Multi-server wireless multihop networks (MWMNs) are critical for modern communication systems, enabling efficient data transmission between devices and servers. However, the complexity of determining optimal server selection and multihop path planning in such networks often results in high interference, high network latency, low network capacity, and reduced network performance. To address these challenges, this paper proposes a two-stage network optimization scheme for MWMNs, using Broad Learning System and Q-learning, called BLSQ. First, the Broad Learning System (BLS) is employed to allocate servers to devices based on their location and computational requirements. Second, a Q-learning algorithm is introduced to optimize multihop path selection, aiming to maximize network capacity while minimizing interference. The proposed approach is evaluated based on different path selection methods in extensive simulations. Results demonstrate that our method significantly reduces network interference, increases network capacity, and achieves lower transmission time, providing a possible approach for optimizing wireless in MWMNs.

With the rapid evolution towards Beyond 5G and future 6G networks, multi-access edge computing (MEC)-enabled wireless networks are expected to support massive device connectivity, ultra-low latency, and high network capacity. However, meeting these stringent requirements in multi-server wireless multihop networks essentially requires the joint orchestration of server selection, multihop routing, and interference management. This paper develops a novel three-stage optimization scheme named broad learning system with Q-learning (BLSQ), consisting of a broad learning system-based server allocation stage, a signal-to-interference-plus-noise ratio-driven Q-learning-based multihop path selection stage, and a consensus transmit power control stage for adaptive interference mitigation. Furthermore, a consensus transmit power control mechanism is incorporated to adaptively adjust the transmit power of user devices, aiming to balance interference mitigation and throughput enhancement. The proposed scheme is particularly suitable for various mission-critical and dynamic scenarios, such as emergency communication in disaster-stricken areas, multihop data exchange between rescue teams and command centers, and flexible network deployment in large-scale events using unmanned aerial vehicles. Extensive simulation results demonstrate that the proposed BLSQ schemes outperforms existing related approaches in terms of network capacity, task completion time, interference management, and quality of servers, validating the superiority and robustness of our design for future MEC-enabled wireless networks.

A wireless multihop network (WMN) is set of wirelessly connected nodes without an aid of centralized infrastructure that can forward any packets via intermediate nodes by a multihop fashion. In the WMN, there are still some issues that need to be resolved, like due to any source node may choose an uncertainty path to send their packets through the multihop fashion and this leads to the performance of network capacity can degrade drastically. To solve this problem, in this research, we propose two novel path selection algorithms called SNR-based learning path selection (NLPS) algorithm and SINR-based learning path selection (INLPS) algorithm, which are incorporated with the deep reinforcement learning (DRL) to select the best multihop path from any source node to a destination node with highest end-to-end (E2E) throughput. Besides that, a factor graph (FG) approach and a nested lattice code (NLC) representation are used to reduce the computation time. According to the numerical studies with the NLC is applied, our simulation results reveal that the proposed NLPS and INLPS algorithms can improve the overall average network capacity up to 3.1 times and 10.5 times compared to FG, respectively. However, the overall average computation time are highly increased for NLPS and INLPS, i.e., about 0.627 s and 1.221 s, respectively compared to FG, which is about 0.006 s. In other words, both NLPS and INLPS algorithms can achieve high network capacity and moderate computation time.

In the Industrial Internet of Things (IIoT), Multi-access Edge Computing (MEC) emerges as a transformative paradigm for managing computation-intensive tasks, where task offloading plays an important role. However, due to the complex environment of IIoT, existing deep reinforcement learning-based schemes suffer from significant shortcomings in accuracy and convergence speed during model training when addressing the issue of task offloading. In this paper, to solve this problem, we propose an online task offloading scheme based on reinforcement learning, leveraging the double deep Q network (DQN) and dueling DQN with a prioritized experience replay mechanism, called thePrioritized experience-basedDoubleDuelingDQNtask offloading scheme (P-D3QN). P-D3QN enhances action selection accuracy using double DQN and mitigates Q-value overestimation by decomposing state and advantage using dueling DQN. Additionally, we adopt the prioritized experience replay mechanism to enhance the convergence speed of model training by selecting transitions that induce a higher training error between the evaluation network and the target network. Experimental results demonstrate that P-D3QN outperforms several state-of-the-art schemes, achieving a reduction of 21.0% in the average cost of the task and improving the completion rate of the task by 19.5%.

Wireless Multihop Network (WMN) is set of wirelessly connected nodes without an aid of centralized infrastructure that can forward any message via relaying nodes by multihop fashion. In WMN, there are still some issues that need to be resolved, like due to the uncertainty of source node choosing a path to send the message and the nature of multihop fashion, the performance of network capacity can degrade drastically. To solve these problems, in this research we propose two novel path selection algorithms called SNR-based learning path selection (NLPS) algorithm and SINR-based learning path selection (INLPS) algorithm, which are incorporated with the deep reinforcement learning (DRL) to select the best multihop path from source node to destination node with highest endto-end throughput. Factor graph (FG) representation is used to reduce the computation time. Our simulation results reveal that both NLPS and INLPS can achieve high network capacity and moderate computation time. Meanwhile, nested lattice code (NLC) is used in compute-and-forward strategy to reduce the time slots. As a result, the network capacity can increase more.

