The Emerging of Named Data Networking: Architecture, Application, and Technology

Named Data Networking (NDN) is developed to accommodate future-oriented internet traffic. In recent years, NDN’s popularity has grown due to the evolution of the Internet of Things, Artificial Intelligence, Cloud Services, and Blockchain. NDN’s key promising features are dynamic content management, privacy, mobility, and trust. This manuscript provides a comprehensive overview of the motivations and promises of NDN in the context of recent evolution technologies and applications. The recent reviews highlight the trend of NDN Blockchain and 5G technology with 73% of research interest, while transportation applications dominate 64% of the research interest. Afterward, we discuss existing NDN Architecture, Operation, Services, and applications, as well as the most popular NDN technologies. Our discussions are based on investigating current solutions with criticism. These significantly highlight the emergence of the NDN and the potential to revolutionize internet traffic management and security.


I. INTRODUCTION
Due to the expansion of the Internet, the Internet of Things (IoT), and new network technology in recent decades, the modern Internet has been forced to comply with even more demanding standards than its creators had envisioned. Accordingly, the efforts of academic and industrial professionals are focusing more on building the next-generation network, overcoming the new Internet's myriad flaws, and ensuring its long-term growth [1]. The massive data sharing and mobility support must tackle different aspects [2]. Statistics show that mobility traffic users will increase more than fixed IP traffic [3]. Shortly, there will be a large number of digital internet-enabled gadgets needing access in any location and at any time. Network architecture should provide dynamic mobility to ensure dependable connectivity and data The associate editor coordinating the review of this manuscript and approving it for publication was Ting Yang . availability anywhere. Because the traditional architecture, TCP/IP, is connection-oriented while going around in one connection, it causes additional network latency. The consumer must submit a request every time because there is an interruption.
Named Data Networking (NDN) caches content on all nodes, eliminating the need to constantly request content from the originating server. It demonstrates the advantages of a well-designed NDN infrastructure through various standard and new applications. NDN enables these qualities to implement reliable networks and applications due to the nature of the underlying architecture design. NDN can enhance performance by minimizing per-content cache consumption and providing a faster time to the first byte, which improves user experience [4]. Furthermore, NDN permits numerous mechanisms that are difficult to implement using IP addresses [5]. Additionally, partial data retrieval, the ability to automatically switch to a better path, transparent failover, and the  ability to retrieve content from numerous sources are essential qualities that simplify an NDN infrastructure and improve user experience [6].
The article has four major sections, as shown in Fig 1. Section I explains the NDN's introduction, including contribution, article selection method, and outline. Section II describes the NDN architecture and operation and summarizes NDN services. Section III demonstrates the NDN application that covers the Internet of Things, smart homes, smart healthcare, intelligent vehicles, and smart cities. Section IV discusses NDN technology inclusively on the cloud, edge, VANET, and 5G. Section V delivers the conclusions and suggestions for future research.

II. NDN ARCHITECTURE, OPERATION, AND SERVICES
As a technology aimed at addressing the current Internet's limits, NDN offers a high-level architecture in which content moves from the origin of the content rather than the physical location. The future architecture of the Internet has shifted standards and repercussions due to many new applications and online services, appropriate mobility, sustainability, trustworthiness, and privacy. NDN is identified to meet the requirements of newly developed Internet applications [22].
NDN routers cache data with satisfied bandwidth [23]. NDN employs a hop-by-hop transmission method as an alternative to the usual paradigm. In this manner, NDN enables multicast and network-based caching features. Any node uses the gateway closest to the current route instead of the external server, saving data retrieval time. This considerably reduces redundancy traffic load throughout the network for the same content, significantly increasing overall network capacity.

A. ARCHITECTURE
The producer, consumer, and NDN router are the three main components of NDN. The producer is a service provider who distributes the content via NDN. Meanwhile, the consumer is a content subscriber who asks NDN for specific structures to fulfill user requests for the requested content, as described in Table 3: CS, PIT, and FIB. NDN retrieves content by name; naming schemes are application-specific and networkindependent. A content name (CN) operates by accessing specific course content in its hierarchical structure. Fig 2  These features make NDN well-suited for applications that require secure, scalable, and flexible data retrieval, such as content delivery, IoT, and vehicular networks.

