Next Article in Journal
A Multimodal Sensing Device for Simultaneous Measurement of Dissolved Oxygen and Hydrogen Ions by Monolithic Integration of FET-Based Sensors
Previous Article in Journal
Aurora Image Classification with Deep Metric Learning
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sensors
Volume 22
Issue 17
10.3390/s22176668
sensors-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

Comparison of Named Data Networking Mobility Methodology in a Merged Cloud Internet of Things and Artificial Intelligence Environment

by
Wan Muhd Hazwan Azamuddin
1,*,
Azana Hafizah Mohd Aman
1,
Rosilah Hassan
1 and
Norhisham Mansor
2
1
Center for Cyber Security, Faculty of Information Science & Technology, Universiti Kebangsaan Malaysia, Bangi 43600, Malaysia
2
Department of Electrical Technology, Advance Technology Training Center (ADTEC) Batu Pahat, Batu Pahat 83000, Malaysia
*
Author to whom correspondence should be addressed.
Sensors 2022, 22(17), 6668; https://doi.org/10.3390/s22176668
Submission received: 24 July 2022 / Revised: 31 August 2022 / Accepted: 31 August 2022 / Published: 3 September 2022
(This article belongs to the Section Sensor Networks)
Browse Figures
Review Reports Versions Notes

Abstract

:
In-network caching has evolved into a new paradigm, paving the way for the creation of Named Data Networking (NDN). Rather than simply being typical Internet technology, NDN serves a range of functions, with a focus on consumer-driven network architecture. The NDN design has been proposed as a method for replacing Internet Protocol (IP) addresses with identified content. This study adds to current research on NDN, artificial intelligence (AI), cloud computing, and the Internet of Things (IoT). The core contribution of this paper is the merging of cloud IoT (C-IoT) and NDN-AI-IoT. To be precise, this study provides possible methodological and parameter explanations of the technologies via three methods: KITE, a producer mobility support scheme (PMSS), and hybrid network mobility (hybrid NeMO). KITE uses the indirection method to transmit content using simple NDN communication; the PMSS improves producer operation by reducing handover latency; and hybrid NeMO provides a binding information table to replace the base function of forwarding information. This study also describes mathematical equations for signaling cost and handover latency. Using the network simulator ndnSIM NS-3, this study highlights producer mobility operation. Mathematical equations for each methodology are developed based on the mobility scenario to measure handover latency and signaling cost. The results show that the efficiency of signaling cost for hybrid NeMO is approximately 4% better than that of KITE and the PMSS, while the handover latency for hybrid NeMO is 46% lower than that of KITE and approximately 60% lower than that of the PMSS.

1. Introduction

Named data networking (NDN) is a new type of architecture based on the Internet Protocol (IP) architecture that makes the Internet work [1]. To connect the various elements of the network, the IP is used as the network layer, incorporating security and device-to-device connectivity. In Internet hourglass architecture, the thin part is at the center of the universal IP network layer. Recent NDN research adopted an adaptive forwarding strategy based on deep reinforcement learning, maintaining the right data delivery equilibrium and allowing users to pick and compete [2]. With its advantages of speed and reliability, NDN is a highly appealing networking solution for Internet of Things (IoT) applications. The goal of the IoT is to connect everyone and everything at any time and from any location. The IoT is being adopted by an increasing number of users in enterprises and academic sectors [3]. Many issues arise as a result of the rapid expansion of IoT devices on the Internet because the current Internet was not designed to handle resource-constrained devices [4]. However, due to the peculiarity of this new method, a new paradigm for IoT must go a step further [5]. This situation means that an alternative Internet design must be investigated—one that takes the restrictions and quantity of devices into account. Other research indicated that NDN is a crucial enabler of network and compute convergence, upon the edge of which artificial intelligence (AI) should be built [6].
In NDN, a consumer sends an interest packet (I_packet) with a content prefix (prefix) [7]. The router records the interest packet’s interface in the pending interest table (PIT) and looks up the forwarding information base (FIB) for information about the content filled by the name-based routing (NBR) protocol. Some routers or content providers send back information to a consumer and store the content in the content store (CS).
Another aspect of NDN that still has no updated solution is mobility. A survey paper [8] derives aspects of mobility that currently remain unresolved. This study discusses NDN mobility methodologies and measures the performance evaluation for each producer mobility operation: KITE [9], a producer mobility support scheme (PMSS) [10], and hybrid network mobility (hybrid NeMO) [11].
Motivation: NDN technology comes with a new perspective as a blank-slate Internet architecture under Information-Centric Networking (ICN). The major goal of NDN technology is to address the existing Internet’s flaws by shifting the IP-based nature of communication to content-transmission communication. Furthermore, NDN was created to facilitate mobile content transmission. Nevertheless, with the rapid development of networking and the proliferation of mobile devices, numerous academics, academic institutions, and organizations have explored ways to manage the mobility of network-connected mobile devices. Moreover, NDN has emerged as the future Internet method of addressing mobility issues. Mobility support enables mobile devices to move between various points of attachment (PoAs) without interfering with content delivery and with minimal handover delay. A PoA makes it possible for mobile nodes to connect to a network. Therefore, mobility was separated into mobility for consumers and producers. According to Saxena et al. [12], the consumer-driven nature of NDN is naturally supported by consumer mobility. When a mobile user moves to a new PoA, the I_packets must be present for transmission to continue normally or to resume after handoff. The mobility of content producers has minimal issues when compared with those of IP architecture [13].
Producer mobility means that a content provider can move without causing problems for consumers and intermediate routers, in terms of content availability and location, with as little time as possible between moves. Consequently, on the basis of a new perspective on mobility support for NDN, Zhu et al. determined that NDN does not support producer mobility [14]. NDN faces a routing table size scaling issue. In addition, when providers relocate to a different location, the naming structure introduces significant scalability challenges [15]. In the same way, when an object moves, a new route must be announced and spread to replace the old routing information.
Another problem that needs to be solved is the long handoff latency and unnecessary I_packet losses during transmission to a producer’s old location [16,17]. When a producer moves, the I_packets keep following the prefix trace in the FIB, which means the I_packets do not reach the producer and are dropped. In addition, indirect points to support producer mobility stretch the data path and cause a problem. Therefore, the circumstances cause a long handoff time. Furthermore, producer mobility requires additional research to find a concrete solution so that the NDN architecture can confidently be used to replace the current Internet architecture, with a lack of problems.
In conclusion, a great deal of optimism has been expressed about NDN becoming a future Internet architecture that is capable of incorporating other networks, without the need for additional mechanisms. Cisco proposed incorporating hybrid ICN into the 5G network [18], which prompted many researchers to improve the transmission of ICN into the 5G network, NDN-5G-SDN support [19], and IoT networking [20,21]. Therefore, more research is needed, especially when it comes to how NDN helps producers move around.
Article Contribution: No recent review papers address the latest NDN mobility issues. Previously, most NDN mobility assessments concentrated on NDN surface design and quality, while ignoring NDN mobility methods and unresolved difficulties. This study provides solutions to some of the questions and problems that remain about NDN mobility. This study’s contributions are as follows: (1) It examines NDN technology, AI, cloud IoT (C-IoT), and NDN-AI-IoT; (2) It provides an explanatory methodological and parameter setup for KITE, the PMSS, and hybrid NeMO; (3) It constructs mathematical equations for signaling cost and handover latency; and (4) It conducts a simulation experiment by using NS3 for network performance analysis.
Article Organization: The study consists of four sections. Section 2 describes NDN, NDN with AI, and NDN with C-IoT, in terms of features and technology applications. Section 3 describes and compares the mobile operation methodology on mobile producer operations that are integrated with AI and cloud services, i.e., KITE, the PMSS, and hybrid operations. Section 4 provides a performance comparison, including the mathematical equations for the NDN-AI-C-IoT mobility method based on signaling cost and handover latency. Section 5 discusses the network performance for signaling cost and handover latency. Section 6 provides the conclusions and recommendations for future research.

