To-date, most multipath propagation parameters are estimated in the element-space domain due to implementation simplicity. Although estimation in the beam-space domain can reduce both the dimensionality of the data size and numerical complexity, it is only beneficial if applied to arrays with a large number of uniformly spaced elements. A new implementation of the space-alternating generalised expectation-maximisation (SAGE) algorithm is presented in joint element-space and beam-space, namely the `hybrid-space' (HS), such that the underlying features of these dual domains can be simultaneously exploited. The resultant algorithm, namely HS-SAGE algorithm, retains the advantages of the individual domains as the user can specify which parameter should be evaluated in each domain, thus optimising the overall performance. Here, the efficient execution flow of the HS-SAGE algorithm is presented, where the estimation quality is demonstrated through results obtained from both synthetic and real-measurement environments employing circular arrays. It is shown that the proposed algorithm can achieve significant timesaving when compared with the classical implementation approach, whilst maintaining accuracy.
Publisher: Institution of Engineering and Technology (IET)
Rose publication type: Journal article
Sponsorship: This work was supported by the Virtual Centre of Excellence in Mobile and Personal Communications (Mobile VCE, www.mobilevce.com). The authors gratefully
acknowledge the funding and support received under the Core 2 research programme.