B. OPERATION
In an NDN operation, the consumer initiates NDN communication by sending an Interest packet. Once the Interest packet grasps the producer, it will return with a Data packet. CN integrates into content, while the Data packet trails the route to ensure it to sent back to the consumer. The linked CN checks the CS for a match when a consumer Interest packet reaches the NDN router. Suppose related content finds, the NDN router handovers it through a Data packet allotted by the producer. The incoming interface for Interest packets adds an Interest aggregation process to the interface list. Consequently, the corresponding Data packet is available, and all attracted consumers accept a copy. If no PIT record exists for an inbound Interest packet, its route to FIB for the longest prefix match (LPM) for a CN.
When a corresponding FIB access this LPMs location, it routes to the relevant next-hop and forms a new PIT entry with its incoming interface. In contrast, a Data packet access to the NDN router searches all PIT records for a matching CN. All of the interfaces listed route the Data packet from the incoming interfaces. Subsequent PIT entry removes, and the content is kept in CS per the local caching policy to serve future requests. The Data packet is lost if there is no matching PIT entry. The operating procedure for the NDN router, represented in Fig 3, indicates that the requestor is looking for the content provided by the producer. Producer yield the content request from the consumer. When the NDN router receives such a request, first check CS to see if the required content is stored locally. The router uses the CN to search for PIT if a connected entry indicates that the NDN router keeps the content locally and sends it based on the subsequent PIT entry.

C. ROUTING
NDN packets are routed and forwarded using names. Consequently, contrasted with the IP architecture, NDN doesn't have addressing issues. No address fatigue issue with the NDN namespace. The routing strategy doesn't involve public and private addressing, and there is no use for Network Address Translation (NAT). There are three types of NDN routing, such as Distance Vector Routing (DVR) [24], Geographical Routing (GR) [25], and Link State Routing (LSR) [26].
In NDN GR routing, packets are delivered to the neighbor if the node does not have a neighbor closer to the destination. The LSR technique seeks the optimal global path in difference towards NDN GR routing. LSR picks all routing information before reach to the producer. Consequently, in LSR routing, each router delivers the data for all potential networks. Another category in NDN routing is proactive routing and [27] reactive routing [28]. Proactive routing uses the router to gather all the information and saves the content into the NDN network. Each node in NDN routing uses the network information before updating the FIB to calculate paths. On-demand routing algorithms based on reactive protocols construct a path for a given destination. Once a path establishes, the node maintains it until it is no longer reachable. FIB in reactive routing neither function as the requesting nor the intermediary node.
Kuai et al. presented a cluster-based routing (CBR) protocol using clustering algorithms for picking cluster heads on alternate paths for information-centric wireless sensor networks (ICWSN) [29]. However, this strategy has significantly increased network lifetime and energy savings and does not rely on caching strategies. Other research has developed a strategy called SCaN-Mob to solve problems handover of producers in wireless networks. It provides a method for a cache replacement strategy and interest forwarding [30].

D. CACHING
Content caching at the NDN router, also known as in-network storage, is critical for supporting the primary notion of NDN's content-centric, with low cost suitable for peer-to-peer data distribution paradigm. In NDN, caching provides numerous advantages. Some advantages are that producer content supports truly separating content from other producers. It eradicates a single point of failure and decreases producer costs by VOLUME 11, 2023   making several copies of the same material available in the NDN network [31]. Due to packet loss, it offers significant advantages to dynamic material in multicast or retransmission. It also reduces network strain and data propagation latency. Cache performance is often measured using the average number of hops, traversed content retrieval delay, and hit ratio. Propagation speed, measured in the time necessary to distribute content to the network core, is another relevant caching performance parameter. Data packet caches in the intermediate routers decide by using the caching policy. Consequently, caching strategies divide into cache placement and cache replacement.