2. Background

This section briefly explains NDN technology, NDN with AI implementation technology, NDN with embedded cloud technology, and NDN with IoT application. The current Internet architecture follows the process of data transmission by referring to the open systems communications (OSI) model, but new evolving data-centric technology uses the data-centric NDN model to transmit content, as depicted in Figure 1. NDN transmission focuses on layer 3 for content transmission, while for the OSI layer, the IP packet is used in layer 3. A new NDN paradigm uses a centralized cloud-based strategy for deploying AI techniques. Recently, NDN-AI has been claimed to be one of the primary enablers of network and computer convergence. Other technologies include NDN embedded with IoT technology, such as NDN C-IoT, as a new IoT technology that is aimed at fast data retrieval. To obtain data, the IoT normally uses IP-based data delivery techniques. However, using IP-based ways to achieve NDN C-IoT is difficult, because it introduces a new data-centric paradigm with a distinct design and operation. Current IoT technology suffers from a severe cascading bottleneck that affects network performance; the solution to this issue was provided by Fu and Yang [22]. This study focuses on NDN-IoT to tackle issues that have not been solved by current IoT applications.

2.1. Named Data Networking

NDN presents a high-level overview of the concept and associated approaches, in which data are identified based on substance or content, rather than on geographical location. Distributed computing is in high demand, as new technologies, such as big data, data mining, and AI, emerge. Increasing numbers of articles are devoted to NDN. The next-generation Internet is built on the foundation of NDN, which is a typical representation and implementation of content-centric networking (CCN) [12]. A new architecture that can allow large-scale deployment of content-based distribution is needed with the increasing use of the Internet, which has shown the limitations of the TCP/IP protocol stack. NDN has emerged as a game-changing alternative approach and a possible option for next-generation Internet architecture. NDN combines several aspects for routing, security, and mobility.
As far as we know, very little research has considered multi-metric limitations on routing techniques. Several methods have been proposed for routing NDN, one of which is the ant colony optimization algorithm (ACO). The ACO has a better function than other traditional NDN routing techniques, in terms of bandwidth, cost, and delay [23]. As shown in Figure 2 the forwarding strategy (FS) is part of routing in NDN, to process content transmission in NDN communications. Another method, known as ant-colony-based extensible forwarding, was upgraded from the ACO strategy; it enhances sending I_packets by automatically assigning each attribute’s weight [24].
FS has been enhanced to improve quality of service (QoS) and to support data-centric networking in NDN [25]. This approach improved performance significantly. In a dynamically evolving network, current fixed rules have been set up inaccurately. FS, based on Q-learning, has implemented a continuous and online learning method that ensures a rapid response to network outages [26]. Caching strategies also provide several improvements from the previous strategy, using the leave copy everywhere (LCE) strategy. The NDN FS strategy has a function in which content is stored in the cache. However, the LCE strategy uses an algorithm that compares two levels of cache nodes and then takes the node degree into account, decreasing the average number of hit hops, the frequency of cache replenishment, and the throughput of content hits [27].
Another aspect of NDN that is being discussed in current research is handover performance. Network mobility (NeMO) [28] has solved the mobility issues on NDN, but no detailed discussion is available on evaluation performance, i.e., signaling cost and handover latency. Other methods have been proposed, e.g., using IPv6 on mobility and applying SINEMO architecture [29]. Although some problems with the NeMO method have been solved, problems such as low efficiency and delay caused by triangular routing, high handoff cost and latency, and a high level of packet droplet and signaling overhead have not been addressed.

2.2. Named Data Networking and Artificial Intelligence

The current TCP/IP network’s security, mobility, and other issues are addressed by NDN, which serves as a prototype for the future network by replacing the IP address with the name of the content. NDN is considered a promising new Internet paradigm to replace the TCP/IP hourglass model’s “thin waist.” A new solution has been obtained by merging NDN technology with AI. With its low power consumption and rapid speed, AI is ideal for IoT systems that require local real-time data analysis. Software implementation of noise distribution normalization [5] adds Gaussian distributed noise, resulting in differential privacy for edge AI. Traffic across the core network and inference latency can be reduced by deploying AI at the edge of networks that are in close enough proximity to the extraction of a large amount of information [30]. Several articles have embedded medical application systems on the IoT, including smartwatches, and smart HVAC systems, with the aim of reducing human effort [31]. A brief discussion was presented by Aman et al. [32], elaborating on the use of the Internet of Medical Things using AI technology. Analyzing local data in real time for the IoT systems is made possible by edge AI, which allows for a low-level operation while protecting user privacy. AI technology has improved data privacy access, based on a large number of data and mobility systems on NDN [33].
In the context of continuing network location changes, mobility refers to the capacity to maintain a unified connection with network entities in AI. Existing host-dimension solutions have exhausted all of the design options and are at a standstill when it comes to new performance advances. Trajectory-driven reachability updating and topology-driven intermediate placement have been proposed to improve overall mobility assistance performance [34]. Figure 3 shows an example of an NDN application that is embedded with AI mobility in e-health technology.
In terms of security, NDN data caching, which is a significant aspect of the technology, can be greatly reduced. Identifying a viable mitigation strategy when NDN routers are exploited by attackers is difficult. ACO [35] has been proposed for probing a safety transmission path and adjusting content retrieval, gathering information about all routers on a network, and bypassing bad routers, thereby preventing the bad routers from continuing to disseminate illegal content and minimizing substance contamination by removing bogus data packets from a cache store during the path traversal. A reactive attack uses content fetch time to track down the items that people are looking for. A proactive attack [36] caches an item before checking whether the victim requests that item.

2.3. Named Data Networking and Cloud Internet of Things

Cloud services are increasingly used in today’s networked and distributed applications. However, relying only on cloud services is not always the best option. This section focuses on cloud IoT applications that collaborate with NDN. As a new paradigm technology, NDN has evolved as an element of the future Internet, enabling content delivery, mobile, privacy, reliability, and access to data, regardless of geographic space. NDN has various functionalities, i.e., cache management, device-naming schemes, regulations for password protection, forwarding strategies, configuration management, and awareness of NDN/CCN in the IoT [37]. NDN has also evolved in various IoT applications. The IoT is a well-known idea that envisions the interconnection of several physical items to facilitate the transmission of data between those items. IoT networks need to be designed for performance in a heterogeneous environment with a high volume of data, for devices with limited resources, and with a high degree of mobility. NDN-based content-forwarding solutions have been proposed for usage in IoT applications in NDN-IoT scenarios, given proper identification [38].
NDN IoT applications have also been used to ensure that rapid disease spread can be handled more efficiently, using IoT applications for e-health. To address the limited resources of IoT devices and the time-sensitive nature of the data transmitted, the smart COVID-19 pandemic controlled eradication over NDN-IoT (SPICE-IT) method has been used to reduce the problem of network congestion and cache overflow [39]. The NDN secure remote health-monitoring system [40] is one of the e-health IoT applications that have been developed to increase online communication among doctors and patients. An e-health security application has been deployed to make sure NDN communication is secure [41]; this technology must be enhanced via NDN mobility technology. NDN IoT applications have also been applied in emergency response cases. Currently, all disaster scenarios apply in simulation-based situations only, but an NDN disaster-response system [42] has been developed to monitor the probability of a disaster. That work [42] came up with a new solution by applying edge infrastructure with cloud technology to achieve a fast response for the handover process, ensure a quick exchange of information, and reduce delays in transmission. Another work in disaster scenarios devised a new solution by implementing a connection-oriented surveillance system [43]. That work [43] improved the communication between the sender and the receiver by enhancing the transmission rate of interest packets requested by the consumer.
New NDN ideas, such as those pertaining to security, in-network caching, hierarchical naming or namespace, named content, and NBR, are used to meet IoT needs [38]. Some of the ideas being proposed in IoT technology [44] by using clustering-based intrusion detection for 5G have improved performance on IoT networks. Other aspects of energy-saving on 5G, especially the computing paradigm on green IoT, have been proposed by Hu et al. to improve communication overhead on the server cloud side [45]. This technique can be used for applications on the NDN C-IoT service. NDN is also compatible with 6G networks, because the sender and receiver are connected to each other. NDN has additional features, such as in-network caching and hop-by-hop transmission. Hu et al. proposed an efficient method that is suitable for NDN C-IoT implementation to improve energy in network computing, to overcome higher latency and signaling costs [45].
Over NDN, hydra [46] automatically provides scalable and efficient data distribution at large volumes and enables distributed control with great robustness. The traditional approach is commonly used in the majority of NDN clouds, in which developers need to communicate with a central server. GitSync [47] can function even in the event of a network partition or inconsistent connectivity, because it has no single point of failure. On the other side of the vehicular perspective, vehicle-to-everything (V2X) networking’s full potential can be realized, with several significant issues that must be resolved. The dual technique of a combined blockchain and NDN framework [48] was developed to secure V2X. The methodology of the NDN cloud is summarized in Table 1.

