LSTM Based Spectrum Prediction for Real-Time Spectrum Access for IoT Applications

Abstract

In the Internet of Things (IoT) scenario, many devices will communicate in the presence of the cellular network; the chances of availability of spectrum will be very scary given the presence of large numbers of mobile users and large amounts of applications. Spectrum prediction is very encouraging for high traffic next-generation wireless networks, where devices/machines which are part of the Cognitive Radio Network (CRN) can predict the spectrum state prior to transmission to save their limited energy by avoiding unnecessarily sensing radio spectrum. Long short-term memory (LSTM) is employed to simultaneously predict the Radio Spectrum State (RSS) for two-time slots, thereby allowing the secondary node to use the prediction result to transmit its information to achieve lower waiting time hence, enhanced performance capacity. A framework of spectral transmission based on the LSTM prediction is formulated, named as positive prediction and sensing-based spectrum access. The proposed scheme provides an average maximum waiting time gain of 2.88 ms. The proposed scheme provides 0.096 bps more capacity than a conventional energy detector.

A Deep Long-Term Joint Temporal–Spectral Network for Spectrum Prediction

“…Baselines and Evaluation Metrics: For comparison, 10 prediction models based on deep learning are considered as baselines, which are LSTM [ 18 ], GRU [ 32 ], BiLSTM [ 33 ], CNN [ 34 ], CNN-LSTM [ 35 ], CNN-BiLSTM [ 25 ], CNN-BiLSTM-attention [ 36 ], seq2seq-LSTM [ 37 ], seq2seq-LSTM-attention [ 38 ], and seq2seq-Bi-ConvLSTM [ 39 ]. In order to fairly compare the performance of models, the input information and output information of all prediction models are all the same and the baseline models use the multi-step principle to output results.…”

“…Recurrent neural network (RNN)-based methods (Long Short-Term Memory (LSTM) and gated recurrent unit network (GRU)) [ 17 ] are capable of mining underlying temporal correlations among spectrum data. In [ 18 ], LSTM was employed to simultaneously predict the Radio Spectrum State (RSS) for two time slots, which requested a large amount of computing resources and suffers from very long training time. To solve the above problems, Ling Yu et al [ 19 ] introduced the Taguchi method and LSTM for time domain spectrum prediction, effectively reducing the time and computational power requirements.…”

Design of wireless battery management system monitoring and automated alarm system based on improved long short-term memory neural network

“…The results show that more accurate battery status can be obtained by fully using the cloud platform’s storage and computing resources. Nandakumar, Ponnusamy & Mishra (2023) proposed a cloud BMS to improve battery systems’ computing power and data storage capacity through cloud computing. Discusses the application of an equivalent circuit model in the digital twin system of the battery system, which improves the computing power, data storage capacity and reliability of BMS.…”