FIG. 3 is a timeline of a MU-MIMO downlink in accordance with an embodiment of the invention incorporating hybrid spatial mapping using both explicit sounding together with crosstalk tracking. The timeline shows a wireless access point (WAP) 300 and associated stations 302A-B of a residential WLAN during various phases of operation associated with an MU-MIMO downlink in an embodiment of the invention where both the WAP and the stations include complementary capabilities to enable hybrid MU-MIMO spatial mapping. These capabilities reduce the frequency of explicit soundings by extending the post sounding interval over which a reliable downlink can be maintained. Each packet in the MU-MIMO downlink has a payload portion which contains distinct blocks of data each destined for a corresponding one of the stations in the MU-MIMO downlink group. The MU-MIMO packet allows transmission of discrete payload data to two or more receiving stations at the same time, thus increasing WLAN throughput. Ideally, each station in the group receives a common MU-MIMO packet header, and its own a distinct block of data. The distinct blocks of data in the MU-MIMO packet payload are said to be spatially mapped during transmission so as to arrive at each corresponding one of the stations, without interference, a.k.a. crosstalk, from the distinct blocks of data destined for other stations in the group. The spatial mapping involves complex signal processing using the multiple antennas of the WAP and a precoding matrix to spatially separate the distinct blocks of data in the MU-MIMO packet payload among the targeted MU-MIMO group of stations. The determination of the precode matrix, requires detailed channel information feedback from each of the targeted stations in response to an explicit sounding. The feedback from each station as to its communication channel with the WAP is aggregated by the WAP and used to determine the precode matrix for transmitting subsequent MU-MIMO downlink packets. The minute changes in the communication channels which would otherwise blur the spatial separation shortly after an explicit sounding are corrected for by using crosstalk feedback from one or more of the targeted stations during the MU-MIMO downlink. This avoids frequent explicit soundings and extends the downlink interval between explicit soundings, e.g. by 500%.