3. Mobility Methodology

KITE, the PMSS, and hybrid NeMo were selected due to the mobility support offered and the need for producer mobility technology. KITE uses the indirection method to transmit content using simple NDN communication; the PMSS improves producer operation by reducing handover latency; and hybrid NeMO provides a BIT to replace the FIB function by using the same cluster for indirection mobility. Thus, these three methods are comparable in measuring network performance for handover latency and signaling cost. On the basis of the NDN architecture, the ndnSIM simulator can be implemented on the network protocol stack framework. The simulator works with any link layer protocol, including point-to-point, CSMA, and wireless, as well as with the deployment of IPv4 and IPv6. The advantage of the ndnSIM simulator is that it can implement heterogeneous scenarios, such as NDN over IP. The ndnSIM simulator uses the C++ programming language to enable several modules in the NDN to operate, including PIT, FIB, content storage, network and application surface, and forwarding strategies. The use of this structured module enables any component to be modified.
NS3 has been patented, using the GNU GPLv2 license to reduce the writing of model procedures in simulations. The NS3 simulator uses the C++ library as its main program, as no graphics version is provided. NS3 can be used on Windows operating systems by adding MinGW modules. To capture the traffic log, the Wireshark app can be combined with the NS3 simulator. Several methods are running for the NDN testbed, especially on mobility perspectives, including KITE with C-IoT, PMSS with C-IoT, and hybrid NeMO with C-IoT, which is the latest technology. Figure 4 shows the setup for these three methods, which include a home router and a destination router; an AP that maintains the connection for mobile producers; and interaction between consumers with FIB to receive I_packets and D_packets.
Simulation is applied by researchers to implement new solution methods before testing them in real situations. Simulation methods can improve performance and meet performance expectations. Simulation combines elements of hardware and software. The combination of these elements is important, because running the simulation process can identify how some elements can communicate with each other and also identify the effect of the use of certain elements on the simulation process [59]. The network simulation process is unavoidable in the research article process, due to the use of open source model systems. The advantage of using open source systems is that researchers can conduct studies and find faults while the simulation process is in progress. With the use of network simulation using open source methods, various applications with different protocols can be run, such as FTP, TFTP, and Telnet. Network simulators can implement the methodological process for a study without implementing it in real situations. They can be implemented in different network topologies.
Various types of network simulators are available in today’s market. Different functions of each network simulator are used, depending on the accuracy of the parameters, i.e., the difficulty level, the number of nodes, the type of traffic on the nodes, the level of CPU usage, the memory usage, and the time rate calculation for the protocol used for a study. Several studies have been conducted on network simulators, such as NS3 [60], OMNET ++ [61], QualNet [62], and NetSIM [63]. Some articles on NDN simulators start with a basic setup to design a testbed for a preliminary test of NS3 [64].

3.1. KITE of NDN with C-IoT and AI

KITE technology with cloud operation uses a server called the rendezvous server (RS), which is located between the consumer and the mobility producer that uses IoT devices. The RS can always be traced by an NDN producer known as KITE. As shown in Figure 5, KITE operates between transmission I_packets from the consumer via accessing the producer by storing the location through the RS. A single route prefix is assumed for simplicity. KITE swaps the public keys to create confidence. It is signed by the mobile producer and has a special trace tag in the name, containing AI technology. The RS acknowledges the data interest for the location of the producer and returns signed tracing data to the mobile producer via the same path as the TI. Intermediary forwarders establish or update FIB entries for the mobile producer data prefix. When a mobile producer moves to other locations, it stores all of the prefixes via cloud storage for any interest packet update. Table 2 explains the operation of KITE technology in cloud storage.

3.2. The PMSS of NDN with C-IoT and AI

Several processes take place while using the PMSS with cloud to enhance producer mobility, as shown in Figure 6. The SR element is added to maintain data packet transmission with add-on cloud storage for IoT implementation and AI technology for processing data at mobile producers that use IoT devices as a transmission mechanism. Table 3 shows the details of the transmission of the PMSS.

3.3. Hybrid of NDN with C-IoT and AI

For hybrid NeMO with cloud operation, as shown in Figure 7, the mobile producer moves from the current location under agent router 1 (AR 1) to the newest location on AR 2. The processing of data in the mobile producer using this hybrid technology uses AI to speed up the transmission process. During this movement, all information of the data packet and the interest packet is currently updated through FIB, but in this technology, BIT reduces the functionality of FIB. AR 1 and AR 2 interact with each other on the NDN network cloud. After BIT has been updated at AR 2, it sends the latest information to MR 2 and, finally, to the mobile producer. After this operation, which is called interNeMO, the consumer obtains the data packet successfully. Table 4 explains the operation of hybrid NeMo in cloud storage.

4. Performance Comparison

As discussed in Section 3, each technology has its own advantages and a variety of operations. All this technology varies their levels of network performance, such as handover latency, signaling cost, and overhead cost.
This section will focus on handover latency and signaling cost on the perspective mobility of NDN. Each operation producer has three well-known methods for mobility: KITE, which is based on indirection technology; the PMSS, which is an evolution of the KITE operation; and hybrid NeMO, which is the newest method for NDN mobility. Handover delay, i.e., the time taken by a mobile producer that has received I_packets before disconnecting from their current connection, continues as new I_packets arrive at a new location. Handover latency is good when the value is smaller during the content transmission. Signaling cost, i.e., the measurement of the number of transmission messages from the node producer to the node consumer from NDN network communication, is good when the value is smaller while the content transmission is in progress.
To confirm this statement, a mathematical equation for KITE, the PMSS, and hybrid NeMO was designed to generate a formula for each network’s performance. For network analysis, ndnSIM was used to compare the performance evaluation for each method. For further setup in ndnSIM, several parameters were defined for each method to measure the performance and impact of mobility on each producer mobility technique. This setup measured the quality of network performance via each method by varying the number of parameters. Each topology used the same network grid of 400 m × 400 m; the range from the NDN router to AP was 100 m; the number of producers and mobility speeds varied for each method; the I_packets’ transmission range used 100 ms; and the same benchmark reference as that of the random waypoint mobility model was applied. Table 5 shows the details of how the parameters were set, including the size of the network, the speed of mobility, the mobility model used, and the simulation software on each testbed for different NDN mobility perspectives. Figure 8 shows the number of nodes for the consumer and the producer, which was used to measure the handover latency and signaling cost.

