A low complexity OFDM receiver with combined GAMP and MF message passing

Abstract

With a unified belief propagation (BP) and mean field (MF) framework, we propose an iterative message passing receiver, which performs joint channel state and noise precision (the reciprocal of noise variance) estimation and decoding for OFDM systems. The recently developed generalized approximate message passing (GAMP) is incorporated to the BP-MF framework, where MF is used to handle observation factor nodes with unknown noise precision and GAMP is used for channel estimation in the time-frequency domain. Compared to state-of-the-art algorithms in the literature, the proposed algorithm either delivers similar performance with much lower complexity, or delivers much better performance with similar complexity. In addition, the proposed algorithm exhibits fastest convergence.

Joint spare channel estimation and decoding for orthogonal frequency division multiplexing using combined message passing

“…In this work, we also investigate the message passing schedules as they may significantly affect the performance of the JCED receivers [11, 27, 28]. The receivers in [4–8, 23, 24] use a simple message schedule, where each message is updated once in each iteration. The other receivers in [27–30] use both inner iteration and global iteration , where the receivers are functionally divided into several blocks, such as channel estimation and decoding, and the messages are iteratively updated in each block, called inner iteration , and then the blocks iteratively exchange messages, called global iteration .…”

“…Orthogonal frequency division multiplexing (OFDM) has achieved great success in wireless communications due to its high spectral efficiency and resilience in combating multi-path distortion [1][2][3]. As receivers with joint channel estimation and decoding (JCED) can achieve much better performance than conventional receivers where channel estimation and decoding are separated, the design of the JCED OFDM receivers has attracted much attention in the literature [4][5][6][7][8][9], which is also the focus of this work. It is well known that broadband wireless channels approximately exhibit sparsity in the lag-domain, and the sparsity can be exploited to significantly improve the performance of the system [10].…”

“…It is well known that broadband wireless channels approximately exhibit sparsity in the lag-domain, and the sparsity can be exploited to significantly improve the performance of the system [10]. Hence, in this work, we will also take the channel sparsity into account, and aim to design low complexity JCED receivers based on the message passing techniques [4][5][6][7][8][9].…”

“…It is worth mentioning that, by exploiting their virtues while avoiding their drawbacks, BP, EP, and MF have been integrated carefully to design superior receivers with low complexity [20][21][22]. With regard to OFDM receiver design, there has been various message passing based approaches to JCED in the literature [4][5][6][7][8]. One of the key issues in the design of messagepassing JCED receivers lies in how to deal with the observation nodes [4,5,8].…”

“…With regard to OFDM receiver design, there has been various message passing based approaches to JCED in the literature [4][5][6][7][8]. One of the key issues in the design of messagepassing JCED receivers lies in how to deal with the observation nodes [4,5,8]. According to the approaches employed for the observation nodes (local functions), the receivers can be divided into two types: the MF-based and the BP-based.…”