There are many heterogeneous intelligent devices with different resources in the field environment lacking in-frastructure. Some of these devices have insufficient computing power and power supply, seriously affecting the functioning of computing-intensive and delay-sensitive applications. To fully utilize the available computing resources scattered in the region and provide deterministic quality of service guarantees for task execution, we designed Universal Identity Tag (UIT), a universal resource identification method. UIT can identify multi-dimensional resources and extract task characteristics, which effectively supports the heterogeneous communication entities computing resource discovery process and provides a basis for task scheduling. Subsequently, we propose Computing-Aware Strategy (CAS), a task scheduling algorithm based on ad hoc cloud and computing power awareness. CAS allows intelligent devices to dynamically perceive the state of other mobile devices to optimize the task scheduling process. Therefore, regional devices can share data and collaborate on computing tasks. Simulation results show that the proposed scheme effectively reduces data traffic overhead and lowers the average service delay.

Mobile Edge Computing (MEC) has solved a sharp increase in data volume caused by various emerging network applications. The edge data center is an essential part of MEC, which connects the edge of the network and the backbone network. Faced with a complex network environment, edge data centers suffer low bandwidth resource utilization and high network latency. This paper proposes Twin Delayed Deep Deterministic policy gradient based Routing Optimization (TRO) algorithm to improve the performance of edge data centers. The TRO algorithm uses Deep Reinforcement Learning (DRL) and Software-Defined Networking (SDN) to achieve routing optimization from two aspects of bandwidth utilization and load balancing. Experiments demonstrate that compared with other algorithms, the TRO algorithm proposed in this paper significantly improves network throughput and reduces average packet latency and average packet latency error.

Aiming at the problem of network performance degradation caused by the characteristics of mobile ad hoc network, this paper proposes a cooperative packet forwarding algorithm based on game combination and third-party altruism. First of all, we analyze the static game among nodes, and calculate the forwarding probability of nodes in the case of static game through the theorem of Nash equilibrium. Then, for the sake of solving the shortcomings of static game, we introduced the self-game model. Next, this paper puts forward the definition of node's benefit, which is vague in many studies, and proposes the concept of incentive factor based on third-party altruism to improve node collaboration. Finally, using the static game and self-game, combined with the incentive factor to calculate the node forwarding probability, the data packet cooperative forwarding algorithm is designed. The simulation results demonstrate that the proposed algorithm improves the performance indicators of MANET such as packet delivery ratio, average transmission delay, throughput and routing overhead.

In recent years, sensor multimodal time series data fusion has attracted widespread attention. One of the key challenges is to extract features from multimodal data to obtain shared representations, which combines the characteristics of time series data to further improve prediction performance. To solve this problem, based on Stacked Sparse Auto-Encoder (SSAE) and Long Short-Term Memory (LSTM), we propose a multimodal time series data fusion model called SSAE-LSTM. SSAE mines the inherent correlation features of multimodal data to extract a good shared representation, which is used as the input of the LSTM neural network to perform data fusion processing. Experiments on real time series datasets demonstrate that SSAE-LSTM can obtain a good shared representation of multimodal data to predict the future development trend. Compared with other neural networks, SSAE-LSTM has better performance in Precision, Accuracy, Recall, F-score and so on.

We propose a task distribution algorithm based on agent correlation to enhance the comprehensive research on the distribution sensing tasks in mobile sensor networks. First, in the proposed algorithm, the score and feature factors of the mobile agents are considered comprehensively in a direct correlation model, which is constructed by combining the direct and indirect correlation samples. Second, we introduce a mobility model based on the exponential distribution, and obtain a calculation method of probability parameter λ according to the analysis presented in this article. Finally, we integrate the constructed correlation model and mobility model; which are then applied to the distribution algorithm. By experimental research, the task distribution algorithm based on relationships of agents we proposed, has a precision improvement of 57.23% over MTPS, which is a fine-grained multitask allocation framework algorithm, and 12.31% of that to sequential MF, which is a recommendation algorithm based on collaborative filtering by exploiting sequential behaviors. These results indicate that the proposed algorithm improves the task distribution performance significantly, and offers a more accurate and reliable service.

With the popularization of intelligent hardware, wireless sensor networks have led to mobile crowdsensing (MCS) systems, which provide solutions for large-scale and complex urban data collection. Task distribution is the most important part of intelligent hardware applications. MCS can improve the task distribution efficiency by accurately predicting the location of a perceived user for task distribution. This paper proposes a task location estimator based on a variable-order Markov time window sensing (TEMTWS) algorithm. This method is based on time window modeling, and the association between user tasks is established by sensing the historical track data of user execution tasks. First, the task execution frequency and task vector are calculated, and the organizer at each position is selected. To obtain more perceptual users, similarity estimation is performed on the users and organizers within the time window, and users with high relevance are grouped into the same cluster. An experiment is conducted with the Gowalla dataset to verify the algorithm. The results show that the proposed algorithm outperforms the standard Markov K-means algorithm and K-means-GA algorithm in terms of the prediction accuracy.