11 October 2019 in IEEE Access
1 October 2020 in IEEE Access
Mobile Edge Computing (MEC) has emerged as an alternative to cloud computing to meet the latency and Quality-of-Service (QoS) requirements of mobile devices. In this paper, we address the problem of server resource allocation in MEC. Due to the dynamic load conditions on MEC servers, their resources need to be used intelligently to meet the QoS requirements of the users and to minimize server energy consumption. We present a novel resource allocation algorithm, called Power Migration Expand (PowMigExpand). Our algorithm assigns user requests to the optimal server and allocates optimal amount of resources to User Equipment (UE) based on our comprehensive utility function. PowMigExpand also migrates UE requests to new servers, when needed due to the mobility of users. We also present a low cost Energy Efficient Smart Allocator (EESA) algorithm that uses deep learning for energy efficient allocation of requests to optimal servers. The proposed algorithms consider varying load of incoming requests and their heterogeneous nature, energy efficient activation of servers, and Virtual Machine (VM) migration for smart resource allocation and, thus, is the first comprehensive approach to address the complex and multidimensional resource allocation problem using deep learning. We compare our proposed algorithms with other resource allocation approaches and show that our approach can handle the dynamic load conditions better. The proposed algorithms improve the service rate and the overall utility with minimum energy consumption. On average, it reduces 26% energy consumption of MESs and improves the service rate by 23%, compared with other algorithms. We also get more than 70% accuracy for EESA in allocating the resources of multiple servers to multiple users.
1 October 2021 in IEEE Internet of Things Journal
Massive Internet of Things (IoT) connectivity requires addressing spectrum congestion caused by spectrum scarcity in wireless communications. Over the past decade, cognitive radio (CR) has been proposed as a promising solution to utilize the licensed spectrum efficiently. Conventional spectrum-sensing approaches are complex and require statistical information about the behavior of licensed users, which is impractical. To overcome this limitation, several reinforcement-learning (RL)-based spectrum-sensing approaches have been proposed that are also highly adaptable to the dynamics of IoT environments. Additionally, cooperative-RL-based spectrum-sensing approaches have been widely used because they are more accurate than noncooperative approaches. However, the advantage comes at the cost of scalability due to increased information-sharing overhead. Furthermore, cooperative spectrum-sensing (CSS) approaches suffer from attacks on the network, such as sensing data falsification (SDF) attacks, which deteriorate sensing accuracy dramatically. In this article, we present a scalable, partially CSS algorithm that is highly resilient to SDF attacks. The novelty of the proposed algorithm lies in partial cooperation through coalition formation, which reduces sensing and information sharing overhead while improving sensing accuracy. Moreover, the algorithm learns to adapt the sensing participation percentage and selects the most rewarding channel for sensing to maximize rewards while minimizing energy consumption. The proposed algorithm outperforms state-of-the-art CSS algorithms in terms of sensing accuracy and overheard. Contrary to centralized CSS algorithms, the proposed algorithm’s performance is directly proportional to the number of devices; hence, it is suitable for massive connectivity.
19 June 2024 in IEEE Open Journal of the Communications Society
Cognitive radio networks (CRNs) mitigate spectrum scarcity by leveraging the holes in the licensed spectrum to enable internet of things (IoT) devices to opportunistically access the spectrum. However, IoT devices need to sense the spectrum before they can access it, which is an energy-intensive process and hinders the practical implementation of opportunistic spectrum access for energy-constrained IoT devices. In this context, reinforcement learning-based algorithms that encourage cooperation among IoT devices to eliminate the need for constant sensing are promising candidates for practical CRN implementation. As exciting as the application of reinforcement learning to CRNs is, benchmarking the performance of different algorithms is a huge challenge due to a lack of standardized comparison metrics, especially for hybrid action spaces that comprise both discrete and continuous actions. We propose a hybrid discrete-continuous space deep reinforcement learning algorithm that maximizes the energy efficiency of CRNs by optimizing sensing, cooperation, and transmission by IoT devices.We also analyze the algorithm’s performance by setting the theoretical upper bound for throughput and find that it reaches 99.4% of the theoretical upper bound, while its discrete action-space version reaches 96% and other baseline algorithms range between 70% and 86%.
13 May 2026 in IEEE Open Journal of the Communications Society
High-altitude platform stations (HAPS) offer a pragmatic backbone for extending 6G non-terrestrial networks (NTNs) into dense urban cores, sparsely connected remote regions, and disaster-stricken areas. In these environments, cognitive radio networks (CRNs) must arbitrate spectrum access without trusting all agents or relying on intact terrestrial backhaul. Accordingly, in the present study, we propose a federated hierarchical reinforcement learning (F-HRL) framework where secondary users (SUs) cooperatively access primary users’ (PUs) spectrum, ground next-generation Node Bs (gNBs) and unmanned aerial vehicles (UAVs) enforce trust-aware scheduling, while HAPS coordinators perform Byzantine-resilient aggregation using satellite backhaul. Area-specific reward shaping allows the same policy stack to prioritize spectral efficiency in urban deployments, coverage in remote regions, and rapid recovery after infrastructure loss. We benchmark F-HRL against non-cognitive baselines, simple heuristics, and the hierarchical deep reinforcement learning framework across the following three key scenarios over 5 000 decision steps: nominal spectrum sharing, disaster recovery with a gNB outage, and adversarial environments with malicious SUs. Experimental results demonstrate that F-HRL achieves a nearly 8-fold higher spectrum efficiency than the non-cognitive operation, maintains 86% SU quality-of-service (QoS) satisfaction with a Jain fairness index of 0.950, and sustains over 95% of nominal throughput during infrastructure failures. Under attack conditions with 25% compromised agents, trust-weighted aggregation successfully isolates malicious updates while preserving legitimate throughput. These findings demonstrate that HAPS-centric F-HRL provides adaptive, secure, and resilient spectrum management for future NTN deployments.