On Fri, 04 Jan 2008 13:18:25 -0500, Jerry Avins <jya@ieee.org> wrote:>Eric Jacobsen wrote: >> On Thu, 03 Jan 2008 22:08:56 -0600, John O'Flaherty >> <quiasmox@yeeha.com> wrote: >> >>> On Thu, 03 Jan 2008 21:17:00 -0500, Randy Yates <yates@ieee.org> >>> wrote: >>> >>>> John O'Flaherty <quiasmox@yeeha.com> writes: > > ... > >>>>> I'm thinking of buying a tower and putting a couple of antennas on >>>>> it, maybe with a rotor. >>>> How about a UHF horn? :) >>> Then I might need a precision rotor... >> >> And you might get an even less reliable signal. >> >> Multipath helps if your equalizer is good enough to take advantage of >> it. >> >> Think of it this way, if 80% of the impinging power at the antenna >> location is in the strongest ray and 10% of the power is in the second >> strongest, an omnidirectional or wide-beam antenna that admits both, >> along with an equalizer that integrates them, will be able to exploit >> at -least- 90% of the available energy (plus however many other >> smaller rays can be integrated). >> >> If the high-gain antenna rejects all of the reflections but the >> strongest ray, i.e., you've done a good job of pointing the horn at >> the strongest reflection, if no other reflections come from that >> direction you only have 80% of the available energy to use. This >> might reduce the SNR enough to put you closer to or below the >> threshold of watchability. > >The usual goal isn't the highest signal, but the highest SNR. A very >directional antenna can reject noise as well as reflections. As the New >York lottery ads go, "Hey, you never know!" > >JerryNoise comes primarily from the LNA and the rest of the AFE, which is all after the antenna. Interference rejection gets better with a narrower antenna beam. Noise, not so much. Eric Jacobsen Minister of Algorithms Abineau Communications http://www.ericjacobsen.org
Online · Live + on demand
February 9-11, 2027
Price goes up Jan 1: register now to lock in today's rate
Register Now →
OT Digital TV
Started by ●January 3, 2008
Reply by ●January 4, 20082008-01-04
Reply by ●January 4, 20082008-01-04
On Fri, 04 Jan 2008 18:00:22 +0800, Steve Underwood <steveu@dis.org> wrote:>Eric Jacobsen wrote: >>> Then I might need a precision rotor... >> >> And you might get an even less reliable signal. >> >> Multipath helps if your equalizer is good enough to take advantage of >> it. >> >> Think of it this way, if 80% of the impinging power at the antenna >> location is in the strongest ray and 10% of the power is in the second >> strongest, an omnidirectional or wide-beam antenna that admits both, >> along with an equalizer that integrates them, will be able to exploit >> at -least- 90% of the available energy (plus however many other >> smaller rays can be integrated). >> >> If the high-gain antenna rejects all of the reflections but the >> strongest ray, i.e., you've done a good job of pointing the horn at >> the strongest reflection, if no other reflections come from that >> direction you only have 80% of the available energy to use. This >> might reduce the SNR enough to put you closer to or below the >> threshold of watchability. >> >> This illustrates the concept of multi-path gain, which is a good thing >> as long as the equalizer can make use of the collected rays. It >> could be that the equalizer in your receiver just isn't very good in >> dynamic conditions, like the wind blowing the reflectors and antenna >> around. > >For many people the wind blowing trees around is a much more significant >thing. Equalisers tend to perform poorly with such rapidly changing >reflections.That's completely implementation dependent. Channel dynamics for a fixed Tx and Rx are typically very slow compared to the ATSC (or DVB-T) symbol rate, so a good equalizer that can keep up with the channel shouldn't be all that tough. If one or both are mobile it gets a lot harder, but that's not usually the case, especially for a display of any substantial size. Eric Jacobsen Minister of Algorithms Abineau Communications http://www.ericjacobsen.org







