Dataset Title:  Attention-Based Receive Beamforming

Dataset Description: Time series of received snapshots and covariance matrices

This dataset has been used in this work (please cite this reference in your work if you make use of this dataset):

• A. Saifaldawla, F. Ortiz, E. Lagunas, S. Chatzinotas, “Attention-based Blind Adaptive Receive Beamforming for Interference Limited NGSO Satellite Systems”, submitted.

Detailed description:

A realistic NGSOs Co-frequency interference (CFI) datasets created via MATLAB satellite communication toolbox.

Detailed description:

Three are three main datasets groups we use in the paper mentioned above:

1. Data1_M64_L64: The received signal snapshots “L” is equal to number of antenna elements “L= M”,
2. Data2_M64_L128: The received signal snapshots “L” is twice the number of antenna elements “L= 2xM”,
3. Data2_M64_L256: The received signal snapshots “L” is four times the number of antenna elements “L= 4xM”.

Each Folder “dataset” has .mat files are as follows:

1. Train Data (N_tr = 9000) : 30 desired satellite passes, repeated 3 times with interference from a random interfering plane at a time from the 10 available interfering planes.

TrData_Y (N_tr, M x 2, L): the array signal snapshots Y_ss [eq. (22)]. (floats: Real and Imaginary parts extracted)
TrData_FFT (N_tr, M x 2, L): the FFT calculated from Y_ss. (floats: Real and Imaginary parts extracted)
TrData_R (N_tr, M x 2, M): the covariance matrix estimate \hat{R} [eq. (17)] calculated from Y_ss. (floats: Real and Imaginary parts extracted)
TrData_H (N_tr, K + 1, M): channels data, only single interference satellite at a time (K=1). The first row is the desired satellite channel (h_d). (complex)
TrData_X (N_tr, L) : the orignal desired signal samples “to be recovered”. (complex)
TrData_No2 : Noise power “sigma_n^2”. (fixed float value= 6.512325902250013e-14)
TrData_DOAs: consists of two main sub-data are:
* globalDOAx (2, N_tr): DOA of the desired satellite [Az ; -El]. (floats)
* globalDOAi (2, N_tr): DOA of the interfering satellite [Az ; -El]. (floats)

1. Test Data (Similar to Train Data but with N_ts = 900) : 3 desired satellite passes, repeated 3 times with interfernce from a random interfernce plane from the 10 available planes.

TsData_Y (N_ts, M x 2, L).
TsData_FFT (N_ts, M x 2, L).
TsData_R (N_ts, M x 2, M).
TsData_H (N_ts, K + 1, M).
TsData_X (N_ts, L).
TsData_No2: Noise power “sigma_n^2”. (fixed float value= 6.512325902250013e-14)
TsData_DOAs: globalDOAx (2, N_ts), globalDOAi (2, N_ts).

DOWNLOAD DATA