4.1. Signaling Cost

The signaling cost is the total traffic demand encountered while transmitting signaling messages during content transmission. According to [65], the handover rate is defined as the likelihood that the user crosses over to the next cell within one movement period. For this purpose, we focused on several messages that can be sent through the network during handover transmission. A low signaling cost corresponds to better performance of the proposed methodology.
  • Signaling cost for KITE operation
Equation (1) describes the signaling cost for the KITE operation, while Table 6 shows the parameter setup.
SCKite = CprodAR + Caccessrouter + Caccessrouter + CprodAR + CconsumerAR + Caccessrouter + Caccessrouter+ CprodAR
= SInterest/data × (2a + 2c) + SInterest × (2a + 2c)
  • Signaling cost for PMSS operation
Equation (2) describes the signaling cost for the PMSS operation, while Table 6 shows the parameter setup.
SCPMSS = CprodAR + Caccessrouter + CconsumerAR + Caccessrouter + Caccessrouter+ CprodAR
= SmobilityInterest × (a + 2c) + SInterest × (a + c)
  • Signaling cost for hybrid NeMO operation
Equation (3) describes the signaling cost for the hybrid NeMO operation, while Table 6 shows the parameter setup.
SChybrid = CprodAR + Caccessrouter + CconsumerAR + CprodAR
= SmobilityInterest × (a + c) + SInterest × (a + c)
  • Average signaling cost for each mobility operation
Equation (4) describes the average signaling cost for each method.
SCAVERAGE = (C10 + C20 + C30 + C40 +C50 + C60 +C70 + C80 + C90 + C100)/10

4.2. Handover Latency

Handover, or handoff, is a method of mobile communication in which cellular transmission (voice or data) is transferred from one base station (cell site) to another without compromising the cellular transmission link. Handover is an important part of mobile transmission because it sets up data sessions or phone calls between always-mobile devices. Handover latency and failure are major problems in mobile data networks that slow down and stop service. We need lower latency to achieve better performance. Equation (5) refers to [66] to measure the speed of latency by calculating the average hop count of content transmission with a variety of mobile producer speeds.
  • Handover latency equation for KITE operation
Equation (5) describes the handover latency for the KITE operation, while Table 6 shows the parameter setup.
HLkite = ((TpTp + (l/v)) + (1 − (TpTp + (l/v)))) × (lpn + a × Lwlup + 2d × LwtrcgInt + a × LwltrcdInt + 2d × LwtrcdIntA2)
  • Handover latency equation for PMSS operation
Equation (6) describes the handover latency for the PMSS operation, while Table 6 shows the parameter setup.
HLpmss = ((TpTp + (l/v)) + (1 − (TpTp + (l/v)))) × (lpn + a × LwlmobInt + 2c × LwlmobInt + a × LWlint + c × LWintA2)
  • Handover latency equation for hybrid NeMO operation
Equation (7) describes the handover latency for hybrid NeMO operation, while Table 6 shows the parameter setup.
HLHybrid = ((TpTp + (l/v)) + (1 − (TpTp + (l/v)))) × (Lpn + a × Lwlhybrid + 2c × Lwlhybrid + a × Lwlint + c × LwintA2)
  • Average handover latency equation for each mobility operation
Equation (8) describes the average handover latency for each method.
HLAVERAGE = (HL10 + HL20 + HL30 + HL40 + HL50 + HL60 + HL70 + HL80 + HL90 + HL100)/10

5. Discussion and Analysis

This section discusses the performance evaluation for KITE, the PMSS, and hybrid NeMO, based on an IoT element embedded in the technology. Two types of performance evaluation were compared: signaling cost and handover latency.

5.1. Signaling Cost

Figure 9 shows how the signaling cost varied with mobile producer speed, starting from 0 ms to 100 ms after the simulation was run using the NS3 software based on the parameter sets.
The results were encouraging, despite the broadcasting nature of the three methods for the mobile producer and the domain restriction application. KITE and the PMSS had minimal handoff signaling costs. However, the handoff signaling cost of hybrid NeMO was nearly equal to or less than that of KITE and the PMSS, with about 4% of signaling cost, compared with those of KITE and the PMSS. The result of hybrid NeMO was good, even though no significant improvement against KITE and the PMSS was found at certain mobile producer speeds, such as 30 to 50 m/s, 50 to 60 m/s, and 70 to 100 m/s.
With regard to hop-count performance, Figure 10 shows that the KITE operation sent a greater number of messages than the PMSS or hybrid NeMO did. When the speed increased, the hybrid operation increased the number of messages to be sent. Therefore, mobility speed did not affect the performance of the transmission data packet and the interest packet to the consumer, because of the stability of BIT. Message drop was also encountered, but message delay generally happened during transmission.
Figure 10 shows that the hybrid method had the highest number of transmitted messages, compared with the PMSS and KITE. Even though a recent study stated that the PMSS and KITE have minimal handover signaling overhead during transmission, compared with anchor and anchorless producer mobility methods, the performance of the hybrid method in terms of signaling cost was more efficient than that of these other two methods.

5.2. Handover Latency

Figure 11 shows the handover latency versus variation in mobile producer speed from 0 ms to 100 ms after the simulation was run using the NS3 software based on the parameter set.
The handover latency of all schemes continued to increase, due to the increase in interest arrival rate. In addition, for the hybrid NeMO scheme, the handover latency was higher than that of KITE and the PMSS, because of the fast handover messages sent by the mobile producer as a pre-process of handoff. Moreover, the signaling cost of KITE was between that of hybrid NeMO and the PMSS, while hybrid NeMo had the lowest handover latency of 46% compared with KITE and 60% compared with the PMSS.
On the basis of a mathematical equations that were constructed for cost signaling and handover latency, a simulation was conducted by using the NS3 simulator to generate network performance with the use of a certain parameter. The techniques were compared to determine the hop count over speed. Figure 12 shows that while the mobile process increased the speed, the hop count to reach data for hybrid NeMo was less than that for the PMSS and KITE, because BIT decreased the usage of FIB, thereby reducing the hop count. This feature was an advantage of hybrid NeMO over the other two methods.
Handover latency can be measured by the time a mobile producer requests I_packets that were received from the last content router and the transmission of new I_packets after the handover operation. When more content needs to be traversed, the latency is increased because of the long distance from the consumer to the producer. The hop count between the consumer and the producer was examined. Figure 12 shows that when the speed of mobile producers increased, the hybrid method reduced the number of hops from the consumer to the producer. The average hop count for hybrid NeMo was, at most, 4 per hop, compared with the average hop count for the PMSS, which was 10 per hop; KITE had a maximum of 8 per hop while the mobile producer speed increased. Hybrid NeMO showed that having a lower average hop count could reduce handover latency, thus resolving the issues of path stretching for KITE.

6. Conclusions

This study adds comprehensively to current research on the merging of NDN, AI, and C-IoT technologies. This paper’s main contribution is the combination of NDN-AI-IoT. It focused on three selected methodological explanations for KITE, the PMSS, and hybrid NeMO. NDNs, rather than being simply an Internet technology, serve a variety of functions, with a focus on consumer-driven network architecture. KITE uses the indirection method to transmit content via simple NDN communication; the PMSS improves producer operations by lowering handover latency; and hybrid NeMO provides a binding information table to replace the base function of forwarding information. In addition, mathematical equations for signaling cost and handover latency were described in this study. The producer mobility operation was highlighted using the network simulator ndnSIM NS-3. Based on the mobility scenario, mathematical equations were developed for each methodology to measure handover latency and signaling cost. We concluded that the hybrid NeMO performs better than KITE and PMSS in terms of handover latency and signaling cost. In the future, this work will be expanded to include additional parameters with a larger network to support scalability and multiple environments.

