In this paper, we propose a novel framework to integrate the intelligent reconfigurable surface (IRS) in terahertz (THz) networks, with co-existing ultra-reliable low-latency communication (URLLC) and enhanced mobile broadband (eMBB) services. To meet URLLC latency constraints, the URLLC traffic is scheduled along with eMBB traffic, which poses a significant resource allocation challenge. To address this concern, in this paper, we propose a joint power and service allocation (JPSA) framework to maximize URLLC reliability while ensuring eMBB data rates. Furthermore, to address the challenging NP-hard mixed-integer nonlinear programming (MINLP) problem, we decompose the resource allocation problem into the URLLC power allocation and service allocation sub-problems. More specifically, we suggest a one-to-one matching game for service allocation. Our simulation results demonstrate that the proposed scheme outperforms baseline methods, particularly in terms of enhancing the reliability of URLLC user equipment (uUEs).

In quasi-static wireless networks characterized by infrequent changes in the transmission schedules of user equipment (UE), malicious jammers can easily deteriorate network performance. Accordingly, a key challenge in these networks is managing channel access amidst jammers and under dynamic channel conditions. In this context, we propose a robust learning-based mechanism for channel access in multi-cell quasi-static networks under jamming. The network comprises multiple legitimate UEs, including predefined UEs (pUEs) with stochastic predefined schedules and an intelligent UE (iUE) with an undefined transmission schedule, all transmitting over a shared, time-varying uplink channel. Jammers transmit unwanted packets to disturb the pUEs’ and the iUE’s communication. The iUE’s learning process is based on the deep reinforcement learning (DRL) framework, utilizing a residual network (ResNet)-based deep Q-Network (DQN). To coexist in the network and maximize the network’s sum cross-layer achievable rate (SCLAR), the iUE must learn the unknown network dynamics while concurrently adapting to dynamic channel conditions. Our simulation results reveal that, with properly defined state space, action space, and rewards in DRL, the iUE can effectively coexist in the network, maximizing channel utilization and the network’s SCLAR by judiciously selecting transmission time slots and thus avoiding collisions and jamming.

This paper investigates a deep reinforcement learning (DRL)-based approach for managing channel access in wireless networks. Specifically, we consider a scenario in which an intelligent user device (iUD) shares a time-varying uplink wireless channel with several fixed transmission schedule user devices (fUDs) and an unknown-schedule malicious jammer. The iUD aims to harmoniously coexist with the fUDs, avoid the jammer, and adaptively learn an optimal channel access strategy in the face of dynamic channel conditions, to maximize the network's sum cross-layer achievable rate (SCLAR). Through extensive simulations, we demonstrate that when we appropriately define the state space, action space, and rewards within the DRL frame-work, the iUD can effectively coexist with other UDs and optimize the network's SCLAR. We show that the proposed algorithm outperforms the tabular Q-learning and a fully connected deep neural network approach.

Terahertz (THz) communication is a promising technology for future wireless communications, offering data rates of up to several terabits-per-second (Tbps). However, the range of THz band communications is often limited by high pathloss and molecular absorption. To overcome these challenges, this paper proposes intelligent reconfigurable surfaces (IRSs) to enhance THz communication systems. Specifically, we introduce an angle-based trigonometric channel model to evaluate the effectiveness of IRS-aided THz networks. Additionally, to maximize the sum rate, we formulate the source-IRS-destination matching problem, which is a mixed-integer nonlinear programming (MINLP) problem. To solve this non-deterministic polynomial-time hard (NP-hard) problem, the paper proposes a Gale-Shapley-based solution that obtains stable matches between sources and IRSs, as well as between destinations and IRSs in the first and second sub-problems, respectively.

In this paper, we propose an innovative self‐organizing medium access control mechanism for a distributed cognitive radio network (CRN) in which utilization is maximized by minimizing the collisions and missed opportunities. This is achieved by organizing the users of the CRN in a queue through a timer and user ID and providing channel access in an orderly fashion. To efficiently organize the users in a distributed, ad hoc network with less overhead, we reduce the sensing period through parallel sensing wherein the users are divided into different groups and each group is assigned a different portion of the primary spectrum band. This consequently augments the number of discovered spectrum holes which then are maximally utilized through the self‐organizing access scheme. The combination of two schemes augments the effective utilization of primary holes to above 95%, even in impasse situations due to heavy primary network loading, thereby achieving higher network throughput than that achieved when each of the two approaches are used in isolation. By efficiently combining parallel sensing with the self‐organizing MAC (PSO‐MAC), a synergy has been achieved that affords the gains which are more than the sum of the gains achieved through each one of these techniques individually. In an experimental scenario with 50% primary load, the network throughput achieved with combined parallel sensing and self‐organizing MAC is 50% higher compared to that of parallel sensing and 37% better than that of self‐organizing MAC. These results clearly demonstrate the efficacy of the combined approach in achieving optimum performance in a CRN.