Wireless full-duplex (FD) transmission is one of the key drivers for improving spectrum utilization and network capacity. Many research studies have examined suppressing the residual self-interference and reducing the co-channel interference from other ongoing transmissions to realize the wireless FD transmission. The trade-off between the interference power and capacity gain still leads to a great challenge in designing the practical FD medium access control (MAC) protocol in the multihop wireless networks. This paper proposes a novel FD MAC protocol with a Mixture of Concurrent and Sequential transmission (MCST) scheme to accomplish a higher throughput by managing the transmissions and optimizing the achievable transmission capacity. Numerical simulations reveal that the proposed FD MAC with MCST scheme can achieve higher average saturation throughput with a reasonable level of average overhead ratio compared to the existing HD MAC and FD MAC protocols.

Full-duplex (FD) wireless communication improves the attainable spectral efficiency of the wireless network with the simultaneous transmission and reception over the same frequency channel at a single timeslot. However, the effect of self-interference (SI) and inter-user interference (IUI) become crucial to take into consideration compared to the current half-duplex system. Many researches have been studied to suppress the SI and IUI, to design the medium access control (MAC) protocol, and to propose resource management techniques for realizing the FD system and truly achieving the double transmission capacity in the network. In this paper, we propose a novel FD MAC protocol with a mixture of concurrent and sequential transmission scheme, namely FD-MCST, for maximizing the transmission capacity in multihop wireless networks. Through FD-MCST, the transmitting node can cooperatively share the transmission capacity status for maximizing the network capacity. As a result, numerical simulations reveal that the proposed FD-MCST can accomplish a higher achievable network capacity of up to 1.7 times and nearly twice the achievable throughput with a reasonable amount of achievable transmission overhead about 7.7%, compared to the existing FD MAC protocols.

Full-duplex (FD) communication is one of the key drivers for improving spectrum utilization. Many researches have been studied to suppress the self-interference for realizing FD communications in addition to reducing the co-channel interference from other ongoing transmissions. Since controlling the transmission power will affect the total interference and improve the overall network capacity, in this paper, we investigate an improved consensus transmit power control (CTPC) algorithm with optimal search technique to maximize the saturation throughput while reducing the transmission power and message exchange overhead in the FD wireless multihop networks. Numerical simulations reveal that the CTPC performance can achieve better performance in terms of saturation throughput, transmit power and message exchange overhead.

The video describes the novel concept of “culturally competent robotics”, which is the main focus of the project CARESSES (Culturally-Aware Robots and Environmental Sensor Systems for Elderly Support). CARESSES a multidisciplinary project whose goal is to design the first socially assistive robots that can adapt to the culture of the older people they are taking care of. Socially assistive robots are required to help the users in many ways including reminding them to take their medication, encouraging them to keep active, helping them keep in touch with family and friends. The video describes a new generation of robots that will perform their actions with attention to the older person's customs, cultural practices and individual preferences.

Smart Home is generally considered to be the final solution for human living problem, especially for health care of the elderly and disabled, power saving, etc. Human activity recognition in smart home is the key to achieve home automation, which enables smart services automatically run according to human mind. Recent researches have made several progresses in this field, however most of them can only recognize default activities which is probably not needed by smart home services. In addition, low scalability makes such researches infeasible out of laboratory. In this work, we unwrap this issue and propose a novel framework to not only recognize human activity, but also predict it. The framework contains three stages: recognition after the activity; recognition in progress and activity prediction in advance. With the help of RFID tags, the hardware cost of our framework is low enough to popularize. And the experiment result shows that our framework can realize good performance in activity recognition and prediction with high scalability.

A physical-layer network coding approach, compute-and- forward based on nested lattice code (NLC), is considered for multi-source multi-relay networks. This paper proposes a design of overlapped chunked code (OCC) which is applied before NLC, which we call OCC/CF. Random linear network coding is applied within each chunk. Only the transmissions from the sources to the relays are considered. The design is based on the empirical rank distribution and the empirical probability distributions of the participation factor of all sources. A consecutive OCC is employed with the proposed design to investigate the performance of OCC/CF. From the numerical results, the design overhead of OCC/CF is low when the probability distribution of the participation factor is dense at chunk size for each source.

This paper proposes a transmission scheme which applies random linear network coding (RLNC) over compute-and-forward (CF), called RLNC/CF, in multi-source multi-relay networks. Instead of solving the full rank failure at relays, this paper compensates for this overhead to increase the possibility of successfully decoding computed messages at the destination. The concept of the overlapped generations is applied with a proposed computing and storing strategy. This paper provides a compensation based on the estimation of the channel state information (CSI) of the previous generation and a compensation based on the learning data of CSI. By comparing to an orthogonal channel transmission scheme, a performance trade-off is considered. An expression for estimated performances of RLNC/CF in function of the probabilities of the parameters related to CSI is provided to help for the decision of selecting transmission scheme. From the numerical result, RLNC/CF scheme works better than a conventional CF transmission scheme in reducing the transmission latency.

