Deep Learning for TDD and FDD Massive MIMO: Mapping Channels in Space and Frequency

Abstract

Can we map the channels at one set of antennas and one frequency band to the channels at another set of antennaspossibly at a different location and a different frequency band? If this channel-to-channel mapping is possible, we can expect dramatic gains for massive MIMO systems. For example, in FDD massive MIMO, the uplink channels can be mapped to the downlink channels or the downlink channels at one subset of antennas can be mapped to the downlink channels at all the other antennas. This can significantly reduce (or even eliminate) the downlink training/feedback overhead. In the context of cellfree/distributed massive MIMO systems, this channel mapping can be leveraged to reduce the fronthaul signaling overhead as only the channels at a subset of the distributed terminals need to be fed to the central unit which can map them to the channels at all the other terminals. This mapping can also find interesting applications in mmWave beam prediction, MIMO radar, and massive MIMO based positioning.In this paper, we introduce the new concept of channel mapping in space and frequency, where the channels at one set of antennas and one frequency band are mapped to the channels at another set of antennas and frequency band. First, we prove that this channel-to-channel mapping function exists under the condition that the mapping from the candidate user positions to the channels at the first set of antennas is bijective; a condition that can be achieved with high probability in several practical MIMO communication scenarios. Then, we note that the channel-to-channel mapping function, even if it exists, is typically unknown and very hard to characterize analytically as it heavily depends on the various elements of the surrounding environment. With this motivation, we propose to leverage the powerful learning capabilities of deep neural networks to learn (approximate) this complex channel mapping function. For a case study of distributed/cell-free massive MIMO system with 64 antennas, the results show that acquiring the channels at only 4-8 antennas can be efficiently mapped to the channels at all the 64 distributed antennas, even if the 64 antennas are at a different frequency band. Further, the 3D ray-tracing based simulations show that the achievable rates with the predicted channels achieve near-optimal data rates when compared to the upper bound with perfect channel knowledge. This highlight a novel solution for reducing the training and feedback overhead in mmWave and massive MIMO systems thanks to the powerful learning capabilities of deep neural networks.

Downlink Extrapolation for FDD Multiple Antenna Systems Through Neural Network Using Extracted Uplink Path Gains

“…The channel samples obtained from QuaDRiGa does not have any idealistic feature, e.g., perfect directional reciprocity between the UL and DL channels. This is quite different from previous works where channel samples are obtained based on analytical channel models [15], [16], [18]- [22], [25], [26].…”

“…The purpose of using the NN is to estimate the DL CSI directly from the UL CSI measured at the BS without having any explicit downlink training. More precisely, the previous works in [15], [16] used the UL OFDM channel H(f ul ) as an input and the DL OFDM channel H(f dl ) as an output to train the NN, i.e.,…”

“…To convert complex-valued data into real-valued ones, the most commonly used method is to divide the complex-valued data into their real and imaginary parts. The real and imaginary parts may form one input layer [14], [15], [21], [22], [25], [28], or each of parts can construct two independent input layers [19], where each input layer is used for a separate network. Not only the real and imaginary parts of data but also the magnitude of data can be added to the input layer as in [20], since more information, even redundant, in the input layer restricts the network from over-fitting.…”

“…The optimal NN structure greatly depends on the data, so the CH-learning and the PG-learning may require different structures. The CH-learning based on CNN and MLP is studied in [14] and [15], respectively. After extensive simulations, we have verified that MLP is better than CNN for the CH-learning while CNN outperforms MLP for the PG-learning.…”

Deep Learning-Based Downlink Channel Prediction for FDD Massive MIMO System

“…Based on the CSI correlations between different base station (BS) antennas, linear and supper vector based regression methods have been proposed to use the downlink CSI of partial BS antennas to predict the whole downlink CSI, which can reduce the overheads of both downlink pilots and uplink feedback [26]. In fact, [27] develops the channel mapping in space and frequent concept which proves that deep neural networks (DNNs) can learn not only the correlation between closely-located BS antennas but also between the base station arrays that are positioned at different locations or different frequency bands in the same environment. By exploiting the mapping relation between the uplink and downlink CSI in FDD massive MIMO systems, an efficient complex-valued based DNN is trained to predict the downlink CSI only from the uplink CSI, i.e., no downlink pilot is needed at all [28].…”