Histograms to Quantify Dataset Shift for Spectrum Data Analytics: A SoC Based Device Perspective

Abstract

Cloud/software-based wireless resource controllers have been recently proposed to exploit radio frequency (RF) data analytics for a network control, configuration and management. For efficient resource controller design, tracking the right metrics in real-time (analytics) and making realistic predictions (deep learning) will play an important role to increase its efficiency. This factor becomes particularly critical as radio environments are generally dynamic, and the data sets collected may exhibit shift in distribution over time and/or space. When a trained model is deployed at the controller without taking into account dataset shift, a large amount of prediction errors may take place. This paper quantifies dataset shift in real wireless physical layer data by using a statistical distance method called earth mover's distance (EMD). It utilizes an FPGA to process in real-time the inphase and quadrature (IQ) samples to obtain useful information, such as histograms of wireless channel utilization (CU). We have prototyped the data processing modules on a Xilinx System on Chip (SoC) board using Vivado, Vivado HLS, SDK and MATLAB tools. The histograms are sent as low-overhead analytics to the resource controller server where they are processed to evaluate dataset shift. The presented results provide insight into dataset shift in real wireless CU data collected over multiple weeks in the University of Oulu using the implemented modules on SoC devices. The results can be used to design approaches that can prevent failures due to datashift in deep learning models for wireless networks.

Confidence Aware Deep Learning Driven Wireless Resource Allocation in Shared Spectrum Bands

“…The authors in [19] present a DL based method to solve the problem of sub-band and power allocation in a multi-cell network. All of the preceding works are based on the assumption of similar distribution for the training and target datasets which is not always true in reality [20]. Therefore, those works can exhibit suboptimal performance when deployed in a real wireless environment [21].…”

Concept Drift Detection Methods for Deep Learning Cognitive Radios: A Hardware Perspective

“…Once the bin size is selected, different threshold values can be selected for each bin. Whenever a corresponding value arrives, the value of the bin is incremented [3]. The histograms are typically normalized by dividing the value of a bin with the total value of bins in the entire histogram.…”

“…To our best knowledge, this is the only work that includes VLSI architecture and FPGA implementation of both histogram computation and concept drift for cognitive radios. FPGA implementation of histogram and a concept drift method is previously presented in [3] with the aid of high level synthesis, and thus, the detail of VLSI architecture is missing in [3].…”

“…However, radio frequency (RF) measurements from real data traffic indicates that the exact distribution of traffic is sometimes unknown and model mismatch is unavoidable [3]. The model mismatch is commonly known as concept drift or dataset shift in the literature [4].…”