Inter national J our nal of Electrical and Computer Engineering (IJECE) V ol. 16, No. 3, June 2026, pp. 1626 ∼ 1644 ISSN: 2088-8708, DOI: 10.11591/ijece.v16i3.pp1626-1644 ❒ 1626 AMA C-L W : Adapti v e medium access contr ol f or long range wide ar ea netw ork with ener gy-awar e r outing So wmya M. 1 , S. Meenakshi Sundaram 2 , P andiyanathan Murugesan 3 , Santhosh K umar K. S. 4 , T ejaswini R. Mur god 5 1 Department of Articial Intelligence and Data Science, Nitte Meenakshi Institute of T echnology , Beng aluru, India 2 Department of Computer Science and Engineering, A CS Colle ge of Engineering, Beng aluru, India 3 Department of Computer Science and Engineering, K oneru Lakshmaiah Education F oundation, V addesw aram, India 4 Department of Articial Intelligence and Machine Learning, Mysore Uni v ersity School of Engineering, Uni v ersity of Mysore, Mysuru, India 5 Department of Articial Intelligence and Machine Learning, B N M Institute of T echnology , Beng aluru, India Article Inf o Article history: Recei v ed Dec 14, 2024 Re vised Jan 21, 2026 Accepted Mar 16, 2026 K eyw ords: Data communication ef cienc y Ener gy-ef cient LoRaW AN netw orks MA C layer Optimized routing algorithm P ack et deli v ery ratio Security ABSTRA CT T o enhance the performance of long range wide area netw ork (LoRaW AN), a routing algorithm and a no v el medium access control (MA C) layer protocol are required. In addition to addressing scalability and security issues, the protocol seeks to impro v e communication ef cienc y , dependability , and po wer consump- tion. It presents a dynamic routing method that reduces ener gy consumption by utilizing machine learning processes, adapti v e routing tactics, and route opti- mization approaches. Simulations in a range of deplo yment situations are used to assess the suggested solutions. These results imply that t he suggested proto- col and routing scheme ha v e the potential to greatly enhance the sustainability , ener gy ef cienc y , and performance of LoRaW AN-based Internet of Things net- w orks. The ef fecti v eness of the proposed solutions is e v aluated through e xten- si v e simulations across di v erse deplo yment scenarios. The results demonstrate that the proposed MA C protocol achie v es a throughput of 350 bps, outperform- ing con v enti onal protocols that typically reach only 220 bps. Latenc y is re- duced to 50 ms from 85 ms, ener gy consumption is decreased to 2.5 joules from 4.5 joules, and the pack et deli v ery ratio (PDR) is impro v ed to 95%, compared to 75% in e xisting approaches. These ndings highlight the potential of the pro- posed protocol and routing scheme to signicantly enhance the performance, ener gy ef cienc y , and sustainability of LoRaW AN-based IoT netw orks. This is an open access article under the CC BY -SA license . Corresponding A uthor: So wmya M. Department of Articial Intelligence & Data Science, Nitte Meenakshi Institute of T echnology Beng aluru-560064, India Email: sanu.196@gmail.com 1. INTR ODUCTION The Internet of Things (IoT) is re v olutionizing digital and ph ysical en vironments, leading to v ast in- terconnected netw orks of sens o r s, de vices, and smart systems. Lo w po wer wide area netw orks (LPW ANs) ha v e become crucial for IoT deplo yments due to their long-range connecti vity and lo w ener gy consumption. Long range wide area netw ork (LoRaW AN) a widely adopted LPW AN protocol, is scalable, e xible, and cost-ef fecti v e, making it suitable for lo w-po wer , battery-operated de vices . Ho we v er , to fully realize the poten- tial of LoRaW AN in lar ge-scale IoT ecosystems, performance-related challenges at both the medium access J ournal homepage: http://ijece .iaescor e .com Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 ❒ 1627 control (MA C) and routing layers need to be addressed. LPW AN technologies enable IoT de vices to commu- nicate o v er long distances with minimal po wer consumption, and LoRaW AN has g ained widespread adoption due to its e xible architecture, lo w deplo yment cost, and suitability for lo w-data-rate, long-range communi- cation scenarios. This study aims to address the need for reliable and ener gy-ef cient communication in IoT deplo yments. Standard LoRaW AN protocols, while adequate for basic operations, struggle in dynamic en vi- ronments with high node density , frequent data transmission, and intelligent decision-making. IoT applications in smart cities, healthcare, and industrial automation require protocols that can adapt in real-time, consume minimal ener gy , and maintain secure data deli v ery . Existing MA C protocols in LoRaW AN lack adaptability and intelligence, prompting t he de v