How does the HP WiFi mouse work?

[NiHaoMike]

http://www.amazon.com/HP-WiFi-Mouse/dp/B00556O4YC
Perhaps it can be a cheap source of parts for connecting microcontrollers to WiFi? (And, of course, there's a nice optical sensor assembly as well.)

[johnnyfp]

From the information and details on the mouse, it looks like it only does peer-to-peer wifi, and a bastardized one at that. I suspect it will have very limiting range and bandwidth to do anything remotely useful and you have to have a Win7 Wifi Certified reciever to use the mouse with.

For slightly less you can get a Microchip wifi module which does proper wifi. Check out http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en548014. Seems promising.

cheers

Jfp

[codeboy2k]

From the information and details on the mouse, it looks like it only does peer-to-peer wifi, and a bastardized one at that. I suspect it will have very limiting range and bandwidth to do anything remotely useful and you have to have a Win7 Wifi Certified reciever to use the mouse with.

For slightly less you can get a Microchip wifi module which does proper wifi. Check out http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en548014. Seems promising.

cheers

Jfp

what are they thinking! WiFi doesn't belong in a mouse, even a bastardized version.  That's even worse, now it's non-standard.  And unless I am mistaken, a wifi adapter in a laptop can do peer to peer or AP mode, but not both. So if my laptop/homenet was already using an AP, then I'd have to switch everything over to peer-to-peer mode instead .. ugh!

[mikeselectricstuff]

I bet the battery life sucks....

[amyk]

It even has an off button. Who would remember to turn off their mouse after use?

Technical information seems to be sparse (haven't seen any mention of Linux drivers either), but my guess is that it relies on the NIC in the host supporting promiscuous mode and just broadcasts packets when it moves. The protocol will probably not be going through the TCP/IP stack.

[T4P]

I bet the battery life sucks....

I hope there isn't a beep everytime the mouse is moved 
Or a "off" button that puts it into a fake "sleep" 

[Short Circuit]

Possible issues like power consumption and protocol compatibility aside, I think its a great idea.
Or better said, I believe it's ridiculous for a device like a laptop needing to have two or three 2.4GHz
radios just because somewhere at firmware level, protocols are not compatible. New chipsets do
wireless display (WiDi) using the the wifi radio, so why not mice and keyboards.

[T4P]

Bad idea remains, how many batteries do you have to freaking swap out? Why are the current options aren't any good?

[G7PSK]

I use a 2.4 GHZ wireless mouse made or marketed by hama, that uses two AAA in parallel. The batteries last about 7 months and I never turn the mouse off , so battery life need not be a concern.

[Short Circuit]

Bad idea remains, how many batteries do you have to freaking swap out? Why are the current options aren't any good?

We have no idea how much current it draws, so why a bad idea? Fact that it has an on/off switch does not say anything, my Logitechs (with proprietary 2.4G dongles) also have on/off switches... And even if it draws too much current, still a good idea, only needing a bit more R&D
Current options suck because they need dongles, which occupy ports (annoying if your daily pc is a laptop with few ports) and point out of the side.

[mikeselectricstuff]

I use a 2.4 GHZ wireless mouse made or marketed by hama, that uses two AAA in parallel. The batteries last about 7 months and I never turn the mouse off , so battery life need not be a concern.

The problem is not with the 2.5GHz, but the use of a very complex protocol like 802.whatever. For a mouse all it needs to do is send very short packets fairly infrequently, but I would expect wifi to require significantly more overhead as it is designed for much higher payload sizes, plus all the security stuff.

[codeboy2k]

..Or better said, I believe it's ridiculous for a device like a laptop needing to have two or three 2.4GHz
radios just because somewhere at firmware level, protocols are not compatible. New chipsets do
wireless display (WiDi) using the the wifi radio, so why not mice and keyboards.

I agree that it's ridiculous for a laptop to have so many radios. Something needs to be done here.

Bluetooth was supposed to provide most of the PAN (Personal Area Network) connectivity, and WiFi the WAN (Wide Area Network) connectivity but I think bluetooth failed miserably to deliver.  Bluetooth mice and keyboards continued to disappoint me and almost everyone I talked to about it.  These days if I go wireless, it's going to be a non-bluetooth mouse and keyboard, and that usually means Logitech for me.  Now, it seems that the manufacturers agree that one radio is easier to manage, so they are starting to put everything on WiFi, but it's just too heavyweight.  Unfortunately, I concede that it's also proven to work, and works well.

I use a 2.4 GHZ wireless mouse made or marketed by hama, that uses two AAA in parallel. The batteries last about 7 months and I never turn the mouse off , so battery life need not be a concern.

The problem is not with the 2.5GHz, but the use of a very complex protocol like 802.whatever. For a mouse all it needs to do is send very short packets fairly infrequently, but I would expect wifi to require significantly more overhead as it is designed for much higher payload sizes, plus all the security stuff.

