After a long delay, I'm back and blogging. I spent most of the past month doing or recovering from (in terms of work in other classes) my Readiness Assessment, my department's loose equivalent of qualifying exams -- and am still catching up in other classes. In the meantime, we've talked about frequency shifting, open fit hearing aids, and other topics that have gone more blurrily by my exam-focused brain; if folks have particular curiosity or questions about an area, holla and I'll see what I can pull off. Today's topic is wireless connectivity.

Hearing aids can receive signals through their microphone, but they can also get analog or digital wireless signals from inputs that are in different places -- whether that's a T-coil (which gives you signals from phones), an FM unit (a remote microphone you individually place elsewhere in the room to capture a signal from the environment -- for instance, clipping it to your teacher's lapel) or a music player (signal from outside the immediate environment, perhaps transmitted via bluetooth to a reciever that then transmits to your hearing aids; bluetooth connectivity is too space and resource intensive to build directly into the hearing aid itself, but more on that later).

Wireless has many benefits, including a better signal-to-noise ratio (SNR), similar to the benefits of directional mics, because a remote microphone can be placed closer to the signal source than your ear. Of course, it also lets you connect to your gadgets, which may have signal sources like the music you're playing or the phone call you're on. In an age where most people who can afford hearing aids also have things like TVs and cell phones and whatnot -- and go to the movies, and to plays, and attend lectures and concerts -- the ability to connect to many of the same gadgets that hearing folks use gives us a better SNR for those lovely things instead of relying on our hearing aids to pick up the movie dialogue over the background noise of the kids eating popcorn behind us. For active geeks like me, gadgets of all sorts are a big part of our lives -- and for older folks who make up a large portion of hearing aid users (and are more likely to have reduced mobility) the phone might be their connection to the world outside their house, and the TV the primary entertainment activity inside it.

There are subtler benefits more easily forgotten. Wireless allows you to take a signal that would normally be just at one ear (like a phone call) and stream it equally to both ears. You can also program hearing aids without plugging in a lot of awkward wires (handy for audiologists) and use them to control gadgets at a distance (handy for mobility-limited people and also just plain ol' lazy ones like me). And if you wear two hearing aids, they can "talk" to each other with wireless, so you can (for example) turn up the volume on one hearing aid and have the other automatically step up amplification in sync -- a benefit that sounds trivial until you've experienced the annoyance of trying to turn two tiny dials behind your ear exactly the same amount in the freezing winter with gloves on. As a little kid, I'd always just leave my volume settings on maximum for both hearing aids; hearing things more loudly in one ear than the other was annoying and I knew that if I turned the dials as high as they would go, each ear would be equally amplified. Audiologists in the audience may wince now.

Many modern hearing aids have wireless technologies, but not all users know about them -- and even if they know about them, they don't often use them. A 2010 survey found that only 44% of US users whose hearing aids had wireless connectivity to their TVs knew about it, and only 12% used them -- that's 27% of the people who knew! And with the TV watching statistics about America, I'm pretty sure it's not because the other 83% who knew and didn't use it were not watching TV. Sometimes it's a convenience factor; even if it only takes 10 seconds to set up wireless streaming, that's 10 seconds you need to spend before every phone call, for example, and it adds up. Sometimes it's a "this isn't worth it" factor; if I'm walking by the living room and checking out the TV just for a few minutes before getting back to work, I'm not going to take one of those three minutes to set up my hearing aids. I'm just going to deal with not getting the dialogue for a while (unless someone has been so kind as to turn on closed captioning). It's a fact of life.

We've had wireless for ages -- the familiar FM system, the bane of my existence in elementary school (clunky blue battery/receiver pack, etc. etc.) is an analog one. Digital wireless is a lot more recent. And it's in the area of wireless transmission that the fidelity advantages of digital start coming through. Although digital signals can never be as high-fidelity as their analog counterparts (because there's necessarily a loss of information in the quantization process), the fidelity of digital signals tends to remain more constant -- since we're transmitting the information as 0's and 1's, we can do things like error checking and correction.

In addition, it's a lot harder to confuse a 0 for a 1 or vice versa in a digital signal. If we transmit the signal via a technique like frequency shift keying (FSK), which transmits 1's as rising voltage slopes and 0's as falling voltage slopes, it takes a lot to mix one up with the other -- even if you receive a value that's slightly off, chances are that it'll look almost like the 0 or the 1 you know it's supposed to be. But if an analog signal gets nudged slightly, there's usually no way to figure out what the original value is, since it could be anything.