Cultural competence is a well known requirement for an effective healthcare, widely investigated in the nursing literature. We claim that personal assistive robots should likewise be culturally competent, aware of general cultural characteristics and of the different forms they take in different individuals, and sensitive to cultural differences while perceiving, reasoning, and acting. Drawing inspiration from existing guidelines for culturally competent healthcare and the state-of-the-art in culturally competent robotics, we identify the key robot capabilities which enable culturally competent behaviours and discuss methodologies for their development and evaluation.

Real-time system is playing an important role in our society. For such a system, sensitivity to timing is the central feature of system behaviors, which means tasks in the system are required to be completed before their deadlines. Currently, almost all the practical real-time systems are equipped within multiple processors, for which the schedule synthesis to make sure that all the tasks can be completed before their deadlines is known to be an NP complete problem. In this paper, to solve the scheduling problem, we propose a scheduling method based on satisfiability modulo theories (SMT). In the method, the problem of scheduling is treated as a satisfiability problem. The key work is to formalize the satisfiability problem using first-order language. After the formalization, a SMT solver (e.g., Z3, Yices) is employed to solve such a satisfiability problem. An optimal schedule can be generated based on a solution model returned by the SMT solver. Moreover, in the SMT-based scheduling method, we define the scheduling constraints as system constraints and target constraints. Such design makes the proposed method apply more widely compared with existing methods.

In this paper we introduce a network coding aware Medium Access Control (necoMAC) scheme that incorporates many protocols such as NCA-2PSP, 2PSP and NCA-CSMA in order to provide data transmission in higher rates with fewer number of transmissions for multirate wireless networks. We create two golden topologies called golden chain and golden triangle, and calculate their energy consumption, overhead ratio, throughput and fairness for each protocol. We also set up a simulation to further analyze the performance of these protocols with the increasing number of nodes and flows. The simulation results show that the proposed scheme provides higher through-put and less energy consumption compared to the conventional CSMA/CA.

In a real-time system, tasks are required to be completed before their deadlines. Due to heavy workload, the system may be in overload condition under which some tasks may miss their deadlines. To alleviate the degrees of system performance degradation cased by the missed deadline tasks, the design of scheduling is crucial. Many design objectives can be considered. In this paper, we focus on maximizing the total number of tasks that can be completed before their deadlines. A scheduling method based on satisfiability modulo theories (SMT) is proposed. In the method, the problem of scheduling is treated as a satisfiability problem. The key work is to formalize the satisfiability problem using first-order language. After the formalization, a SMT solver (e.g., Z3, Yices) is employed to solver such a satisfiability problem. An optimal schedule can be generated based on a solution model returned by the SMT solver. The correctness of this method and the optimality of the generated schedule are straightforward. The time efficiency of the proposed method is demonstrated through various simulations. To the best of our knowledge, it is the first time introducing SMT to solve overload problem in real-time scheduling domain.

Control of data delivery latency in wireless mobile networks is an open problem due to the inherently unreliable and stochastic nature of wireless channels. This paper explores how the current best-effort throughput-oriented wireless services could be evolved into latency-sensitive enablers of new mobile applications such as remote 3D graphical rendering for interactive virtual/augmented-reality overlay. Assuming that the signal propagation delay and achievable throughput meet the basic latency requirements of the user application, we examine the idea of trading excess/federated bandwidth for the elimination of non-negligible data re-ordering delays, caused by temporal transmission failures and buffer overflows. The general system design is based on (i) spatially diverse delivery of data over multiple paths with uncorrelated outage likelihoods, and (ii) forward packet protection based on encoding redundancy that enables proactive recovery of lost or intolerably delayed data without end-to-end re-transmissions. Our analysis is based on traces of real-life traffic in live carrier-grade LTE networks.

The key factor of influencing the network capacity performance is the effect of interference power of receiving nodes, which is obtained from the other transmitting nodes in multihop wireless networks (MWNs) that are simultaneously using the same channel. Minimizing total interference power can improve overall network capacity and reduce total energy consumption. In this paper, we propose a consensus transmit power control (CTPC) algorithm to maximize end-to-end throughput in MWNs. The CTPC algorithm tunes the nodes' transmit powers to maximize the average end-to-end throughput with a consensus coefficient. Simulation results reveal that the CTPC algorithm enables all the traffic flows to accomplish the maximum average end-to-end throughput. At the same time, the total interference power and the total power consumption are decreasing. Only in the dense MWNs, under usual threshold of received signal strength indicator (RSSI) setting, the CTPC algorithm cannot achieve good performance. In addition, an advanced wmediumd emulator over the StarBED testbed is used to further verify the performance evaluation of CTPC algorithm.