Author Contributions

Supervision: A.H.M.A.; validation: R.H. and N.M.; visualization and writing of original draft: W.M.H.A.; review and editing: A.H.M.A. and R.H. All authors have read and agreed to the published version of the manuscript.

Funding

This article was supported under the Fundamental Research Grant Scheme FRGS/1/2019/ICT03/UKM/02/1.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Acknowledgments

The authors would like to acknowledge the support of the Network Communication Lab Technology Research Groups, FTSM, and UKM in providing facilities for this research. This article was supported under the Fundamental Research Grant Scheme FRGS/1/2019/ICT03/UKM/02/1.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Hou, S.; Hu, Y.; Tian, L. A Survey on Geographic location routing and forwarding strategies of VANET and WSN based on NDN. J. Phys. Conf. Ser. 2021, 1738, 012040. [Google Scholar] [CrossRef]
  2. Hao, B.; Wang, G.; Zhang, M.; Zhu, J.; Xing, L.; Wu, Q. Stochastic Adaptive Forwarding Strategy Based on Deep Reinforcement Learning for Secure Mobile Video Communications in NDN. Secur. Commun. Netw. 2021, 2021, 6630717. [Google Scholar] [CrossRef]
  3. Gündoğan, C.; Kietzmann, P.; Schmidt, T.C.; Wählisch, M. A mobility-compliant publish–subscribe system for an information-centric Internet of Things. Comput. Netw. 2022, 203, 108656. [Google Scholar] [CrossRef]
  4. Mohd Aman, A.H.; Yadegaridehkordi, E.; Attarbashi, Z.S.; Hassan, R.; Park, Y.J. A Survey on Trend and Classification of Internet of Things Reviews. IEEE Access 2020, 8, 111763–111782. [Google Scholar] [CrossRef]
  5. Fu, J.; Liao, Z.; Liu, J.; Smith, S.C.; Wang, J. Memristor-Based Variation-Enabled Differentially Private Learning Systems for Edge Computing in IoT. IEEE Internet Things J. 2021, 8, 9672–9692. [Google Scholar] [CrossRef]
  6. Karim, F.A.; Mohd Aman, A.H.; Hassan, R.; Nisar, K. A Survey on Information-Centric Networking with Cloud Internet of Things and Artificial Intelligence. Wirel. Commun. Mob. Comput. 2022, 2022, 7818712. [Google Scholar] [CrossRef]
  7. Karim, F.A.; Mohd Aman, A.H.; Hassan, R.; Nisar, K.; Uddin, M. Named Data Networking: A Survey on Routing Strategies. IEEE Access 2022, 10, 90254–90270. [Google Scholar] [CrossRef]
  8. Azamuddin, W.M.H.; Aman, A.H.M.; Hassan, R.; Abdali, T.A.N. Named Data Networking Mobility: A Survey. In Emerging Technology Trends in Internet of Things and Computing. TIOTC 2021. Communications in Computer and Information Science; Liatsis, P., Hussain, A., Mostafa, S.A., Al-Jumeily, D., Eds.; Springer: Cham, Switzerland, 2022; Volume 1548. [Google Scholar] [CrossRef]
  9. Zhang, Y.; Xia, Z.; Mastorakis, S.; Zhang, L. KITE. In Proceedings of the 5th ACM Conference on Information-Centric Networking, Boston, MA, USA, 21–23 September 2018; pp. 125–136. [Google Scholar] [CrossRef]
  10. Hussaini, M.; Nor, S.A.; Ahmad, A. PMSS: Producer Mobility Support Scheme Optimization with RWP Mobility Model in Named Data Networking. Int. J. Commun. Netw. Inf. Secur. 2018, 10, 329–339. Available online: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85051811566&partnerID=40&md5=2e294ec148f4d3dff72c8096c425022e (accessed on 1 March 2022). [CrossRef]
  11. Yan, Z.; Park, Y.; Leau, Y.; Ren-Ting, L.; Hassan, R. Hybrid Network Mobility Support in Named Data Networking. In Proceedings of the 2020 International Conference on Information Networking (ICOIN), Barcelona, Spain, 7–10 January 2020; pp. 16–19. [Google Scholar] [CrossRef]
  12. Saxena, D.; Raychoudhury, V.; Suri, N.; Becker, C.; Cao, J. Named Data Networking: A survey. Comput. Sci. Rev. 2016, 19, 15–55. [Google Scholar] [CrossRef]
  13. Sinky, H.; Khalfi, B.; Hamdaoui, B.; Rayes, A. Responsive Content-Centric Delivery in Large Urban Communication Networks: A LinkNYC Use-Case. IEEE Trans. Wirel. Commun. 2018, 17, 1688–1699. [Google Scholar] [CrossRef]
  14. Zhu, Z.; Afanasyev, A.; Zhang, L. A New Perspective on Mobility Support. NDN Tech. Rep. 2013, 13, 1–6. Available online: http://new.named-data.net/wp-content/uploads/TRmobility.pdf (accessed on 12 January 2022).
  15. Detti, A.; Bracciale, L.; Loreti, P.; Rossi, G.; Blefari Melazzi, N. A cluster-based scalable router for information centric networks. Comput. Netw. 2018, 142, 24–32. [Google Scholar] [CrossRef]
  16. Farahat, H. Proactive Caching to Support Mobility in Named Data Networks. Ph.D. Thesis, Queen’s University, Kingston, ON, Canada, June 2017. [Google Scholar]
  17. Torres, J.V.; Alvarenga, I.D.; Boutaba, R.; Duarte, O.C.M.B. Evaluating CRoS-NDN: A comparative performance analysis of a controller-based routing scheme for named-data networking. J. Internet Serv. Appl. 2019, 10, 20. [Google Scholar] [CrossRef]
  18. Carofiglio, G. Mobile Video Delivery with Hybrid ICN IP-Integrated ICN Solution for 5G. Available online: https://www.cisco.com/c/dam/en/us/solutions/collateral/service-provider/ultra-services-platform/mwc17-hicn-video-wp.pdf (accessed on 15 January 2022).
  19. Hu, N.; Tian, Z.; Sun, Y.; Yin, L.; Zhao, B.; Du, X.; Guizani, N. Building Agile and Resilient UAV Networks Based on SDN and Blockchain. IEEE Netw. 2021, 35, 57–63. [Google Scholar] [CrossRef]
  20. Choi, J.H.; Cha, J.H.; Han, Y.H.; Min, S.G. A dual-connectivity mobility link service for producer mobility in the named data networking. Sensors 2020, 20, 4859. [Google Scholar] [CrossRef]
  21. Rehman, M.A.U.; Ullah, R.; Kim, B.S. NINQ: Name-integrated query framework for named-data networking of things. Sensors 2019, 19, 2906. [Google Scholar] [CrossRef]
  22. Fu, X.; Yang, Y. Modeling and analyzing cascading failures for Internet of Things. Inf. Sci. 2021, 545, 753–770. [Google Scholar] [CrossRef]
  23. Zheng, Y.; Han, M.; He, L.; Li, Y.; Xing, G.; Hou, R. A QoS-supported Multi-constrained Routing Strategy Based on Ant-Colony Optimization for Named Data Networking. In Proceedings of the 2018 1st IEEE International Conference on Hot Information-Centric Networking (HotICN), Shenzhen, China, 15–17 August 2018; pp. 18–23. [Google Scholar] [CrossRef]
  24. Zhang, Y.; Wang, J.; Gong, L.; Yuan, J.; Lei, K. A quantified forwarding strategy in NDN by integrating ant colony optimization into MADM. In Proceedings of the 2017 IEEE International Symposium on Parallel and Distributed Processing with Applications and 2017 IEEE International Conference on Ubiquitous Computing and Communications (ISPA/IUCC), Guangzhou, China, 12–15 December 2017; pp. 214–220. [Google Scholar] [CrossRef]