E. FORWARDING STRATEGY
NDN uses a stateful forwarding plan to determine which mobile consumers are behind the traffic. NDN is a consumerdriven infrastructure. NDN forwarding techniques (FT) are another feature of NDN. It is a selective broadcast-based forwarding system that can make forwarding decisions based on past hit rate data. A lightweight forwarding technique compatible with IEEE802.15.4 technology [32]. It works with standard transmission NDN, not vehicular transmission. Vehicle communication has also made use of FT. A study has developed a better system for forwarding interest and data packets. It's called ''data-based forwarding.'' Data packets are distributed based on how many nodes are in the contentbased named network.
Unlike distance-based packet broadcasting, a researcher offered novel strategy techniques that analyze nodes' content worries and forward interest in the direction of more significant concern [33]. Another forwarding strategy solution is a new protocol for transferring named data in VANETs. The forwarding methods make it easier to spread interest and find data with unique names [34]. Mobility info of the vehicles in the neighboring database uses to choose the consumer's nexthop forwarders [35]. It offers a sophisticated NDN forwarding that keeps track of personal and nearby PITs to manage data and interest packets. This work also develops a periodic interest-generating mechanism to make nearby nodes aware of a neighboring node's PIT.

F. SECURITY
The critical problems in NDN are architectural security, costeffective operations, security management, trust management, and confidentiality protection [36]. To facilitate access control and confidentiality, NDN employs content encryption. However, no reliable servers are required to apply security restrictions. Furthermore, the decryption key does not need to distribute because it delivers directly to the consumer. Only the private key can decrypt content encrypted with the public key. There are two types of NDN security encryption, asymmetric and symmetric. For symmetric encryption, by using symmetric keys, a secret key that retrieves from a signed Interest packet is required. A data producer can validate the consumer by marking it and returning an encrypted private key using the consumer's public key. Hao and Wang developed a solution for asymmetric encryption by integrating lightweight physical layer identity [37]. This method increases the differentiation rate and correct authentication probability for NDN security issues on the monitoring plane. Work in [38], a monitoring system for detecting anomalies using Bayesian Network. The last invasion comes from Ko and Mambo [39], which proposes Public Key Encryption Keyword Search (PEKS) for forwarding packets. It has overcome the issue of the cryptographic operation.

III. NDN APPLICATIONS
According to recent research, NDN applications are able to support users' emerging needs and requirements. Intelligent vehicles, smart cities, smart healthcare, and smart homes are the current NDN applications, as shown in Fig 4. The NDN application provides enormous benefits to human life through its enabled services. There are much research related to NDN application, as shown in Table 4. The most popular NDN application is the smart vehicles, followed by IoT applications, while the entertainment sector provides video. NDN also supports other categories, such as innovative environment and education.
Among the NDN applications used in smart vehicles are smart parking, intelligent entertainment, autonomous assistance, sensor systems, and route location identifiers, as shown in Table 5. NDN application that uses IoT has highlighted the interest of researchers, academicians, and industry players. For smart healthcare, the evolution of NDN application technology resulted in numerous advantages for patient happiness and gratification [40]. Wireless body area networks (WBAN) or RFID used in wearable devices that communicate with cloud computing are examples of NDN that use IoT technology in healthcare [41]. These devices provide an appealing alternative for mobile applications and monitoring systems, including blood pressure, ECG, and oxygen saturation [42]. Other healthcare applications include adversative drug reaction monitoring, diabetes prevention rehabilitation systems, inventory, medicine, and wheelchair management systems. IoT devices in NDN applications permit surgeons to observe patient health remotely.
NDN application also mainly report on multimedia application such as video streaming and video conference. Implementing NDN in video streaming improves stable links and minimizes broken links during transmission [43]. Other NDN applications allow synchronized Network Time Protocol by implementing a structure of NDN Hybrid [44]. Besides that, climate or weather monitoring, smart agriculture, and smart farming, waste management are examples of NDN applications in the environmental sector. Waste management is a critical concern for researchers globally. Some countries confuse garbage collection and waste management, yet NDN has improved this solution by applying an intelligent waste management system. Other applications, such as climate monitoring, weather predictions, and smart agriculture, produce more products at a lower cost. Using NDN technology, smart cities may provide intelligent industries, smart facilities to people, and smart homes [45]. This placement uses smart tools and applications for the environment, building more accessible VOLUME 11, 2023 healthcare and transportation. Citizens' smart digital ID is another feature that has many functions and uses. Fig 5 shows how NDN apps use IoT devices to administer, automate control, maintain buildings and factories, and monitor. It operates in many industries, including food, where automatic systems can trace, monitor, and monitor the freshness of food along the supply chain to improve manufacturing and shipping.

