Intelligent Path Selection Research Articles

An Agile Super-Resolution Network via Intelligent Path Selection

In edge computing environments, limited storage and computational resources pose significant challenges to complex super-resolution network models. To address these challenges, we propose an agile super-resolution network via intelligent path selection (ASRN) that utilizes a policy network for dynamic path selection, thereby optimizing the inference process of super-resolution network models. Its primary objective is to substantially reduce the computational burden while maximally maintaining the super-resolution quality. To achieve this goal, a unique reward function is proposed to guide the policy network towards identifying optimal policies. The proposed ASRN not only streamlines the inference process but also significantly boosts inference speed on edge devices without compromising the quality of super-resolution images. Extensive experiments across multiple datasets confirm ASRN’s remarkable ability to accelerate inference speeds while maintaining minimal performance degradation. Additionally, we explore the broad applicability and practical value of ASRN in various edge computing scenarios, indicating its widespread potential in this rapidly evolving domain.

Design of Real-Time Measurement and Intelligent Path Selection System for Network Quality under Big Data

With the emergence and development of new businesses such as video and language, people's requirements for the internet are constantly increasing. This article mainly uses the intelligent path selection method to design a real-time measurement system for network quality. On this basis, this article uses instructions to send messages to the target host, then waits for the receiver's response message, and parses the message to obtain performance indicators such as latency. This article uses intelligent path selection to query and analyze network devices, and cooperates with network devices to actively group and send data to the network. It can measure the inflow and outflow of device interfaces. The packet loss rates of path 1 during T1 and T2 are 0.066% and 0.044%, while the packet loss rates of path 2 are 0.41% and 0.11%, respectively. From the data conclusion, it can be seen that intelligent path selection can effectively improve network quality and provide users with a better internet experience.

Intelligent Path-Selection-Aided Decoding of Polar Codes

CRC-aided successive cancellation list (CA-SCL) decoding is a powerful algorithm that dramatically improves the error performance of polar codes. Path selection is a major issue that affects the decoding latency of SCL decoders. Generally, path selection is implemented using a metric sorter, which causes its latency to increase as the list grows. In this paper, intelligent path selection (IPS) is proposed as an alternative to the traditional metric sorter. First, we found that in the path selection, only the most reliable paths need to be selected, and it is not necessary to completely sort all paths. Second, based on a neural network model, an intelligent path selection scheme is proposed, including a fully connected network construction, a threshold and a post-processing unit. Simulation results show that the proposed path-selection method can achieve comparable performance gain to the existing methods under SCL/CA-SCL decoding. Compared with the conventional methods, IPS has lower latency for medium and large list sizes. For the proposed hardware structure, IPS's time complexity is O(klog2(L)) where k is the number of hidden layers of the network and L is the list size.

An Intelligent Path Selection Algorithm Based on Deep Reinforcement Learning

Path selection is the most important algorithm in intelligent devices such as robots. At present, the traditional path-planning algorithm has achieved some results, but it lacks the ability of environmental perception and continuous learning. In order to solve the above problems, this paper proposes an intelligent path selection algorithm based on deep reinforcement learning, which uses the learning ability of deep learning and the decision-making ability of reinforcement learning to realize the autonomous path planning of robots and other equipment. Simulation results show that the proposed algorithm has faster convergence, efficiency and accuracy.

Precision agriculture with cluster‐based optimal routing in wireless sensor network

SummaryAgriculture played a vital role in the growth of the Indian economy. New technologies can overcome the various issues in agriculture. One such ever‐growing area is Precision Agriculture (PA) with advancements in Wireless Sensor Network (WSN) technology. It finds applications in various segments of agriculture. The data packet of each application expects different services from the underlying network. Most of the techniques have not found much on service differentiation and arrival rate. Due to the dynamicity of the network, the traffic rate may get vary and load the selected paths which degrade performance expected from the network. These issues addressed by the proposed protocol employing Decision Support System (DSS) for intelligent path selection based on traffic rate. The proposed protocol outperforms in terms of minimized delay, energy consumption, achieving high throughput, and packet delivery ratio.