  25. Rotinsulu, M.R.; Susilo, B.; Presekal, A.; Pramono, E.; Sari, R.F. Measuring Quality of Services (QoS) of several forwarding strategies on Named Data Networking(NDN) using ndnSIM. In Proceedings of the 2017 IEEE International Conference on Cybernetics and Computational Intelligence (CyberneticsCom), Phuket, Thailand, 20–22 November 2017; pp. 45–49. [Google Scholar] [CrossRef]
  26. Martins, V.F.; Tomczyk, Ł.; Amato, C.; Eliseo, M.A.; Oyelere, S.S.; Akyar, Ö.Y.; Motz, R.; Barros, G.; Sarmiento, S.M.A.; Silveira, I.F. A Smart Ecosystem for Learning and Inclusion: An Architectural Overview; LNCS; Springer International Publishing: Berlin/Heidelberg, Germany, 2020; Volume 12249, ISBN 9783030587987. [Google Scholar]
  27. Sun, X. Advances in Artificial Intelligence and Security. In Proceedings of the 7th International Conference, Dublin, Ireland, 19–23 July 2021. [Google Scholar]
  28. Islam, S.; Abdalla, A.H.; Hasan, M.K. Novel multihoming-based flow mobility scheme for proxy NEMO environment: A numerical approach to analyse handoff performance. ScienceAsia 2017, 43, 27–34. [Google Scholar] [CrossRef]
  29. Rahman, M.S.; Bouidel, O.; Atiquzzaman, M.; Ivancic, W. Performance comparison between nemo bsp and sinemo. In Proceedings of the 2008 IEEE Global Telecommunications Conference, New Orleans, LA, USA, 30 November–4 December 2008; pp. 1–5. [Google Scholar] [CrossRef]
  30. Alfredo, L.; Gennaro, G.; Piro, G.; Jararweh, Y.; Eds, C.C.; Goos, G. Ad-Hoc, Mobile, and Wireless Networks; Springer: Berlin/Heidelberg, Germany, 2020. [Google Scholar]
  31. Mohd Aman, A.H.; Hassan, W.H.; Sameen, S.; Attarbashi, Z.S.; Alizadeh, M.; Latiff, L.A. IoMT amid COVID-19 pandemic: Application, architecture, technology, and security. J. Netw. Comput. Appl. 2021, 174, 102886. [Google Scholar] [CrossRef]
  32. Ali, E.S.; Hasan, M.K.; Hassan, R.; Saeed, R.A.; Hassan, M.B.; Islam, S.; Nafi, N.S.; Bevinakoppa, S. Machine Learning Technologies for Secure Vehicular Communication in Internet of Vehicles: Recent Advances and Applications. Secur. Commun. Netw. 2021, 2021, 8868355. [Google Scholar] [CrossRef]
  33. Zhang, Y. Artificial Intelligence and Cognitive Computing for Communications and Networks. IEEE Access 2018, 8, 144105–144111. [Google Scholar] [CrossRef]
  34. Langa, C.; Tarwireyi, P.; Adigun, M. Evaluating Named Data Networking forwarding strategies in different loT topologies. In Proceedings of the 2020 International Conference on Artificial Intelligence, Big Data, Computing and Data Communication Systems (icABCD), Durban, South Africa, 6–7 August 2020. [Google Scholar] [CrossRef]
  35. Cui, W.; Li, Y.; Zhang, Y.; Liu, C.; Zhan, M. An ant colony algorithm based content poisoning mitigation in named data networking. In Proceedings of the 2019 18th IEEE International Conference On Trust, Security And Privacy In Computing And Communications/13th IEEE International Conference On Big Data Science And Engineering (TrustCom/BigDataSE), Rotorua, New Zealand, 5–8 August 2019; pp. 176–183. [Google Scholar] [CrossRef]
  36. Compagno, A.; Conti, M.; Losiouk, E.; Tsudik, G.; Valle, S. A Proactive Cache Privacy Attack on NDN. In Proceedings of the NOMS 2020–2020 IEEE/IFIP Network Operations and Management Symposium, Budapest, Hungary, 20–24 April 2020. [Google Scholar] [CrossRef]
  37. Aboodi, A.; Wan, T.C.; Sodhy, G.C. Survey on the Incorporation of NDN/CCN in IoT. IEEE Access 2019, 7, 71827–71858. [Google Scholar] [CrossRef]
  38. Ahed, K.; Benamar, M.; El Ouazzani, R. Content delivery in named data networking based internet of things. In Proceedings of the 2019 15th International Wireless Communications & Mobile Computing Conference (IWCMC), Tangier, Morocco, 24–28 June 2019; pp. 1397–1402. [Google Scholar] [CrossRef]
  39. Khan, M.T.R.; Saad, M.M.; Tariq, M.A.; Akram, J.; Kim, D. SPICE-IT: Smart COVID-19 pandemic controlled eradication over NDN-IoT. Inf. Fusion 2021, 74, 50–64. [Google Scholar] [CrossRef]
  40. Kar, P.; Dong, Y.; Li, Y.; Ma, X.; Ding, X.; Wang, Y. NDN Based Plug-n-Play and Secure Remote Health Monitoring System. In Proceedings of the ICC 2021-IEEE International Conference on Communications, Montreal, QC, Canada, 14–23 June 2021. [Google Scholar] [CrossRef]
  41. Hasan, M.K.; Islam, S.; Sulaiman, R.; Khan, S.; Hashim, A.H.A.; Habib, S.; Islam, M.; Alyahya, S.; Ahmed, M.M.; Kamil, S.; et al. Lightweight Encryption Technique to Enhance Medical Image Security on Internet of Medical Things Applications. IEEE Access 2021, 9, 47731–47742. [Google Scholar] [CrossRef]
  42. Tran, M.N.; Kim, Y. Named data networking based disaster response support system over edge computing infrastructure. Electronics 2021, 10, 335. [Google Scholar] [CrossRef]
  43. Okamoto, K.; Mochida, T.; Nozaki, D.; Wen, Z.; Qi, X.; Sato, T. Content-Oriented Surveillance System Based on ICN in Disaster Scenarios. In Proceedings of the 2018 21st International Symposium on Wireless Personal Multimedia Communications (WPMC), Chiang Rai, Thailand, 25–28 November 2018; pp. 484–489. [Google Scholar] [CrossRef]
  44. Hu, N.; Tian, Z.; Lu, H.; Du, X.; Guizani, M. A multiple-kernel clustering based intrusion detection scheme for 5G and IoT networks. Int. J. Mach. Learn. Cybern. 2021, 12, 3129–3144. [Google Scholar] [CrossRef]
  45. Hu, N.; Tian, Z.; Du, X.; Guizani, M. An Energy-Efficient In-Network Computing. IEEE Trans. Green Commun. Netw. 2021, 5, 1722–1733. [Google Scholar] [CrossRef]
  46. Liu, S.; Patil, V.; Yu, T.; Afanasyev, A.; Feltus, F.A.; Shannigrahi, S.; Zhang, L. Designing hydra with centralized versus decentralized control: A comparative study. In Proceedings of the Interdisciplinary Workshop on (de) Centralization in the Internet, Virtual Event, Germany, 7 December 2021; pp. 4–10. [Google Scholar] [CrossRef]
  47. Ma, X.; Zhang, L. GitSync: Distributed version control system on NDN. In Proceedings of the 8th ACM Conference on Information-Centric Networking, Paris, France, 22–24 September 2021; pp. 121–123. [Google Scholar] [CrossRef]
  48. Rawat, D.B.; Doku, R.; Adebayo, A.; Bajracharya, C.; Kamhoua, C. Blockchain Enabled Named Data Networking for Secure Vehicle-to-Everything Communications. IEEE Netw. 2020, 34, 185–189. [Google Scholar] [CrossRef]
  49. Baldoni, G.; Loudet, J.; Cominardi, L.; Corsaro, A.; He, Y. Facilitating distributed data-flow programming with Eclipse Zenoh: The ERDOS case. In Proceedings of the 1st Workshop on Serverless mobile networking for 6G Communications, WI, USA, 25 June 2021; pp. 13–14. [Google Scholar] [CrossRef]
  50. Wang, X.; Cai, S. Named Data Networking-Based Macrovehicular Cloud Framework. IEEE Syst. J. 2021, 15, 5646–5653. [Google Scholar] [CrossRef]