IV. NDN TECHNOLOGY
Various technologies enable NDN communication. The most popular ones in NDN technology are IoT, followed by VANET, 5G, edge, and cloud computing, as shown in Fig 6. Microcontrollers, microprocessors, and network interfaces are part of IoT devices. The 5G technology covers mobile devices' telecommunication networks, while the VANET provides a control mechanism for moving vehicles, and for both, the NDN significantly supports better performance. As for edge computing, the NDN is preferred for data storage management support. Table 6 maps the most popular simulation tools used for NDN technologies research.

A. BLOCKCHAIN
Blockchain, known as ledger technology, improves security and enhances the NDN to catch the content. The use of blockchain technology ensures a secure and transparent transfer of data between vehicles and other road entities. The data can include traffic conditions, road conditions, vehicle maintenance information, and more. Applying a reputation-based mechanism Blockchain improve caching in Vehicle Named Data Network (VNDN) and enhance trust for content access through consumer [46]. The VNDN network aims to improve road safety, reduce traffic congestion and make driving more efficient. Another issue with VNDN is the current structure dependent on IP base transportation. Implementing Blockchain on NDN solves the issues and increases network security [47]. Other works have developed a new solution in VNDN by leveraging Blockchain and NDN [48]. This framework enhances the intelligent transportation system increasing trust and transparency. Secure data transmission needs to highlight, and a policy hiding scheme has come up with a solution for data revocation in VNDN [49]. Work in, [50] developed a reputation based on Blockchain to secure forwarding planes strategy. These techniques assure privacy and trust. Ahmad et al. [51] develop a solution by applying key management with a consensus algorithm to support privacy-aware on the VNDN network. Work in [52] proposes using key management based on blockchain technology to encounter mutual trust with anchors and improve verification efficiency.
Blockchain NDN (BoNDN) proposed by [53] supports broadcasting in NDN to generate blocks from subscriptions. Another solution to the NDN healthcare monitoring application issue solves by applying IoT blockchain technology. Fabric Hyperledger is used in work [54] to apply some functionality to security architecture on NDN.
Work in [55] covers Internet-of-Battlefield-Thing (IoBT) topics by applying Interest groups to overcome resource constraints. Distributed data management encounters security and privacy issues, and some problems solve by applying Blockchain. Work in [56] uses identifiers on Blockchain in the NDN environment to safely store the content. Alowish et al. came up with a survey on Blockchain over NDN that covers the benefit of Blockchain in decentralized payment, e-health care, and cloud computing [57].

B. CLOUD AND EDGE TECHNOLOGY
New-era research continues with VANET over cloud base technology. NDN came out with a unique solution called NDN-VC for a cloud-based that provides reliability and an intelligent transportation system [50]. Content priority causes issues in VNDN communication because the content is important before being transmitted. Thomas et al. propose a new algorithm that increases the hit ratio and reduces the  signal overhead compared to IP architecture [58]. The latest discussion on NDN cloud is application distribution and network architecture. Work in [59] uses Geographical Vehicular Central Data Networking (GeoVCDN) to distribute vehicle content and improve efficiency compared to rendezvous and initial NDN networks. Macrovehicular cloud comes with an VOLUME 11, 2023   issue of reverse data disruption and data cost. NDN-based aVC framework (NVCF) improves data acquisition and content transmission success [60].
NDN over-edge cloud computing (ECC) gives many benefits in terms of security, flexibility, and resiliency. Work in [61], hybrid NDN over ECC gives a higher response time than traditional edge computing.
Other work in edge computing integrates with IoT applications. One research found that the previous Internet of Things Cloud (IoTC) improved data retrieval based on IP architecture, but it's hard to embed IP addressing on IoT technology. Based on IoT, NDN proposes to overcome IP addressing issues, improve data efficiency by almost 9 percent, and reduce the cost to 42 percent compared to the traditional IoT approach. Other solutions also give better results to reduce cost and increase data retrieval by integrating virtual cloud and NDN [62].

