynamic Bandwidth Allocation Algorithm Based on Convex Optimization Theory (CO-CDBA) in Digital Video Broadcast-Return Channel via Satellite system [43]. They analysed a Dynamic Bandwidth Allocation (DBA) problem which converted to a convex optimization problem with constraints and can be solved by using a fast interval iterative algorithm. The allocation result, utility value, users’ fairness performance and allocation procedure based on CO-CDBA were simulated.

Yuying Liu et al. (2018) have focused on the design of DBA protocol for OFDM-PON supporting multi-service. A DBA mechanism was also proposed under the MAC framework of the widespread Ethernet PON, such that OFDM-PON can be compatible with the existing PON [44]. They further proposed to guarantee the ultra-low latency of haptic communication and achieve the fair transmission of regular services. Performance results showed that the proposed mechanism was effective in guaranteeing the round-trip latency of haptic communication with less impact on the regular services.

Oscar J. Ciceri et al. (2018) have introduced a mechanism for the support of multi-ONU service level agreements (SLAs) in dynamic bandwidth allocation (DBA) algorithms for Ethernet passive optical networks (EPON). The employment of SLAs for multiple optical network units (ONUs) instead of individual ONUs allows better utilization of the bandwidth reserved for these ONUs. [44]. Their proposed DBA mechanism allows customers owning multiple ONUs to redistribute the aggregated bandwidth of the group of ONUs to better balance the bandwidth utilization and proposed DBA can be employed in different use cases such as mobile backhauling/fronthauling, PON virtualization, and multi-site enterprise networking. Their simulation results showed that the proposed DBA improves the network performance.

Syed Baqar Hussain et al. (2018) have demonstrated an online gated service DBA algorithm for NG-EPON to assign a flexible number of wavelengths and a dynamic grant size on each upstream wavelength. The assignment was based on the difference between the start time of the wavelengths and the effectiveness in reducing delay by transmitting on multiple wavelengths [44]. To evaluate the transmission effectiveness, an optimum criterion alpha was introduced in relation to the number of ONUs in the network. The proposed DBA algorithm’s performance was evaluated in terms of average packet delay, grant utilization, number of wavelengths used, and resequencing delay in comparison to modified-IPACT, the recently proposed First-Fit DBA (FF-DBA), and water-filling-algorithm-based DBA (WF-DBA) algorithms. Their simulation results confirmed the validity of the proposed DBA algorithm.

Tania Panayiotou et al. (2019) have proposed a data-driven bandwidth allocation (BA) framework for periodically and dynamically reconfiguring an elastic optical network according to predictive bandwidth allocation (PBA) models [44]. Their proposed framework is scalable to the number of network connections and also adaptive to the increasing traffic of each network connection separately and to the overall network load as well and that was achieved by formulating the BA problem as a Partially Observable Markov Decision Process, which constitutes were reinforcement learning (RL) model.

Be the first to comment

Leave a Reply

Your email address will not be published.


*


For security, use of Google's reCAPTCHA service is required which is subject to the Google Privacy Policy and Terms of Use.

I agree to these terms.