elopment of enhanced solutions. Machine learning-based techniques are inte grated into routing and scheduling mechanisms to ensure data inte grity and minimize ener gy consumption. LoRaW AN operates on the LoRa modulation technique, which uses chirp spread spectrum (CSS) for long-range communication. Ho we v er , traditional adv ocates of linux open-source ha w aii association (ALOHA)-based ac- cess schemes result in increased collision probability , pack et loss, and ener gy inef ciencies. Researchers are e xploring more adv anced MA C and routing protocols for performance enhancement and ener gy optimization. LoRaW AN is a v ersatile IoT communication protocol that of fers long-range communication, lo w po wer consumption, scalability , and cost-ef fecti v eness. It supports thousands of end de vices per g ate w ay and is ener gy-ef cient, allo wing de vices to spend most of their time in lo w-po wer sleep modes. LoRaW AN also supports bidirectional communication, enabling data transmission and remote control. It also supports mobil- ity to some e xtent. Ho we v er , LoRaW AN has limitations, such as a lo wer data rate compared to other wireless technologies, increased netw ork congestion and pack et collisi ons in dense deplo yments, and limited quality of service (QoS) due to its reliance on the ALOHA protocol. Basic implementations may also lack adv anced intrusion detection or anomaly mitig ation mechanisms. These trade-of fs highlight the need for continued opti- mization in LoRaW AN protocol design to meet the e v olving demands of IoT systems. LoRaW AN is a wireless communication system that consists of three classes of end de vices are class A, class B, and class C. Class A is the most ener gy-ef cient mode, ensuring minimal ener gy usage for battery-po wered sensors used in agricul- ture or utility metering. Class B synchronizes de vices with beacons from the g ate w ay , allo wing for predictable do wnlink communication. Class C, the most responsi v e class, k eeps the de vice’ s recei v er open e xcept during transmission, pro viding the lo west latenc y for do wnlink messages. Each class caters to dif ferent IoT use cases, with class A for ultra-lo w po wer needs, class B for balanced control and po wer , and class C for latenc y-critical operations. Ho we v er , the performance of LoRaW AN in demanding applications can be signicantly impro v ed through the inte gration of intelligent MA C layer protocols and ener gy-a w are routing strate gies. These enhance- ments are essential for scaling LoRaW AN to meet the requirements of ne xt-generation IoT netw orks. As depicted in Figure 1, the adapti v e MA C protocol for LoRaW AN (AMA C-L W) is structured in a lay- ered and modular f ashion, enabling intelligent coordination of communicati on processes within IoT netw orks. At the top, the application layer interf aces directly with the AMA C-L W protocol, which serv es as the central medium access control entity responsible for managing transmission requests, ackno wledgment s, and commu- nication orchestration between end-de vices and g ate w ays. The AMA C-L W protocol is composed of three core components: the MA C common part sublayer (MCPS), which f acilitates da ta transmission and reception; the MA C layer m anagement entity (MLME), which o v ersees essential netw ork management tasks such as joining procedures and scheduling; and the MA C information base (MIB), which stores runtime congurations and operational state data required for protocol e x ecution. In addition to these primary elements, AM A C -L W in- corporates se v eral adapti v e enhancements that further optimize performance. Dynamic duty c ycling adjusts de vice acti vity periods in response to traf c load and ener gy conditions, while congestion-a w are routing selects optimal paths to alle viate netw ork bottlenecks. Ener gy ef cienc y optimization mechanisms are embedded to prolong bat tery life without sacricing reliability . Security is reinforced through inte grated encryption, au- thentication, and anomaly detection techniques. Scalability is addressed through dynamic parameter tuning, enabling seamless netw ork e xpansion. Moreo v er , the protocol benets from machine learning enhancements that le v erage historical patterns to inform smarter decisions in both routing and MA C operations. Collecti v ely , these features enable AMA C-L W to deli v er high-performance, ener gy-ef cient, and scalable communication suitable for modern IoT en vironments. The e xisting medium access control (MA C) protocols in LoRaW AN f ace signicant limitations when deplo yed in dense or lar ge-scal e netw orks. The basic LoRaW AN MA C layer , which follo ws a pure ALOHA scheme, suf fers from high pack et collision rates, lacks ef cient scheduling and resource allocation strate- gies, resulting in suboptimal throughput, increased latenc y , e xcessi v e ener gy consumption, and challenges in AMA C-L W : Adaptive medium access contr ol for long r ang e wide ar ea network with ... (Sowmya M.) Evaluation Warning : The document was created with Spire.PDF for Python.