  51. Ogle, C.; Reddick, D.; McKnight, C.; Biggs, T.; Pauly, R.; Ficklin, S.P.; Feltus, F.A.; Shannigrahi, S. Named Data Networking for Genomics Data Management and Integrated Workflows. Front. Big Data 2021, 4, 1–16. [Google Scholar] [CrossRef]
  52. Wang, X.; Wang, X.; Wang, D. Cost-efficient data retrieval based on integration of VC and NDN. IEEE Trans. Veh. Technol. 2021, 70, 967–976. [Google Scholar] [CrossRef]
  53. Neto, R.C.J.; Merindol, P.; Theoleyre, F. Transformation Based Routing Overlay for Privacy and Reusability in Multi-Domain IoT. In Proceedings of the 2020 IEEE 19th International Symposium on Network Computing and Applications (NCA), Cambridge, MA, USA, 24–27 November 2020. [Google Scholar] [CrossRef]
  54. Wilhelm, G.; Fouchal, H.; Ayaida, M. Vehicular Context Cloud Networking for Intelligent Transport Systems. In Proceedings of the ICC 2020-2020 IEEE International Conference on Communications (ICC), Dublin, Ireland, 7–11 June 2020; Volume 2020. [Google Scholar] [CrossRef]
  55. Wang, X.; Cai, S. Secure healthcare monitoring framework integrating NDN-based IoT with edge cloud. Futur. Gener. Comput. Syst. 2020, 112, 320–329. [Google Scholar] [CrossRef]
  56. Ajmi, N.; Helali, A.; Hafnaoui, T.; Mghaieth, R. A virtual sensor network for (WSNs): Architecture and routing. In Proceedings of the 2020 IEEE International Conference on Design & Test of Integrated Micro & Nano-Systems (DTS), Hammamet, Tunisia, 7–10 June 2020. [Google Scholar] [CrossRef]
  57. Shannigrahi, S.; Mastorakis, S.; Ortega, F.R. Next-generation networking and edge computing for mixed reality real-time interactive systems. In Proceedings of the 2020 IEEE International Conference on Communications Workshops (ICC Workshops), Dublin, Ireland, 7–11 June 2020; pp. 1–6. [Google Scholar] [CrossRef]
  58. Park, M.; Kim, J.; Kim, Y.; Cho, E.; Park, S.; Sohn, S.; Kang, M.; Kwon, T.T. An SGX-Based Key Management Framework for Data Centric Networking. IEEE Access 2020, 8, 45198–45210. [Google Scholar] [CrossRef]
  59. Chaudhary, R.; Sethi, S.; Keshari, R.; Goel, S. A study of comparison of Network Simulator -3 and Network Simulator -2. Int. J. Comput. Sci. Inf. Technol. 2012, 3, 3085–3092. [Google Scholar]
  60. Afanasyev, A.; Moiseenko, I.; Zhang, L. ndnSIM: NDN simulator for NS-3. In NDN, Tech. Rep. NDN-0005; Revision 2, October 2012; pp. 1–7. Available online: http://named-data.net/techreport/TR005-ndnsim.pdf (accessed on 1 March 2022).
  61. Varga, A. HURRAY-TR-070903. In REMPLI Discrete Event Simulation System; Version 0, 20 September 2007; pp. 1–15. Available online: http://cister-labs.pt/docs/rempli_discrete_event_simulation_system/401/view.pdf (accessed on 7 April 2022).
  62. Dinesh, S. Goyal Sonal Qualnet simulator. Int. J. Inf. Comput. Technol. 2014, 4, 1349–1354. [Google Scholar]
  63. Iii, L.B.; Barnett, B.L. (TR-92-2). In NetSim: A Network Performance Simulator Netsim: A Network Performance Simulator; Math and Computer Science Technical Report Series; June 1992; pp. 1–11. Available online: https://core.ac.uk/download/pdf/232768405.pdf (accessed on 14 April 2022).
  64. Ching, T.W. Performance Analysis of Internet of Things Routing Protocol: Energy, Overhead and Paket Delivery. In Proceedings of the 2021 3rd International Cyber Resilience Conference (CRC), Langkawi Island, Malaysia, 9–31 January 2021. [Google Scholar]
  65. Sadr, S.; Adve, R.S. Handoff rate and coverage analysis in multi-tier heterogeneous networks. IEEE Trans. Wirel. Commun. 2015, 14, 2626–2638. [Google Scholar] [CrossRef]
  66. Yan, Z.; Zeadally, S.; Zhang, S.; Guo, R.; Park, Y.-J. Distributed mobility management in named data networking. Wirel. Commun. Mob. Comput. 2016, 16, 1773–1783. [Google Scholar] [CrossRef]
Sensors 22 06668 g001 550
Figure 1. Comparison of the OSI model and the NDN model.
Figure 1. Comparison of the OSI model and the NDN model.
Sensors 22 06668 g001
Sensors 22 06668 g002 550
Figure 2. NDN forwarding strategies.
Figure 2. NDN forwarding strategies.
Sensors 22 06668 g002
Sensors 22 06668 g003 550
Figure 3. NDN AI mobility using smart technology.
Figure 3. NDN AI mobility using smart technology.
Sensors 22 06668 g003
Sensors 22 06668 g004 550
Figure 4. General NDN simulation scenario.
Figure 4. General NDN simulation scenario.
Sensors 22 06668 g004
Sensors 22 06668 g005 550
Figure 5. KITE operation with cloud storage.
Figure 5. KITE operation with cloud storage.
Sensors 22 06668 g005
Sensors 22 06668 g006 550
Figure 6. The PMSS with cloud operation.
Figure 6. The PMSS with cloud operation.
Sensors 22 06668 g006
Sensors 22 06668 g007 550
Figure 7. Hybrid NeMO with cloud operation.
Figure 7. Hybrid NeMO with cloud operation.
Sensors 22 06668 g007
Sensors 22 06668 g008 550
Figure 8. Network topology setup for KITE, the PMSS, and hybrid NeMO.
Figure 8. Network topology setup for KITE, the PMSS, and hybrid NeMO.
Sensors 22 06668 g008
Sensors 22 06668 g009 550
Figure 9. Signaling cost versus speed.
Figure 9. Signaling cost versus speed.
Sensors 22 06668 g009
Sensors 22 06668 g010 550
Figure 10. Average signaling by each method.
Figure 10. Average signaling by each method.
Sensors 22 06668 g010
Sensors 22 06668 g011 550
Figure 11. Handover latency (hop count) versus speed.
Figure 11. Handover latency (hop count) versus speed.
Sensors 22 06668 g011
Sensors 22 06668 g012 550
Figure 12. Average hop count by each method.
Figure 12. Average hop count by each method.
Sensors 22 06668 g012
Table
Table 1. NDN cloud application.
Table 1. NDN cloud application.
ReferenceMethodDescription
[49]ERDOSIntegration of edge-native data flow and edge computing.
[50]NDN-based aVC framework (NVCF)Improve aVC data gathering success rates and lower the cost of aVC data retrieval.
[51]NDN GenomicsWith in-network caching of widely used datasets, NDN for genomic information operations improves data insights, accelerates extraction leveraging ubiquitous resources, and enables society to create architecture.
[52]VC NDNBy using the NDN’s benefits, it provides cost-effective and value-based data retrieval.
[53]NDN Routing overlayIn addition to reducing network and cache burden, NDN CSs that offer data aggregation and transformation also enforce privacy naturally.
[54]VCCNVehicle networks have the potential to spread new data via multi-hop.
[55]NDN IoT EdgeCipher texts and signatures are used to ensure the security of medical data transport and the advantages of NDN.
[56]VSN on NDNRouting data between virtual sensors is a solution to the current paradigm.
[57]NDN Mixed Reality Real-TimeBased on NDN, an AR/VR computational architecture that potentially addresses these issues is developed by utilizing a hybrid edge-cloud model.
