Interference Between 700 MHz Cellphone Signals and DTV Signals
April 11, 2012
I presented two papers at the recent IEEE Conference on Consumer Electronics on interference to broadband wireless communications by other broadband signals. GPS was one of my topics, and the other was interference to cellular telephony in the 700 MHz band (former Channels 52-69) from DTV transmitters on Channels 14 to and including Channel 51. In this column I want to talk about interference to DTV reception in the UHF band by cellphones and their base stations as more and more of these are getting on-the-air in the 700 MHz band.
Suppose I am watching DTV and someone in the house makes a call on their cellphone (one of the new 700 MHz units, that is). If my antenna is on my rooftop, everything is probably OK, but if my antenna is indoors, it may pick up enough signal power from a nearby cellphone to block DTV reception. Or as more and more base stations transmitting on former TV channels go "on-air," even my roof top antenna may deliver enough of that signal to cause blocking of my DTV receiver.
But how can these cellular phone signals get into my DTV receiver?
This is sometimes referred to as "blocking." The undesired signal may be strong enough to overload the tuner of your DTV tuner.
BLOCKING: YESTERDAY AND TODAY
I first encountered blocking in San Francisco well before color TV, and even before UHF broadcasting. A customer complained that at 8:30 p.m., his brand new RCA set quit. I investigated and sure enough, at 8:30 it quit receiving TV signals. I put my head close to the loudspeaker and heard very faintly "Hotel Mark Hopkins, San Francisco, California, the United States of America." A call to the hotel disclosed that there was indeed a radio station there and I was connected with the engineer. I explained where I was and he replied that at 8:30 he turned on his 50,000 Watt short wave transmitter six blocks from my customer's house. That TV set had unshielded IF amplifier inductors and its IF was 21.25-25.75 MHz; this SW transmitter was on 11.8 MHz. Its second harmonic overloaded the receiver so its AGC circuit turned the RF and IF gains down, way down. That is blocking, also known as "de-sensitization."
Fast forward to today, modern broadband receivers usually have very little RF selectivity so a strong signal can pass through the RF amplifier to the RF Automatic Gain Control circuit where it is rectified and this DC voltage cuts off the RF amplifier. This too is blocking and can be done with less than 50 Watts near a 700 MHz band base station. Or this strong undesired signal could be downconverted to frequencies near the IF where the first IF amplifier rectifies this signal, lowering the gain of the IF amplifier and this too is blocking. It is sometimes also called "Brute Force Overloading."
Fig. 1: Two DTV signals on Channels N+3 and N+6 may generate third order distortion products seen here as "Bee-Hives" centered on channels N and N+9. The noise in Channel N is 6 dB greater than in Channel N+/-1, same for noise in Channels 8N+8 and N+10 relative to noise in Channel N+9.
Now, let's turn the tables around. Consider two strong DTV signals on Channels 47 and 50: They generate third order intermodulation products extending from Channels 43 to (former Ch. 54 Blocks A, B and C). This is shown in Fig. 1. Two DTV signals are on Channels N+3 and N+6. These generate third order intermodulation products which look like "Bee-Hives" in Fig. 1. Each Bee-Hive is three channels wide because this is third order distortion. One is centered on Channel N, the other on Channel N+9. A receiver tuned to either Channel N or N+9 may be unable to work because of the high noise level in these two channels (actually third order IM). If the receiver was tuned to Channel N+2, N+4, N+8 or N+10, it would "see" 6 dB less noise and would have a much better chance of working. If the receiver were tuned to channel N-2 or N+11 where there is no noise, reception would be OK.
Note that I didn't say DTV receiver. A handheld cellphone may also pick up these two DTV signals. If they overload a cellphone receiver's front-end a noise power spectrum extending from channel 43-Block C inclusive will be generated which may block reception. If the cellphone is tuned to a former DTV channel below 43 it will work. If the cellphone is tuned to any former channel up to, and including Block C, the noise drives its AGC circuit to turn the gain down and it can't hear what its base station is sending, in other words blocking. All the cellphone user knows is that his cell phone is not reliable.
WHAT IS THE REMEDY?
This is what I told my audience: these 700 MHz band cellphone receivers need to have improved RF selectivity and mixers with much greater dynamic range so they aren't overloaded by pairs of strong DTV signals. The way to extend the linear dynamic range of receivers is to increase the current drawn from the battery by the RF amplifier. But that will reduce play time between battery charges—darn! Now a mixer with much improved dynamic range would require much higher power injected by the local oscillator, but where does this power come from? Why from the rechargeable lithium ion battery, and that is why these radios don't have wide dynamic range mixers—consumers demand long playing time between charges.
So what about better RF selectivity? Well that would require high Q-tuned circuits between the antenna and the RF amplifier and between the RF amplifier and the mixer. These tuned circuits use inductors. Inductors of really high Q factor are needed in the UHF band and while they are not costly, they take up valuable space inside the slim cellphone. The public probably would shun larger cell phones, so what is the answer? And that is where my speaking time expired—I don't have an answer.
Perhaps this problem will affect future cellphone network planning. Perhaps those UHF channels so coveted by cellphone engineers aren't really so great as cell mates with megawatt UHF transmitters.
Charles Rhodes is a consultant in the field of television broadcast technologies and planning. He can be reached via e-mail at firstname.lastname@example.org.