1628 ❒ ISSN: 2088-8708 maintaining scalability and security . T o address these shortcomings, an enhanced MA C protocol architecture, called AMA C-L W , is introduced. This protocol i nte grates adapti v e communication mechanisms and intelligent decision-making capabilities, ensuring reliable, ener gy-ef cient, and secure data transmission across IoT net- w orks. Ener gy-ef cient routing is also crucial in LoRaW AN, as traditional mechanisms often ignore dynamic changes in netw ork topology and node ener gy status, leading to une v en ener gy depletion, bottlenecks, and reduced netw ork lifetime. An ener gy-ef cient and congestion-a w are routing algorithm is proposed to ensure sustained performance and balanced ener gy usage across the netw ork. Figure 1. Adapti v e MA C protocol for LoRaW AN (AMA C-L W) architecture 2. RELA TED W ORK Li et al. [1] proposed the CGBS-LoRa MA C protocol, which signicantly impro v es the scalabil- ity of LoRa netw orks and reduces de vice collisions. The protocol maintains a high pack et deli v ery rate and lo w latenc y , e v en as the LoRaW AN netw ork gro ws. Chasserat et al. [2] introduced LoRaSync, a time-slotted ALOHA-based access method that uses a cl ock drift model from real lo w-cost de vices. Their solution enhances throughput and ener gy ef cienc y , with performance v alidated through simulations and testbed implementation. P aul et al. [3] de v eloped a frame w ork to assist LoRaW AN netw ork designers in selecting or creating models tailored to specic application requirements such as delay and de vice lifetime. The frame w ork incorporates simulation-based comparisons of single-hop and multi-hop routing. Jouhari et al. [4] conducted a comprehen- si v e surv e y of LoRaW AN scalability issues at the ph ysical and MA C layers, focusing on capacity e xpansion and interference reduction. The study highlights e xisting solutions such as spreading f actor optimization, channel assignment, and alternati v e topologies. Chen et al. [5] modeled class-A LoRaW AN de vices using probabilistic timed automata (PT A) to capture timing beha vior , transmission schedules, and collision dynamics. The y used the PRISM model check er for quantitati v e analysis under v arious conditions. Leonardi et al. [6] also emplo yed PT A and PRISM to model and analyze class-A LoRaW AN beha vior , emphasizing MA C layer interactions and performance e v aluation under dif ferent netw ork scenarios. Ahmar et al. [7] proposed a t ime-synchronized cryptographic frequenc y hopping MA C protocol t hat enhances LoRa’ s scalability , security , and reliability . The protocol demonstrates superior performance and re- sistance to selecti v e jamming compared to con v entional LoRaW AN. Chen et al. [8] analyzed denial-of-service (DoS) vulnerabilities in LoRaW AN’ s MA C layer and proposed tw o tar geted attacks on conrmed transmis- sions. Their ndings sho w that e v en a small number of attack ers can signicantly reduce pack et success rates and ener gy ef cienc y . Li et al. [1] further emphasized impro v ements in ALOHA-based LoRaW AN commu- nication through geographical se gmentation and optimized transmission parameters, sho wing enhanced scala- bility and collision reduction in dense netw orks. Prasetyo et al. [9] proposed the LoRa multi-communication (LMC) protocol at the application layer . Designed for IoT de vices with limited ener gy resources, the protocol impro v es battery life and pack et reception based on e xperimental v alidation. Pirri et al. [10] proposed enabling LoRaW AN end-de vices to support multiple MA C protocols to meet the quality of service (QoS) demands in industry 4.0 en vironments. Their method uses o w mapping to address v ari ous latenc y and reliability needs, while also highlighting se v eral design challenges. Chen et al. [11] in v estig ated the impact of greedy beha viors by compromised nodes in LoRaW AN’ s MA C layer . The y proposed a double judgment de tection mechanism. Although LoRaW AN remains f airly Int J Elec & Comp Eng, V ol. 16, No. 3, June 2026: 1626-1644 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 ❒ 1629 resilient, e xtensi v e gr eedy beha vior de grades performanc e. Ho we v er , their approach ef fecti v ely detects such acti vities. Leonardi et al. [12] conducted a simulation-based e v aluation of the listen before talk adapti v e fre- quenc y agility (LBT -AF A) MA C protocol for LoRaW AN under v arying MA C parameters and node densities. Their ndings guide MA C parameter optimization for impro v ed latenc y and consistent performance. Dieng et al. [13] introduced a real-time, collision-free scheduling technique for LoRaW AN based on graph coloring. This method signicantly enhances scalability and reliability for time-sensiti v e IoT applications, as e videnced by NS-3 simulation results sho wing reduced pack et loss and impro v ed deadline adherence. Li [14] proposed a h ybrid access technique that combines S-ALOHA and TDMA to support both periodic and b urst transmis- sions in LoRaW AN. MA TLAB simulations demonstrated reduced collision rates and better channel utilization compared to standard LoRaW AN. Tsakmakis et al. [15] de v eloped an adapti v e h ybrid MA C protocol based on learning automation. The protocol w as sho wn through simulat ion to substantially reduce transmission latenc y when compared with con v entional LoRaW AN approaches. Cheikh et al. [16] pro vided a tutorial re vie w of machine learning approaches for LoRaW AN resource optimization, including transmission po wer control and spreading f actor tuning. The y also identied accessible datasets, simulation tools, and outlined directions for ML-dri v en enhancements. Banti et al. [17] present ed a comprehensi v e surv e y of LoRaW AN MA C protocols with a focus on ener gy ef cienc y . The study compares e xisting techniques, ident ies limitations, and suggests future research directions. Alahmadi et al. [18] pro- posed the SBTS-LoRa MA C protocol, where nodes adjust transmission parameters based on their distance from the g ate w ay . Simulation results sho wed signicant impro v ements in throughput and scalability—up to 14×—compared to adapti v e data rate (ADR) schemes. Xiao et al. [19] e xplored the inte gration of collision decoding techniques with MA C protocols in LoRa netw orks. The y cate gorized and analyzed v arious decoding methods, e xamined their impact on MA C strate gies, and proposed future research opportunities for massi v e IoT connecti vity . T riantafyllou et al . [20] i ntroduced TS-VP-LoRa, a time-slotted MA C scheme that incor p o- rates g ate w ay-coordinated scheduling and channel hopping to enhance LoRaW AN’ s scalability . Simulations demonstrated impro v ements in pack et deli v ery , reduced latenc y , and fe wer collisions, all while preserving en- er gy ef cienc y in dense deplo yments. Zhong and Springer [21] proposed a time-slotted MA C protocol and edge-ackno wledging architecture to impro v e reliability and ener gy ef cienc y in Lo R aW AN conrmed messag- ing. Their approach enhances pack et reception and ener gy sa vings, though it introduces additional delay in lar ger netw orks. Chasserat et al. [22] presented TREMA, a traf c-a w are and ener gy-ef cient MA C protocol for LoRa. TREMA dynamically switches between s yn c hrono us and asynchronous communication to balance ener gy con- sumption and netw ork capacity under v arying traf c conditions. F arooq [23] de v eloped a multi-hop routing protocol with a softw are-dened netw orking (SDN) e xtension for LoRa, aimed at achie ving high data rates and e xtended co v erage. Ev aluations indicated a 5× increase in pack et reception ratio and reduced ener gy consump- tion compared to traditional LoRa settings. Chinchilla-Romero et al. [24] proposed the CARA method, which uses ef cient resource allocation, le v eraging multi-channel access and spreading f actor orthogonali ty . Both simulation and e xperimental results sho wed up to a 95.2% increase in LoRaW AN capacity and full compati- bility with e xisting de vices. T riantafyllou et al. [25] also proposed FCA-LoRa, a beacon-based MA C protocol designed to impro v e throughput in dense LoRaW AN netw orks. Simulations sho wed up to a 50% throughput impro v ement o v er enhanced ALOHA-based protocols. Garrido-Hidalgo et al. [26] implemented a practical lo w-o v erhead synchronization and scheduling system for class A LoRaW AN de vices. Experimental results using SF12 achie v ed pack et deli v ery ratios of up to 29%, v alidating its performance in high-load scenarios. Leonardi et al. [27] e v aluated the performance impacts of updated ETSI re gulation constraints on LoRaW AN MA C protocols, focusing on pure ALOHA and listen before talk (LBT). Their w ork pro vides simulation-based comparisons to assist with protocol selection under realistic traf c loads. T able 1 presents a comparati v e analysis of se v eral recent LoRaW AN MA C protocol enhancement s based on selected w orks from the literature. These protocols aim to im pro v e aspects such as throughput, ener gy ef cienc y , security , and scalability in IoT communication. F or e xample, the w ork by Li et al. [1] introduces a grouped bit-slot approach to impro v e channel utili zation, while Chasserat et al. [2] focus on ener gy-ef cient synchronization through LoRaSync. Some studies, lik e that of Ahmar et al. [7], emphasize rob ust communication by handling pack et collisi ons and ensuring f airness. Meanwhile, Chen et al. [8] and Jouhari et al. [4] analyze securi ty concerns and DoS vulnerabilities in the MA C layer . Ov erall, the table sho ws that while mos t protocols enhance throughput and ener gy usage, only a fe w gi v e detailed att ention to security and lar ge-scale deplo yment support. AMA C-L W : Adaptive medium access contr ol for long r ang e wide ar ea network with ... (Sowmya M.) Evaluation Warning : The document was created with Spire.PDF for Python.
1630 ❒ ISSN: 2088-8708 This highlights the need for a comprehensi v e solution lik e AMA C-L W , which addresses all these as- pects ef fecti v ely . T able 2 presents recent MA C protocol de v elopments tailored for LoRaW AN. These protocols v ary from ener gy-ef cient solutions lik e TREMA [22] and LoRaSync [2] to rob ust and secure models lik e the one proposed by Chen et al. [8]. Hybrid and planning-a w are approaches [3, 6] enhance e xibility and netw ork- wide optimization. Adapti v e and e v ent-triggered designs [1, 15] sho w promise in dynamic IoT en vironments. T able 1. Comparison of selected LoRaW AN MA C protocols rele v ant to AMA C-L W W ork Throughput Ener gy Ef cient Security Scalability Li et al. [1] ✓ ✓ ✗ ✓ Chasserat et al. [2] ✓ ✓ ✗ ✓ P aul et al. [3] ✓ ✓ ✗ ✓ Jouhari et al. [4] ✓ ✓ ✓ ✓ Chen et al. [5] ✓ ✗ ✓ ✓ Leonardi et al. [6] ✓ ✓ ✗ ✓ Ahmar et al. [7] ✓ ✓ ✓ ✓ Chen et al. [8] ✓ ✗ ✓ ✗ Prasetyo et al. [9] ✓ ✓ ✗ ✗ Pirri et al. [10] ✓ ✓ ✗ ✓ Chen et al. [11] ✓ ✗ ✓ ✓ Leonardi et al. [12] ✓ ✓ ✗ ✓ Dieng et al. [13] ✓ ✓ ✗ ✓ Li [14] ✓ ✓ ✗ ✓ Tsakmakis et al. [15] ✓ ✓ ✗ ✓ Cheikh et al. [16] ✓ ✓ ✗ ✓ Banti et al. [17] ✓ ✓ ✗ ✓ Alahmadi et al. [18] ✓ ✓ ✗ ✓ Xiao et al. [19] ✓ ✓ ✗ ✓ T riantafyllou et al. [20] ✓ ✓ ✗ ✓ Zhong and Springer [21] ✓ ✓ ✓ ✓ Chasserat et al. [22] ✓ ✓ ✗ ✓ F arooq [23] ✓ ✓ ✗ ✓ Chinchilla-Romero et al. [24] ✓ ✓ ✗ ✓ T riantafyllou et al. [25] ✓ ✓ ✗ ✓ Garrido-Hidalgo et al. [26] ✓ ✓ ✗ ✓ Leonardi et al. [27] ✓ ✓ ✗ ✓ T able 2. State-of-the-art MA C Protocols for LoRaW AN Ref MA C Protocol / Scheme F ocus Area Strengths Limitations [1] Circular Re gion Grouped Bit-Slot Collision Reduction Ef cient channel use No security support [2] LoRaSync Synchronization Ener gy-ef cient sync Sync o v erhead [16] Multi-layered MA C Ener gy Model Ener gy Ef cienc y Cross-layer optimized Inte gration comple xity [22] TREMA T raf c A w areness Ener gy-ef cient load Comple x coordination [8] Secure DoS-resilient MA C Security Jamming defense Crypto o v erhead [3] LoRaW AN Planning-A w are MA C Netw ork Optimization Impro v ed co v erage/QoS Static assumptions [9] No v el MA C for Non-LoRaW AN Alternati v e Frame w ork Independence Compatibility limits [6] Combined MA C Schemes Hybrid Design V ersatile operations Coordination o v erhead [15] Adapti v e MA C for Ev ent Detection Ev ent-Dri v en Access Reduced latenc y Hardw are dependenc y [21] Conrmed T raf c-A w are MA C Reliability Enhanced A CK handling Conrmation o v erhead 3. PR OBLEM ST A TEMENT The current MA C protocols in LoRaW AN ha v e limitations in data communication ef cienc y , relia- bility , ener gy consumption, security , and scalability . A comprehensi v e analysis of these protocols is needed to de v elop a ne w protocol tailored to specic IoT applications. Ener gy-ef cient routing algorithms are crucial for minimizing ener gy consumption in LoRaW AN netw orks, impacting de vice longe vity and netw ork per - formance. Challenges include maintaining de vice security while reducing ener gy usage, managing netw ork congestion, and adapting to dynamic netw ork conditions. Inno v ati v e solutions incorporating machine learn- ing and articial intelligence techniques, dynamic routing adaptations, and h ybrid approaches are needed to enhance the performance of routing algorithms and ensure ener gy ef cienc y and security within LoRaW AN en vironments. Int J Elec & Comp Eng, V ol. 16, No. 3, June 2026: 1626-1644 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 ❒ 1631 4. PR OPOSED SOLUTION Addressing the challenges identied in the MA C layer protocols and routing algorithms in LoRaW AN includes the follo wing stages. 4.1. Pr oposed model: adapti v e MA C pr otocol f or LoRaW AN (AMA C-L W) model The adapti v e MA C protocol for LoRaW AN (AMA C-L W) is designed to address the k e y chal lenges of data communication in LoRaW AN netw orks is as sho wn in Figure 2, focusing on optimizing ef cienc y , reliability , ener gy consumption, security , and scalability . Belo w are the main features and functionalities of the proposed model: a. Dynamic duty c ycling: The protocol implements adapti v e duty c ycling strate gies that allo w nodes to adjust their acti v e and sleep periods based on real-time traf c conditions. This optimizes ener gy consumption by ensuring that nodes are a w ak e only when data needs to be transmitted or recei v ed, e xtending battery life. b . Congestion-a w are routing: AMA C-L W inte grates a congestion-a w are routing mechanism that continuously monitors netw ork traf c and identies congested paths. When congestion is detected, the protocol reroutes data through alternati v e paths, reducing the lik elihood of pack et loss and enhancing o v erall netw ork perfor - mance. c. Ener gy ef cienc y optimization: The model emplo ys ener gy-ef cient routing algorithms that minimize the distance data must tra v el and the number of hops required. By optimizing these f actors, the protocol reduces ener gy consumption, which is crucial for battery-operated IoT de vices. d. Security inte gration: security measures are seamlessly inte grated into the prot ocol without signicantly af fecting ener gy consumption. The model emplo ys lightweight encryption methods to ensure data con- dentiality and inte grity during transmission, addressing concerns o v er data pri v ac y in LoRaW AN netw orks. Figure 2. Proposed model for ener gy-ef cient MA C layer protocol with optimized routing in LoRaW AN T able 3 outlines the inte gration of multiple security mechanisms within the proposed AMA C-L W frame w ork. At the ph ysical (PHY) and MA C layers, AES-128 or AES-256 encryption ensures data con- dentiality during transmission using MCPS services. Authentication is enforced using a message inte grity code (MIC), particularly during the join procedure, le v eraging MLME services to ensure message authenticity and pre v ent spoong attacks. K e y management is handled dynamically through the NwkSK e y and AppSK e y , which are managed via MIB congurations and MLME runtime services. Replay protection is achie v ed by v erifying frame counters within the MLME layer , safe guarding ag ainst message duplication and delay attacks. Additionally , adapti v e security is int e grat ed through machine learning models that detect anomalies in traf- c patterns—such as abnormal message frequenc y or structure—allo wing the system to adapti v ely respond to emer ging threats. This layered and inte grated approach pro vides a rob ust foundation for ensuring both security and ener gy ef cienc y in LoRaW AN-based IoT applications. AMA C-L W : Adaptive medium access contr ol for long r ang e wide ar ea network with ... (Sowmya M.) Evaluation Warning : The document was created with Spire.PDF for Python.
1632 ❒ ISSN: 2088-8708 a. Scalability: the proposed model is designed to be scalable, allo wing it to adapt to v arying netw ork sizes and densities. It can ef ciently manage lar ge numbers of nodes and dif ferent communication patterns, making it suitable for di v erse IoT applications. b . Machine learning enhancements: AMA C-L W incorporates machine learning techniques to impro v e routing decisions based on historical and real-time data. This allo ws the protocol to learn and adapt to changing netw ork conditions, enhancing its performance o v er time. c. Simulation-based e v aluation: The model will be e v aluated through simulations that replicate v arious de- plo yment scenarios, assessing its performance in terms of throughput, latenc y , ener gy consumption, and reliability ag ainst e xisting protocols. T able 3. Security inte gration in AMA C-L W : adapti v e medium access control for LoRaW AN Security Layer T echnique Inte gration Point Encryption AES-256 Applied at PHY and MA C payload le v els through MCPS Services Authentication MIC (Message Inte grity Code) Enforced through Join Request/Accept via MLME Services K e y Management NwkSK e y and AppSK e y Managed via MIB runtime and supported by MLME layer Replay Protection Frame counter v erication Implemented within MLME Services to pre v ent duplication Adapti v e Security Machine Learning detection Inte grated through MLME and Security Inte gration module The AMA C-L W model represents a comprehensi v e approach to enhancing LoRaW AN data commu- nication by focusing on ener gy ef cienc y , dynamic adaptability , and rob ust security features. By addressing the limitations of e xisting MA C protocols and incorporating inno v ati v e routing strate gies, the model aims to impro v e the o v erall ef fecti v eness and reliability of IoT applications in LoRaW AN netw orks. 4.2. Pr oposed method: implementation of the adapti v e MA C pr otocol f or LoRaW AN (AMA C-L W) This proposed protocol inte grates ener gy-ef cient rout ing, dynamic netw ork adaptations, and AI- dri v en security , with emphasis on IoT de vice longe vity and ef cient data communication within LoRaW AN netw orks. a. Conduct a thorough literature re vie w of e xisting MA C protocols in LoRaW AN to identify their strengths and limitations. This analysis helps to inform the desi gn of the AMA C-L W by highlighting areas for impro v ement, such as ener gy consumption, reliability , and scalability . b . The proposed method for de v eloping the AMA C-L W in v olv es a structured approach that be gins with thor - ough analysis and design, follo wed by algorithm de v elopment, simulation testing, and iterati v e renement. c. By focusing on the unique needs of IoT applications and le v eraging adv anced routing and ener gy-ef cient strate gies, the AMA C-L W aims to enhance the ef fecti v eness of data communication in LoRaW AN netw orks signicantly . 4.3. Mathematical model The proposed adapti v e medium access control for long range wide area netw ork (AMA C-L W) denes v arious parameters, v ariables, and equations representing the core components of the protocol. The model captures duty c ycling, routing algorithms, ener gy consumption, and performance metrics. 4.3.1. K ey parameters and v ariables − N : T otal number of nodes in the netw ork − D i : Data rate of node i (bits/second) − L : A v erage pack et size (bits) − R : T ransmission range of each node (meters) − T acti v e : T ime a node remains acti v e (seconds) − T idle : T ime a node remains idle (seconds) − P trans : Po wer consumed during transmission (w atts) − P recv : Po wer consumed during reception (w atts) − P idle : Po wer consumed during idle state (w atts) − E total : T otal ener gy consumed by a node (joules) − E battery : Battery ener gy capacity of a node (joules) Int J Elec & Comp Eng, V ol. 16, No. 3, June 2026: 1626-1644 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 ❒ 1633 − T latenc y : End-to-end latenc y (seconds) − T throughput : Netw ork throughput (bits/second) − C : Channel capacity (bits/second) − P success : Probability of successful pack et deli v ery − C cong : Netw ork congestion metric The AMA C-L W protocol incorporates ener gy-a w are and traf c-adapti v e mechanisms for ef cient MA C scheduling. The mathematical model bel o w details ener gy consumption, latenc y , reliability , and adapti v e beha viors. 4.4. Duty cycling Duty c ycling reduces idle listening and o v erall ener gy consumption: D C i = T i acti v e T i acti v e + T i idle (1) The node adapts its duty c ycle based on the local traf c load: T i acti v e = T min + κ · Load i (2) where T min is a minimum acti v e duration, κ is a tunable coef cient, and Load i is a number of pack ets in the b uf fer of node i . 4.5. Ener gy consumption model a. Per -state ener gy E trans = P trans · T trans (3) E recv = P recv · T recv (4) E idle = P idle · T idle (5) b . T otal ener gy E i total = E i trans + E i recv + E i idle (6) c. Ener gy per bit E bit = E total L (7) d. Ener gy ef cienc y η E = L · P success E total (8) 4.6. Latency and delay model T latenc y = T queue + T access + T trans + T prop + T proc (9) where T queue is a queuing delay , T access is a channel access delay , T trans is a transmi ssion delay , T prop is a propag ation delay , and T proc is a processing delay . AMA C-L W : Adaptive medium access contr ol for long r ang e wide ar ea network with ... (Sowmya M.) Evaluation Warning : The document was created with Spire.PDF for Python.
1634 ❒ ISSN: 2088-8708 4.7. Thr oughput and efciency Netw ork throughput: T throughput = L · P success T latenc y (10) Alternati v e form (if N success pack ets transmitted in T total time): T throughput = N success · L T total (11) a. MA C throughput ef cienc y η MA C = T throughput C (12) b . Channel utilization U = P N i =1 T i acti v e N · T c ycle (13) 4.8. Routing and pack et deli v ery a. Routing ef cienc y P success = Number of successful transmissions T otal transmissions (14) b . Congestion metric C cong = T otal traf c load C (15) where T otal traf c load = N X i =1 D i · T acti v e (16) c. Success probability P success = N success N sent (17) d. Routing cost metric Z = α · 1 − E i E max + β · 1 T latenc y + γ · P success (18) where α , β , γ is the tunable weighting f actors, E i is the current ener gy le v el of node, and i E max is the maximum (initial) ener gy . 4.9. Netw ork lifetime a. Lifetime estimate T lifetime = E battery ¯ E total /T c ycle (19) b . F orecasting ener gy consumption Using an AutoRe gressi v e inte grated mo ving a v erage (ARIMA) model: ˆ E t +1 = ϕ 1 E t + ϕ 2 E t − 1 + · · · + θ 1 ϵ t + · · · (20) where ϕ i is a autore gressi v e coef cients, θ i is a mo ving a v erage coef cients, and ϵ t is a white noise at time t . Int J Elec & Comp Eng, V ol. 16, No. 3, June 2026: 1626-1644 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 ❒ 1635 c. Objecti v e function The performance optimization goal of AMA C-L W is: max Z [ w 1 · η E + w 2 · η MA C + w 3 · J − w 4 · C cong ] (21) Subject to the constraint: E i total ≤ E i battery , ∀ i ∈ N (22) d. Performance e v aluation The protocol is e v aluated using a multi-objecti v e optimization frame w ork tar geting: − Minimize total ener gy consumption: E total − Maximize netw ork throughput: T throughput − Maximize deli v ery success: P success These objecti v es are subject to system constraints and resource limitations. 5. ALGORITHM FOR THE AD APTIVE MA C PR O T OCOL FOR LORA W AN (AMA C-L W) The algorithm inte grates the proposed model’ s components including adapti v e duty c ycling, dynam ic routing, and ener gy-ef cient communication in a LoRaW AN en vironment. Algorithm 1 outlines the w orking of AMA C-L W , an adapti v e medium access control protocol designed for LoRaW AN with inte grated ener gy-a w are routing. The core idea is to optimize netw ork communication by dynam ically adjusting node acti vity based on traf c load and selecting ener gy-ef cient paths for data transmission. The protocol be gins with initializing netw ork parameters and disco v ering neighboring nodes to form a communication graph. It emplo ys adapti v e duty c ycling, allo wing nodes to enter acti v e states only when necessary , thereby conserving ener gy . When a node detects high traf c, it prepares for transmission and e v aluates multiple routing paths using a scoring function that considers hop count, ener gy , and latenc y . The best-scoring route is chosen to forw ard the data. After transmiss ion, the node updates its ener gy consumption st atus and uses time-series forecas ting (ARIMA) to predict future ener gy trends. The system continuously monitors netw ork changes, such as node f ailures or additions, and updates its topology accordingly . This adapti v e approach ensures ef cient communication, prolonged netw ork lifetime, and resilience to dynamic changes in the LoRaW AN en vironment. 6. RESUL TS AND DISCUSSION 6.1. Dataset The LoRaMA C layer dataset pro vides an e xtensi v e o v ervie w of features follo wing the LoRaW AN specication v1.0.4 and the re gional parameters specication RP2-1.0.1, with resources lik e source code and documentation a v ailable on GitHub . The LoRaMA C layer supports LoRaW AN Classes A, B, and C, of fer - ing v aried communication modes suited for dif ferent application needs, from basic to continuous listening with minimal latenc y . Re gions co v ered incl u de multiple ISM bands such as EU868, US915, CN779, A U915, and more , making it adaptable for v arious global requirements. The layer inte grates with popular radios lik e SX1272, SX1276, SX126x, and LR1110, supporting e xible deplo yments. Additionally , it f acilitates both o v er -the-air acti v ation (O T AA) and acti v ation by personalization (ABP) for secure netw ork joi ning. Re gion selection is adjustable at runtime, enhancing adaptability , and t he data structures enable comprehensi v e opera- tions across the MA C layer , such as request handling, conrmation, and indication processes. Th i s structured API and rob ust implementation mak e it suitable for scalable and customizable LoRaW AN applications. In the three scenarios e v aluated, the Adapti v e MA C protocol for LoRaW AN (AMA C-L W) is compared ag ainst se v eral established protocols under v arying traf c conditions: lo w (10 nodes), me dium (50 nodes), and high (100 nodes). Scenario 1 (lo w traf c) in Figure 3 sho ws that AMA C-L W signicantly outperforms other protocols in throughput, achie ving a v alue of 360 compared to ALOHA ’ s 160 and LoRaW AN Class A ’ s 185. Ho we v er , latenc y remains higher for AMA C-L W (48 ms) than ALOHA (115 ms), indicating that while it is ef cient in data transmission, it may incur some delay . The ener gy consumption is minimal f o r AMA C-L W (2.4 mJ), demonstrating its ener gy ef cienc y . AMA C-L W : Adaptive medium access contr ol for long r ang e wide ar ea network with ... (Sowmya M.) Evaluation Warning : The document was created with Spire.PDF for Python.