The superposition of waves caused by multipath propagation was for a long time considered an unavoidable nuisance in radio communication links. The discovery that multipath interference was central to enable much larger data rates was a breakthrough at the turn of the century. Multiple-input multiple-output (MIMO) spatial multiplexing (SM) allows unprecedented efficiency in the use of the radio spectrum, however, this comes at the cost of high complexity at the receiver because the underlying symbol detection problem belongs to the class of problems of highest computational complexity. In general, lattice problems are simple to describe but rather hard to solve optimally; finding algorithms to deal with the problem has been a central topic in the last decade of research in MIMO SM. This paper contributes to a deeper understanding of the most important types of receivers for SM with a unifying lattice perspective. Capitalising on that, two novel receivers are proposed. The geometric relation between the primal and the dual lattice is clarified, leading to the proposal of a pre-processing technique that greatly reduces the number of candidate solutions via geometric considerations. Then, looking at lattices from a group theory perspective, it is shown that it is possible to approximate the typical lattices encountered in MIMO by a lattice having a trellis representation, translating the problem for the first time into one manageable by the Viterbi algorithm, well known to the semiconductor industry.