Digital wireless technologies can get pretty nifty. For instance, some Widex hearing aids have partner monitoring -- if you've got two hearing aids paired up, and one of them has a battery die, or gets taken off, or otherwise gets out of communication, the other one will signal this so the user knows that hey, maybe I'm about to leave one of my hearing aids on the table. There are price differences as well; replacing analog FM units with digital wireless companion mics (which can be given to conversational partners to clip on their lapel) gives you the benefits of improving SNR without the susceptibility to electromagnetic interference -- and a single companion mic cost less than one of the two FM boots you'd have to get to enable FM functionality in your hearing aids. And remember T-coils? I'd rather not; those things are finicky and usually involve magic voodoo positioning of the phone receiver relative to your hearing aid to get a good signal. If your neck starts to cramp, or you don't like holding your elbow over your head, or if you accidentally shrug, the signal quality shifts. Digital phone pairing lets me stroll around talking with my phone in my pocket.

Wirelessly-linked hearing aids can also communicate between each other to figure out feedback better -- if "feedback" is coming from both ears, it's probably an actual environmental noise, but if it's only in one ear, it's probably feedback and should be gotten rid of post-haste. In general hearing aids are able to make "smarter" decisions with more data -- and giving them data from two ears instead of one doubles the data available. I mean, think about it -- what can you do with 4 microphones (2 ears, front and back) instead of 2? 4-microphone beamforming! (Okay, maybe it's because I'm an electrical engineer, but I think that's cool.) Wirelessly-linked hearing aids can also synchronize things like compression -- which sounds like another "Mel, you're an engineer" thing you remember that we localize sounds in space based on the tiny, tiny timing differences with which they hit our ears, and that hearing aids (by modifying the timing with which these signals hit our ears) really mess with that ability. But if the signal processing gets synced up wirelessly, we can get some of it back.

For my mother, who sometimes worries that everything is known to the state of California to cause cancer (oh no, won't wirelessly-paired hearing aids mean that the signal is passing through YOUR BRAIN?!?!), I offer this reassurance: wireless hearing aids won't give me cancer any more than the stuff I'm already doing. I get more electromagnetic exposure standing in front of a dishwasher or a halogen lamp than I do with hearing aids wirelessly talking with each other. Since they're so close to each other and are exchanging (typically) very limited amounts of information ("she pressed the volume-up button!" "she pressed it again!") the power of the signals is super-low, and they're also super-infrequent.

Now into more geeky stuff: codecs! We can have fun with digital signals -- we can compress them so they take up less bandwidth and space (but be careful of the quality when you do it), we can encode them in a redundant way so that we can do error detection and correction... (but be aware that this uses more bandwidth -- and thus more power -- and thus more battery...)

Because, of course, there are tradeoffs with wireless -- as there are tradeoffs with everything. High radio frequencies (for transmission) drain more current than low frequencies and are absorbed more by the human body ... but they also need smaller antennae. The more data you want to send and the faster you want to send it, the faster you're going to go through a battery.

Why not use Bluetooth for everything? It's everywhere, and it's cheap (less than $3 per chip, if you're a device designer adding Bluetooth functionality to your device). Well, it's also high-latency. Because it's a generic codec, it's got a lot of stuff in it to make sure it'll work for lots of devices. It also uses resubmission as a way to handle errors -- if it finds a mistake, it usually just asks the device to send the whole data dump again. All these things lead to a delay of 75-125ms, which is far over the visually detectable threshold; visual inputs like lipreading are no longer synced to the audio. Bluetooth also sucks power like a monster -- about 10x the typical rate that a modern hearing aid consumes a battery.

So Bluetooth can be used with hearing aids... via a relay device, which takes on the burden of the associated problems (power draw especially; I need to charge my relay device after a full day of use, whereas my hearing aids chug along on a single battery for two weeks). The relay device takes in Bluetooth and outputs something that a hearing aid can work with -- a NFMI (near-field magnetic induction) signal or a far-field one that uses radio transmitters and receivers at license-free frequencies (900MHz or 2.4 GHz).

Since the "MI" in NFMI stands for "(electro)magnetic induction," the "awkward T-coil positioning juggle" issue is somewhat present here; orientation and positioning affect sound quality. It's security-coded to particular hearing aids -- good, if you don't like eavesdroppers on your phone conversations -- but has a range of about a meter, and signal strength drops off with the cube of distance, meaning that it's good for things like transmission between hearing aids and for body-worn relay devices.

Any farther and we need to go far-field, which gives us a range of 5-7 meters, with signal strength dropping off by the square (rather than the cube) of distance. Far-field signals often use a proprietary codec (wheee!) made by the hearing aid manufacturer. They also use frequency hopping to avoid interference a technique invented by Hedy Lamarr (better known for her movie starlet roles, but also one heck of a mathematician).

(Ironically, I'm writing this post sitting in class without my wireless mic -- or my hearing aids, for that matter. My right hearing aid started eating batteries at the rate of one per day, so it's in for repairs, and getting the sound streaming into just one ear was just too weird for me today. That means I'm powering through without amplification, just like I did for most of my life. And geez, the world is quieter, and information's missing -- I don't hear it missing, of course, but I know it's there, invisible.)