The "receiver noise bandwidth" is the figure that you need - the equivalent brick-wall bandpass bandwidth of the
receiver's response. You'll get noise from the input stage(s) of the receiver and from the air (interference from artificial
and natural sources, and ultimately the 2.7K background radiation from space). Also resistive losses in any cable between receiver
and antenna will introduce noise too (more at higher RF frequencies). Only the resistive elements of the losses will
be at 300 kelvin. Interference may dominate if you are in a city, for instance.
Point a good antenna straight up and you'll often get a noise temperature well below 300K, but also interference from
satellites, the sun, Jupiter (again RF frequency dependent).
Given the antenna bandwidth is so wide it implies you are well into the UHF bands, I think this means less interference
in general, so thermal noise will dominate.
And you should use natural logs, not base 10 for that equation.... k T ln(bw)