Improved Clustering with Optimization and Intelligent Path Selection

Wireless sensor networks (WSN) are widely used networking paradigm in research areas and clustering is one of the trusted mechanism for storing and processing data among the nodes in the network. The clustering makes use of a single node for transferring data to the sink node with sustaining the energy of all other nodes in the network and does the procedure of storing and processing of data using an optimization technique called as Ant Colony Optimization (ACO). The ACO is proposed in accordance along clustering which is the optimization technique used for robust communication between source node and destination node. The clustering mechanism used for sending data along the best path among the source and destination in the network so that the data can reach the destination rapidly and that helps in availing the faster response between source node and destination node. The ACO will greatly influence the dropped packets and that the delay occurring in the network based on the traffic available among the nodes in the network.

Intelligent Path Selection in Wireless Networks using Jaya Optimization

In this work, routing and mobility problems are addressed using a new Jaya Optimization Algorithm for Intelligent path selection in networks. In general, mobile nodes (MNs) are connected via wired mode or wireless mode. But in Mobile ad hoc network (MANET) MNs connected through wireless mode because nodes do not have an access point. Devices can be connected each other called as infrastructure less network whereas infrastructure based network devices can be connected through without any access point. The prevalent challenge of this network is mobility and routing. Mobility is processes of nodes are moving from one location to another location. A new Jaya Optimization algorithm is compared with other well-known mobility models like Random Waypoint, Random Direction, and Random Walk model. Also, in this study routing protocols were examined here. Furthermore, the performance results reveal that Jaya algorithm gives better results when compared with others in terms of Packet Delivery Ratio, Average Delay, Throughput and Routing Overhead.

On Efficient Network Planning and Routing in Large-Scale MANETs

In mobile ad hoc networks (MANETs), hierarchical architecture and distributed approaches are more practical than flat architecture and centralized approaches. In this paper, we propose a suite of protocols that achieve a distributed planning and routing scheme for MANETs. The proposed suite, which is composed of three protocols, offers scalability and extends network lifetime. The first protocol, i.e., the fast distributed connected dominating set (FDDS), constructs the virtual backbone by designing a fast distributed hierarchical algorithm that finds a connected dominating set (CDS) in the network graph. The constructed virtual backbone takes into account the node's limited energy, mobility, and traffic pattern. The second protocol, i.e., FDDS-M, proposes a distributed maintenance protocol that preserves the integrity of the hierarchical structure constructed by FDDS. The third protocol, i.e., FDDS-R, uses an intelligent path-selection fuzzy logic controller that can easily be incorporated in any existing link state routing protocol to select energy-efficient routes. We conducted extensive simulations that compare the operational properties (energy efficiency and network lifetime) of our schemes with others. The results show that our proposed schemes can achieve scalability and energy efficiency and outperform some of well-known approaches in the literature.

Robust and Efficient Aggregate Query Processing in Wireless Sensor Networks

Wireless sensor networks have been widely used in many applications, such as soil temperature monitoring for plant growth and abnormal event detection of industrial parameters. Among these applications, aggregate queries, such as SUM, COUNT, AVERAGE, MIN and MAX are often used to collect statistical data. Due to the low quality sensing devices or random environmental disturbances, sensor data are often noisy. Hence, the idea of moving average, which computes the average over consecutive aggregate data, is introduced to offset the effect. The high link loss rate, however, makes the result after averaging still inaccurate. To address this issue, we propose a PCM-based data transmission scheme to "make up" the possibly lost data. Specifically, we focus on obtaining robust aggregate results under high link loss rate. In order to reduce the communication traffic that dominates the energy consumption of the sensor network, we also design an intelligent path selection algorithm for our scheme. Our extensive simulation results have shown that this technique outperforms its counterparts under various sensor network conditions.