I am not deeply familiar with the WiFi protocols over the air, but perhaps in the future there will be a light wifi, or something that just just uses the bottom-most communications layer for light traffic like mouse and keyboards and other devices (like bluetooth, but actually works) and perhaps there can even be a service class, that uses less power from the laptop radio to transmit these bottom layer lighter protocols.  Then you only have one radio in the laptop, the protocol is WiFi at the top layer, but just a packet layer at the bottom, and the transmitter can adjust its output power for different classes of peripheral.

Giving your mouse, keyboard, and display an IP address and using TCP/IP, a fully routable protocol, is still pretty ridiculous. Soon they will amend that to include a harddisk, currently the realm of wireless USB.  I know, it's just a 32-bit number, and it's basically local, and probably given from the auto pool 169.254.0.0/16 , but still.... I can't get over it. Perhaps that's because I grew up on the Internet, long before household NAT gateways and firewalls, and when my desktop Sun or Apollo workstation was directly connected to the Internet through our router.

[amyk]

Regular WiFi 802.11 is essentially Ethernet frames on an RF physical layer. The layers above that are indepedent (although usually TCP/IP) and there's a field in the ethernet header to identify the upper-layer protocol. Wireless mice like these probably are using just that, same Ethernet frames as everything else but a different, lighter upper-level protocol.

A bit of digging reveals HP calls the protocol "Link-5" and might be USB related... USB over WiFi?

I think Mike would be interested in getting one to teardown?

[Short Circuit]

http://www.ozmodevices.com/release_16.php

OZMO2000 - Features and Functionality:
 
•No Dongle: The OZMO2000 connects seamlessly with the Wi-Fi chip in the notebook. Unlike proprietary 2.4GHz solutions, no dongle is occupying a USB port.
•Battery Life: The advanced power management features provide a solution battery life that is two to three times longer than legacy wireless solutions like Bluetooth®.
•Low Cost: as a System-on-a-Chip (SoC), the OZMO2000 only requires a few external components which results in a cost-effective solution compared to Bluetooth and proprietary 2.4GHz technology.
•Highly Scalable: the OZMO2000 supports up to 24Mbps data rates with very low power consumption. This provides an ideal solution for low data rate applications like mouse and keyboard, as well as for advanced designs with higher data rate requirements like touch features or integrated video and audio.

Wi-Fi Direct Certification
 The OZMO2000 is compatible with Windows 7 SoftAP as well as Intel's MyWi-Fi Technology. Furthermore, the OZMO2000 is designed to be compatible with the upcoming Wi-Fi Direct standard. The Wi-Fi Alliance plans this new certification program to be available later in the year.

I guess the fairly high price and limited OS support are slowing market (or device manufacturer) adoption a bit, but if these marketing statements are true, then I'd expect this to become the dominant wireless HID in a few years. (These things can take some time, USB took a few years to become adopted as well, and look now...)

[T4P]

Well the current crop of wireless mouses are 2.4GHz BUT they are very different from what we call "WiFi"
Instead of sending packets like the HP wifi mouse does ( which IS a waste of power ), it establishes a "2-wire" TX-RX simple serial connection data connection
It only sends what it needs to send, wifi works a little differently, every once in a while it has to send a few keep-alive packets or else the connection is dropped

Well you know that laptops only have 1 wifi RECIEVER and they can only recieve 1 station and if HP decides to use more then 1 radio on their laptops that's their issue, other laptops only have 1 and if the mouse doesn't come with a reciever ... tough luck, not gonna sell

Well on the other hand i do know that wifi recievers can be hacked into transmitters (IF you use a android since 2.2 you probably know already)
but that still remains, you can transmit but can't send once you set it to send until you set it to send, it's like a man, it can only do 1 thing at a time

But history taught us, no matter what sort WiFi, it gobbles power like a monster

[Wartex]

I bet the battery life sucks....

5-6 months per review

[rr100]

Bluetooth was supposed to provide most of the PAN (Personal Area Network) connectivity, and WiFi the WAN (Wide Area Network) connectivity but I think bluetooth failed miserably to deliver.  Bluetooth mice and keyboards continued to disappoint me and almost everyone I talked to about it.  These days if I go wireless, it's going to be a non-bluetooth mouse and keyboard, and that usually means Logitech for me.  Now, it seems that the manufacturers agree that one radio is easier to manage, so they are starting to put everything on WiFi, but it's just too heavyweight.  Unfortunately, I concede that it's also proven to work, and works well.

