A Simple Recursively Computable Lower Bound on the Noncoherent Capacity of Highly Underspread Fading Channels
Accepted version
Peer-reviewed
Repository URI
Repository DOI
Change log
Authors
Abstract
Real-world wireless communication channels are typically highly underspread: their coherence time is much greater than their delay spread. In such situations it is common to assume that, with sufficiently high bandwidth, the capacity without Channel State Information (CSI) at the receiver (termed the noncoherent channel capacity) is approximately equal to the capacity with perfect CSI at the receiver (termed the coherent channel capacity). In this paper, we propose a lower bound on the noncoherent capacity of highly underspread fading channels, which assumes only that the delay spread and coherence time are known. Furthermore our lower bound can be calculated recursively, with each increment corresponding to a step increase in bandwidth. These properties, we contend, make our lower bound an excellent candidate as a simple method to verify that the noncoherent capacity is indeed approximately equal to the coherent capacity for typical wireless communication applications. We precede the derivation of the aforementioned lower bound on the information capacity with a rigorous justification of the mathematical representation of the channel. Furthermore, we also provide a numerical example for an actual wireless communication channel and demonstrate that our lower bound does indeed approximately equal the coherent channel capacity.
Description
Keywords
Journal Title
Conference Name
Journal ISSN
1558-2248