[58]NDN SGX-BasedData access control keys are distributed and maintained efficiently and flexibly.
Table
Table 2. KITE operation.
Table 2. KITE operation.
Flow of TransmissionProcess StatusDescription
Process 1Before handoffTypically, a consumer sends an I_packet over an NDN router to the network to request data. The NDN router then determines if the material is available; if not, then it sends an I_packet to the NDN network.
Process 2Initiate connection with a content routerThe I_packet’s prefix data name traverses NDN routers to reach the location of the producer. If the data are cached by any router in the network, then the router reacts promptly with the cached data. If not, then the routers, along with the path, store the interest information as entries in the PIT and FIB tables, then forward it until it reaches the producer. The producer then provides the data in breadcrumb format to the consumer.
Process 3Handoff startedThe producer abruptly decides to switch from the old PoA to the new PoA. After the connection, a new content name prefix is generated, and the producer prepares to notify the anchor router of the new name.
Process 4Update RSThe producer overwhelms the network with trace I_packets destined for the immobile anchor router or the RS to notify it of the new name prefix.
Process 5Establish data trace from producer to NDN routerThrough intermediate routers, the anchor router or RS responds with trace data packets and establishes a trace between the mobile producer and the anchor router or RS.
Process 6Store PIT in RSThe consumer saves PIT in the NDN router or RS.
Process 7After the handoff operationThe NDN router forwards the consumer I_packet through data tracking at the mobile producer.
Process 8Producer acknowledgesThe mobile producer replies to the D_packet to the consumer through the NDN router.
Table
Table 3. The PMSS’s operation.
Table 3. The PMSS’s operation.
Flow of TransmissionProcess StatusDescription
Process 1Before transmissionBefore handover transmission between the producer and the consumer, consumers transmit I_packets for requesting data to the mobile producer. The mobile producer checks the available content with SR1. If the content is not available, then it searches with other SRs in their neighborhood.
Process 2BroadcastInterest packet time is the time to retrieve the requested content and the time it takes to send it from the source to the destination.
Process 3Processing dataThe I_packet goes through the nearby RS until the mobile producer is reached. Processing data from the consumer to the mobile producer is similar to the processing of data in KITE operation behavior. While the handover process starts, the mobile producer cuts off the connection with RS1 and tries to find a greater signal with another RS from another zone. From the RS, a new naming prefix and a new mobile interest packet are created while reaching a new RS.
Process 4FIB updateThe next step is to broadcast the information to update the FIB that contains the routing information of the I_packet. Routing information is important to control the broadcast domain to make sure that no collision of I_packets occurs between the domain.
Process 5Producer updateAfter this process is completed, the consumer resends a new interest packet to obtain new information from the producer. To maintain connectivity, NDN uses the best route strategy to reduce collision and forward the I_packet to the mobile producer’s new location.
Process 6Cloud updateCloud storage is used to update all transmission information and to store it in case the current connection is interrupted or fails. Thus, its advantage is that transmission is maintained and not disrupted.
Table
Table 4. Hybrid Operation.
Table 4. Hybrid Operation.
Flow of TransmissionProcess StatusDescription
Process 1Exchanging informationMovement occurs from MR 1 to MR 2, and content is stored at MR2.
Process 2ForwardingMR 2 sends a signaling packet to AR 2 when AR 1 sends an alert on movement.
Process 3Creating BITAR 2 and MR 2 create BIT for each entry of I_packets. BIT consists of information on consumer nodes, PoA, and face numbers.
Process 4Matching BITBIT has a similar function to FIB, with I_packets searching for FIB to create a movement from the producer to the consumer.
Process 5Forwarding I_packetsIf the I_packet matches the information request by the consumer, then the content is directed without referring back to FIB.
Table
Table 5. Parameter simulation setup.
Table 5. Parameter simulation setup.
Mobility TechniqueNetwork Size (m2) Distance Router NDN from APMobile Producer QuantityMobility Speed (m/s)Interest Range (ms)Segment Size (bytes)Mobility ModelSimulation SoftwareBenchmark Comparison
KITE400 × 40011 nodes/100 m121001024Random waypoint mobility
model		ndnSIMKITE
PMSS400 × 400100 m250, 200, 350 ms100, 200, 3001024Random waypoint mobility
model		ndnSIMMBMA, CDBMA, CDBMA, IBMA
Hybrid NeMO400 × 400100 m5100 ms100, 2001024Random waypoint mobility
model		ndnSIMKITE
Table
Table 6. Network analysis parameter setup.
Table 6. Network analysis parameter setup.
ParameterUnitsDescriptionParameter/Value
SdatabytesSize data packet2000 bytes
SInterestbytesSize Interest packet40 bytes
SInterest/SmobilityInterestbytesSize Interest packet40 bytes
abytesPacket transmission latency and cost between consumer and producer1
cbytesPacket transmission latency and cost between the old NDN router and the new NDN router5
TpmsPaused time0 ms, 100 ms
abytesPacket transmission latency and cost between the consumer and the producer1
LwmsWired link delay2 ms
dbytesPacket transmission latency and cost between the consumer and the producer and the server9
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Azamuddin, W.M.H.; Aman, A.H.M.; Hassan, R.; Mansor, N. Comparison of Named Data Networking Mobility Methodology in a Merged Cloud Internet of Things and Artificial Intelligence Environment. Sensors 2022, 22, 6668. https://doi.org/10.3390/s22176668

AMA Style

Azamuddin WMH, Aman AHM, Hassan R, Mansor N. Comparison of Named Data Networking Mobility Methodology in a Merged Cloud Internet of Things and Artificial Intelligence Environment. Sensors. 2022; 22(17):6668. https://doi.org/10.3390/s22176668

Chicago/Turabian Style

Azamuddin, Wan Muhd Hazwan, Azana Hafizah Mohd Aman, Rosilah Hassan, and Norhisham Mansor. 2022. "Comparison of Named Data Networking Mobility Methodology in a Merged Cloud Internet of Things and Artificial Intelligence Environment" Sensors 22, no. 17: 6668. https://doi.org/10.3390/s22176668

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop