The rapid development of smart cities using edge technologies brings forth a demand for managing IoT data and critical operations in real-time systems [1,2]. Wireless systems and next-generation IoT are combined in different ways for significant growth in healthcare systems [3,4]. Future sensing technologies [5,6] with IoT systems observe the targeted areas and ease the rapid data analysis for health operations [7,8]. Cloud-centric data processing models have struggled to meet such applications’ dynamics, especially when balancing scalability with sustainability [9,10]. Despite advancements, several IoT-based healthcare applications rely on centralized cloud storage, vulnerable and ineffective for data breaches [11,12]. These systems often lack resilience and scalability research issues for large-scale architecture [13,14]. Consequently, due to the demands of end users, there is an increasing shift toward distributed architectures for healthcare systems with edge computing to ensure continuity and fault-tolerant communication [15–17]. In addition, establishing trust between constrained devices is another significant way to attain a reliable monitoring system and resource optimization. In smart cities, enhancing security is crucial for avoiding malicious traffic and preventing unauthorized access to sensitive data [18,19]. It can compromise the network infrastructure and disrupt essential services, thus adapting robust and secured measurement promotes resilient communication against attacks to guarantee a trustworthy environment [20,21]. The contributions of the proposed model are highlighted as follows:

It investigated stochastic routing techniques with multiple factors and network uncertainties to develop a recursive decision-making system for network optimization.

The research work is structured as follows. Section 2 presents related work. The proposed model is described in Section 3. Section 4 provides performance results. In the end, Section 5 concludes this research work.

Edge computing with IoT systems has performed a crucial role in the growth of smart cities and sustainable development [22,23]. It enables seamless and timely interaction between wireless technologies by utilizing gateway devices and ensuring efficient communication in a real-time environment [24,25]. The rapid evolution of edge-driven processing systems, and autonomous IoT networks facilitate and support dynamic data analysis with learning techniques [26–28]. The environment continues monitoring the targeted data and forwards it to the cloud users based on their demands to attain a real-time surveillance system. The authors in [29] proposed an IoT network architecture for energy vehicles (EV) by using a data fusion technique based on fuzzy logic. The proposed data fusion algorithm considers human input from the crowd and sensory data and computes the location-specific congestion. Moreover, an open source routing (OSRM) determines the shortest congestion-aware route by reacting to the collected real-time traffic updates. Based on fuzzy logic, authors [30] developed a Trusted Routing Protocol for Vehicular Cloud Networks (FTRP), which constructs the protected paths for the delivery of data. A node candidacy value is computed using fuzzy logic and accordingly route is either selected or rejected. Cloud assigns a confidence score to each node based on the information it obtained from other nodes. In addition to analyzing multiple factors, the proposed approach computes the trust for selecting a secure route. A novel blockchain-based framework is proposed in [31] to guarantee data security and integrity for IoT systems. It comprised various phases and integrated with diverse security mechanisms for robust security in terms of privacy and data integrity. The experimental analysis of the proposed SecPrivPreserve revealed that its performance improved than existing studies. Authors [32] introduced an intelligent approach to determine the set of optimal border nodes by exploring the Lion Swarm Optimization algorithm (LSOA). The network area is divided into clusters and determines the set of border nodes within each cluster and associated inter-cluster communication. The main factors for finding the optimal border nodes are maximum energy and minimum distance. Later, a routing mechanism is generated based on the selection of optimal border nodes. Fuzzy logic-based secure hierarchical routing scheme is proposed [33] by utilizing the firefly algorithm (FSRF). It is developed for IoT-based healthcare systems and is composed of three main frameworks. To evaluate the trust, a fuzzy logic-based trust framework is presented that tackles with different routing attacks. In addition, using various metrics, the proposed framework supports the clustering of IoT devices and identifies the effective cluster head nodes. Moreover, FSRF provides an on-demand routing framework to attain consistent and energy-efficient routes for data transmission. Authors [34] integrate software-defined networking (SDN) and blockchain technology to cope with energy efficiency and security research obstacles. Consequently, a novel energy protocol combined with a cluster structure was proposed to develop a secure and blockchain-enabled SDN controller architecture for IoT networks. Because it uses public and private blockchains for peer-to-peer (P2P) communication between IoT devices and SDN controllers, eliminating Proof-of-Work, the distributed trust architecture is suitable for resource-constrained IoT devices. Authors [35] proposed a security strategy for the detection and prevention of forwarding attacks from healthcare IoT systems, and composed of five phases. Initially, a topology is developed for finding the cluster heads and discovering the best route. In addition, to identify the selective forwarding attacks, the proposed solution performed packet validation based on the details of transmitted and received packets. Table 1 presents a benchmarking comparison of existing solutions alongside our proposed model.

This section proposes a model for addressing the research challenges that are demanded in healthcare IoT applications. Our proposed model introduces resource optimization, enhancement in healthcare security, and lightweight processing of real-time data using intelligence of network edges. The following subsections describe the flow of the proposed model.

In this section, we evaluate the performance of the proposed model with related work through simulations. The testing environment is created in NS-3 and scripting files are used to observe the behavior of the network. The collected data from health sensors are securely transmitted through a combination of blockchain transactions and maintain privacy and data integrity. The edges are responsible for resource management, data encryption, trust evaluation, and blockchain validation. The network field is fixed to 5000 m × 5000 m, populated by 50 to 200 health sensors, with 2j of initial energy level. 35 edge devices are deployed for the support of data aggregation and validation. To evaluate the real-world applicability of the proposed model, Table 3 depicted the varied key parameters.

The evaluation of the packet drop ratio for the proposed model and existing solutions under varying transmission rates and bandwidth is depicted in Fig. 3a,b. The experimental results illustrate that the proposed model outperforms existing approaches by an average of 33% and 38%. It employs the concept of edge computing and intelligent route identification while attaining reliable communication channels for the healthcare system. Robust decision-making using stochastic optimization also decreases the chances of network disruption and enhances resource optimization in e-health operations. Furthermore, various parameters are re-computed to balance the load on the transmission links with efficient traffic distribution on the constraint devices. Additionally, edges are mobile and crucial components of the proposed model, they intelligently collected the IoT aggregated data and decreased the transmission distance with effective trust management. Moreover, the security analysis is another significant stage of the proposed model, it reduces the communication threats and provides more trustworthiness forwarding based on blockchain transactions.

The proposed model’s experimental results compared to the existing solution are illustrated in Fig. 4a,b. The proposed model balances resource consumption and node congestion to optimize sensor performance with the use of distributed computing, which is based on multi-facet analysis. It was noticed that the proposed model increases the energy efficiency by an average of 41% and 52% under different transmission rates and bandwidths, owing to the selection of several criteria parameters by exploring network conditions. The network edges not only limit the transmission power of the health sensors but also eliminate congested and faulty links for data transmission. Accordingly, the next-hop is more reliable and consistent for continuous routing and offers high performance in e-health data delivery. The routing paths are continuously updated by identifying nearby trustworthy neighbors based on the trust scores, and enhancing the routes’ lifetime while decreasing traffic load on network-wide IoT devices. In addition, routes are authenticated using blockchain integration and marked as infectious entries incurred during data transmission.

Compared to existing solutions, the proposed model improves the computational time for varying the data generation rate and bandwidth by an average of 38% and 46%, as illustrated in Fig. 5a,b. This is because forwarding routes are re-evaluated at both the device and edge level. In case, any device in the route is malicious, the entire route is labeled as faulty, and the route is reformulated for transmission of e-health records. In addition, mobile edges perform a crucial role in the identification of faulty channels based on trust and blockchain integration. It copes to identify the unauthentic devices on the medium and avoid the keys or resending the fake route requests. If any channel closer to the edge device is suspicious, then the edge device announces alarming messages to inform the neighbors that fall in its proximity. In such cases, the proposed algorithm effectively manages the communication cost and latency rate.

In Fig. 6a,b, the performance evaluation of the proposed model is compared with existing solutions in terms of blockchain overhead. Based on the results, it was noticed that performance results improved for varying data generation rate and bandwidth by an average of 45% and 49% by utilizing a more reliable anomaly detection mechanism using integration of lightweight blockchain computation. Moreover, the support of network edges along with the lightweight cryptographic methods, dynamic thresholds, and validation patterns, ensure the proposed model for efficiently detecting misclassification of normal data traffic as communication threats. Unlike existing work, this strategy enables the optimal usage of network resources and prevents additional blockchain overhead caused by false positives. Consequently, the proposed model reduces the chances of false alerts and offers more trustworthiness in blockchain-based security mechanisms for crucial healthcare systems.

Smart cities are interconnected using advanced network technologies and intermediate wireless services to observe the unpredicted environment. In crucial applications, the bounded limitation of sensors, IoT gateways, etc., are not enough to provide an earlier response in the decision-making process and raise significant challenges for network optimization. In addition, the security of distributed systems also poses another research issue in attaining data preservation and authentication. This work presents a blockchain-driven secured and load-balanced routing model for health applications with efficient data flow supported by edge intelligence. It leads to a lightweight communication system using stochastic optimization and enhances trustworthiness between connected devices despite network threats. Moreover, the proposed model secures the exchange of IoT data using integrated blockchain transactions and reduces the probabilities of anomalous behaviors. However, to enhance the robustness of the proposed model, adaptive resource allocation algorithms with lightweight encryption techniques are demanded to decrease the computing load on low-powered devices.