Intelligent Dynamic Spectrum Resource Management Based on Sensing Data in Space-Time and Frequency Domain

Abstract

Edge computing offers a promising paradigm for implementing the industrial Internet of things (IIoT) by offloading intensive computing tasks from resource constrained machine type devices to powerful edge servers. However, efficient spectrum resource management is required to meet the quality of service requirements of various applications, taking into account the limited spectrum resources, batteries, and the characteristics of available spectrum fluctuations. Therefore, this study proposes intelligent dynamic spectrum resource management consisting of learning engines that select optimal backup channels based on history data, reasoning engines that infer idle channels based on backup channel lists, and transmission parameter optimization engines based genetic algorithm using interference analysis in time, space and frequency domains. The performance of the proposed intelligent dynamic spectrum resource management was evaluated in terms of the spectrum efficiency, number of spectrum handoff, latency, energy consumption, and link maintenance probability according to the backup channel selection technique and the number of IoT devices and the use of transmission parameters optimized for each traffic environment. The results demonstrate that the proposed method is superior to existing spectrum resource management functions.

Intelligent Dynamic Real-Time Spectrum Resource Management for Industrial IoT in Edge Computing

“…In intelligent dynamic spectrum resource management, case-based reasoning (CBR) is divided into CR Mater and CR User roles as shown in Figure 5 , and retrieving, reusing and retain operation based on the case DB created through data mining is performed [ 13 , 33 ].…”

“…In this study, the process of analyzing the performance of intelligent dynamic real-time spectrum resource management using data mining and case-based reasoning in a CR network environment that is dynamically variable in the space–time and frequency domains consists of three steps, as shown in Figure 9 below [ 13 , 18 , 37 ].…”

“…The number of spectrum handoffs is defined as the number of times that data transmission is stopped and switched to another idle channel according to the incumbent user detection during one session of CR data transmission in the idle channel reasoning process [ 13 , 45 , 46 ]. However, as frequent spectrum handoff operations can degrade the communication performance based on latency and increase the energy consumption, this simulation aims to minimize unnecessary spectrum handoffs in order to realize an efficient CR network.…”

“…The spectrum handoff latency represents the number of channel searches that occur during the handoff process to an idle channel according to the detection of an incumbent user [ 13 , 45 , 47 ]. To verify a channel selection technique that can quickly handle an idle channel with a small number of searches during the reasoning period, as shown in Equations ( 25 ) and ( 26 ), the number of channel searches occurring during the handoff process for each time slot is counted according to the backup channel order: …”

“…To achieve this, the previously studied intelligent dynamic spectrum resource management [ 13 ] constitutes an infrastructure-based WRAN in which a base station utilizing the TV White Space (TVWS) band controls IoT devices in a suburban area with limited broadband access support according to the IEEE 802.22 WRAN standard [ 14 , 15 , 16 ]. Each base station forms a single wireless network and has a learning, reasoning, and optimization engine structure made possible through database access, enabling dynamic spectrum resource management for coexistence between cells of the same network type existing in adjacent channels and locations.…”