A large-scale mimo detection method, device and storage medium
A detection method and a detection device technology, which are applied in transmission monitoring, advanced technology, diversity/multi-antenna systems, etc., can solve problems such as performance loss of MIMO detection methods, and achieve the effects of improving convergence performance, low complexity, and improving performance
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0092] like figure 1 As shown, a massive MIMO detection method includes the steps:
[0093] Step 101, constructing an approximate expectation propagation network based on the deep learning network, introducing learnable linear correction parameters, and training to obtain a trained approximate expectation propagation network model;
[0094] Step 102, obtain the received signal vector y, the channel matrix H and the noise variance according to the acquired received signal, channel information and noise information
[0095] Step 103, receive the signal vector y, the channel matrix H and the noise variance The data is fed into a trained approximate desired propagation network model, resulting in an estimate of the transmitted signal.
[0096] Preferably, the approximate expectation propagation network model training method includes:
[0097] 1) Based on the construction of the deep learning network, an approximate expected propagation network is obtained, which is recorded ...
Embodiment 2
[0135] like image 3 As shown, a massive MIMO detection device includes:
[0136] The training module is used to train the approximate expectation propagation network model, build the approximate expectation propagation network based on the deep learning network, and each layer of the approximate expectation propagation network corresponds to each iterative process of the EPA algorithm; in each network layer, a learnable Linear correction parameters to correct the coefficients of the second-order term of the unnormalized cavity edge distribution at each iteration in the EPA algorithm; approximating the final estimate of the expected output transmit signal of the last layer of the propagation network; building said approximate expectation for The propagation network is trained to obtain the learnable linear correction parameters after training, and the learned approximate expected propagation network model is obtained by fixing the learnable linear correction parameters;
[01...
Embodiment 3
[0164] A computer-readable storage medium storing computer-executable instructions for executing the massive MIMO detection method in any one of Embodiment 1.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More