Congratulations!  You have been automatically subscribed to IEEE DataPort and can access all datasets on IEEE DataPort!
First Name: 
Last Name: 

Datasets & Competitions

The dataset consists of training and test data and label matrices for single-pixel compressive DoA estimation for mmWave metasurface. The dataset will be uploaded soon. Currently, a small part of the dataset can be accessed through this repository. For detailed information, please visit Graph Attention Network Based Single-Pixel Compressive Direction of Arrival Estimation through


This dataset covers cellular communication signals in the SCF format. There is a total of 60000 signal instances, 36000 of them are reserved as training data and the rest is for the test. The SNR levels are between 1 dB and 15 dB.


For each SNR level, the training dataset has four files. The user can concatenate these files. The same procedure is valid for the test dataset.


The labels mean:


0 -> AWGN (no signal in the spectrum)


1 -> UMTS


2 -> LTE


3 -> GSM


The experimental measurement setup is performed in an anechoic chamber to secure the LOS transmission at the Millimeter Wave and Terahertz Technologies Research Laboratories (MILTAL), Scientific and Technological Research Council of Turkey (TUBİTAK), Kocaeli. The dimensions of anechoic chamber are 7m × 4m × 3m. The dataset includes channel impulse response in between 240GHz and 300GHz for the transmitter-receiver distance of 20cm, 30cm, 40cm, 60cm, and 80cm.


The file contains channel impulse responses for five transmitter-receiver separations. All information about the data set can be found in the article "Modeling and Analysis of Sub - Terahertz Communication Channel via Mixture of Gamma Distribution". If you use the data set, please cite this paper. Do not hesitate the ask question about the dataset.


In order to increase the diversity in signal datasets, we create a new dataset called HisarMod, which includes 26 classes and 5 different modulation families passing through 5 different wireless communication channel. During the generation of the dataset, MATLAB 2017a is employed for creating random bit sequences, symbols, and wireless fading channels. 



Documentation will be available soon.