Well, there isn't much difference either in the reliability or the complexity of setup in bluetooth versus wifi. There's just much higher incentive for the end user to get wifi working as usually no wifi means no internet but most of the other PAN bluetooth devices (like bluetooth headphones or mice) can be replaced either by their wired counterparts (which the user still has and don't have the problem of requiring batteries) or by wireless devices with their own dongles.
And frankly for mice there isn't much to it, logitech has some using their proprietary (and tiny) dongles (that can even work with multiple devices so you don't need one for keyboard and one for mouse). Battery life is supposed to be many months (and even some years) which won't be matched by any heavier protocol.

I do have a regular bluetooth mouse (since quite a few years) and it works well for trips (I really didn't want any extra dongle). Never run out of battery, even after months of use but I do turn it off (and it has two AA batteries even if it's a small mouse). You do have to pair it of course, but at least bluetooth mice are supported without extra drivers in any modern OS (including linux of course).

One important point here: with any non-dongle-based device you still need to fiddle with the OS to recognize it, however you want to call the procedure (bluetooth pairing, etc).

[rr100]

Since my last post (above) I left my bluetooth mouse on/standby (as in no turning off after use). Today the LED started blinking orange (by the way they have rather nice programming, LED blinks a little while after you stop moving the mouse, probably the best way to get your attention; also rather spooky as it blinks just after take your hand away).
Frankly I don't know with so high standby drain - you still need to click to pull it from standby so they could have nearly zero standby current.

[T4P]

Since my last post (above) I left my bluetooth mouse on/standby (as in no turning off after use). Today the LED started blinking orange (by the way they have rather nice programming, LED blinks a little while after you stop moving the mouse, probably the best way to get your attention; also rather spooky as it blinks just after take your hand away).
Frankly I don't know with so high standby drain - you still need to click to pull it from standby so they could have nearly zero standby current.

What do you think? It's bluetooth! I don't think you want to have to go though the connecting process everytime you wake it up from standby! And don't say whatever connecting process doesn't need to exist... IT'S WHY THE BT protocol is a terrible one

[Simon]

It even has an off button. Who would remember to turn off their mouse after use?

Me, I had a cheap chinese wireless mouse with a switch. Great for my laptop. If you stick a battery mouse in a laptop bag have fun expecting the batteries to be usable when you come to get it out.

I think it is playing on the windows 7 ability to work as a wireless router, which just adds further complication and makes it incompatible with other versions of windows.

[rr100]

What do you think? It's bluetooth! I don't think you want to have to go though the connecting process everytime you wake it up from standby!

It has to and it does go through the connecting process, that's fine. But when it's in standby is dead, dead, it's not registering movement (other mice do and it doesn't seem to be that much of battery drain!), is not seen by the laptop, etc. Maybe I should wire it up one of these days and see what current uses in standby.

By the way does anybody know if somebody makes some kind of "dummy batteries" with thin wires attached? I've seen this is popular for some cameras (instead of a DC-in socket they have basically a hole for the wires and you can buy a special power supply+dummy battery from the manufacturer). But just the common (AA, AAA) sizes would be very useful to evaluate current consumption in a more comfortable fashion than "get your probes in there while pushing the battery in one direction and the spring in the other".

[NiHaoMike]

Regular WiFi 802.11 is essentially Ethernet frames on an RF physical layer. The layers above that are indepedent (although usually TCP/IP) and there's a field in the ethernet header to identify the upper-layer protocol. Wireless mice like these probably are using just that, same Ethernet frames as everything else but a different, lighter upper-level protocol.

A bit of digging reveals HP calls the protocol "Link-5" and might be USB related... USB over WiFi?

I think Mike would be interested in getting one to teardown?

I did get one to teardown. Fry's had a special ($12) last week. The most interesting part is that I looked up the FCC ID (EMJMMOWFFKUL) and it seems to be able to operate on both the 2.4GHz and 5GHz bands! It's definitely very unexpected for a low power and low cost device.

[amyk]

Please do post a teardown if you can.

[rr100]

Found it on sale at Fry's for 12.99 and got 3 for good measure. What a POS. Got it to peer with the laptop but it won't move the mouse on the screen. But to top it off it makes noises when you left click!!! First time I thought it has some piezo speaker inside, it sounds precisely like a cheap  meater's continuity tester!!!
Actually I wanted it to replace my bluetooth mouse that had some lag issues when the laptop was in the docking station/external monitors/keyboard, so just a bit further away (normally on a laptop you type and the mouse is just beside it); I was thinking wifi might reach without issues but it seems it doesn't really work at all. It might be a personal firewall issue on the laptop but I'm not going to mess up with it just to make the mouse working, already I wasted too much time I'm not getting back.
They're going back as soon as I manage to go there again.
In the end maybe the proprietary route is the right one here - at least with the good Logitech ones they just work, no lag, you forget to change the batteries, etc. The only problem is they take one USB port.

[T4P]

That's the issue i have with laptops ... They have so many more free un-hubbed ports and they won't route them out on the low-end ones 
Therefore 1 keyboard + mouse and you're doomed on a laptop