C. ARTIFICIAL INTELLIGENCE
To establish a value-oriented ecosystem, upgrading from IP-based to knowledge-driven improves NDN technology significantly. Digital tokenization enhances storage, computing, and networking by embedding blockchain technology [63]. In edge computing, lack of content identifiers but more on host-centric communication urge toward NDN AI. NDN has developed a solution for edge computing by implementing a machine algorithm suited for user-centric communication [64]. Edge computing is constantly issuing big data and cloud computing issues. NDN AI evolution comes with a solution in edge computing that resolves big input data and reduces delay latency [65]. NDN AI also comes with a solution to enhance security and stability routing. A collusive Interest Flooding Attack (CIFA) is a vulnerable attack that affects producers to respond with malicious Interest packets.

D. INTERNET OF THINGS
IoT uses 802.15.4 protocol to communicate at a satisfactory rate, increase data range and reduce the level of energy. To enable integration of IoT device IoT protocol communication such as 6lowPAN, NDN is the alternative to resolve mobility, and content is human-readable. A solution with Mr-IoT to control access to billion data on the NDN network [67]. Smart agriculture is one solution that uses NDN over IoT to reduce energy consumption and cost overhead [68]. Touati et al. find that in IoT, billions of data reproduce every single time. It solves the issue of dynamic computing for content delivery, managing high-volume data traffic and service response time [69]. Table 6 shows the summary of simulation tools of NDN technology research.

V. CONCLUSION
This manuscript provides a concise explanation of NDN's architecture, operation, and services, as well as the most recent technological development and application. Bringing to light the most current, being the primary area of concentration and widely researched NDN technology, is the Internet of Things. Other fields, such as Artificial Intelligence, Blockchain, and Cloud-Edge, are also receiving a lot of interest from researchers. In the meantime, in descending order, the NDN applications sector that is the most popular is the transportation sector, followed by the Internet of Things sector, the streaming sector, and finally, the healthcare sector. Eventually, to be more specific, the smart vehicle, smart healthcare, and smart city categories are considered to be the subfields of application within the sectors. AZANA HAFIZAH MOHD AMAN received the B.Eng., M.Sc., and Ph.D. degrees in computer and information engineering from International Islamic University Malaysia. She is currently a Senior Lecturer with the Research Center for Cyber Security, Faculty of Information Science and Technology (FTSM), Universiti Kebangsaan Malaysia (UKM), Malaysia. Her research interests include computer system and networking, information and network security, the IoT, cloud computing, and big data.
HASIMI SALLEHUDDIN is currently a Senior Lecturer with the Faculty of Information Science and Technology, Universiti Kebangsaan Malaysia. His research interests include computer security and networks, and management information systems.
KHALID ABUALSAUD (Senior Member, IEEE) is currently with the Department of Computer Science and Engineering, Qatar University, Qatar. He has more than 25 years of professional experience in information technology. He teaches courses in hardware and software systems. He is active in getting research funding from different sources, including the Qatar National Research Foundation, the Supreme Committee for Delivery and Legacy (FIFA 2022), and some other organizations in Qatar. He is also a LPI of the NPRP 10-1205-160012 Research Project, which achieved significant outcomes. He has participated actively in organizing several IEEE international conferences in Qatar, namely ICIoT 2020, IEEE WCNC 2016, PLM 2015, AICCSA 2014, RelMiCS 2011, and AICCSA 2008. His research work has been presented at international conferences and journals. His research interests include health systems, wireless sensors for IoT applications, cybersecurity, cloud computing, and computer network protocols. He has served as a technical program committee (TPC) member and the chair for various reputable IEEE conferences. He received several awards from different local and international organizations. He served as the Guest Editor for Connected Healthcare Special Issue for IEEE Network. He is an Associate Editor of IET Quantum Communication.
NORHISHAM MANSOR received the M.Sc. degree in computer science from the Technical University of Malaysia (UTeM), in 2014. He is currently a Vocational Training Officer with the Advance Technology Training Center (ADTEC) Batu Pahat. He has published several articles in peer-reviewed international journals. His current research interests include networking security, future internet technology, and NDN security. VOLUME 11, 2023