Figure 6: Theoretical coverage maps for different distances between the plugged-in device and theWi-Fi router.
The black dots denote the positions for these devices. The red region represents points in the 2D space where a passive
Wi-Fi transmitter can be located, while ensuring that the signal from it to the Wi-Fi router is at least -85 dBm.
2.4.1 Understanding Deployment Scenarios
1. I want to deploy passive Wi-Fi devices in my home.
Where do I place the plugged-in device so as to maximize
their range? Fig. 5 shows the theoretical signal
strength at theWi-Fi receiver as a function of its distance
from the passive Wi-Fi transmitter. We show the results
for different distances between the passive Wi-Fi transmitter
and the plugged-in device. We set Gt , Gr, Gpassive,
Pt to 6 dBi, 0 dBi, 2 dBi, and 30 dBm respectively. The
plot shows that, in general, as the distance between the
passive Wi-Fi transmitter and Wi-Fi receiver increases,
the received signal strength reduces. More importantly,
as the distance between the passive Wi-Fi transmitter
and plugged-in device decreases, the coverage range increases.
This is because, from our analysis, the signal
strength can be increased either by reducing the distance
between the passive Wi-Fi transmitter and the pluggedin
device or that between the passive Wi-Fi transmitter
and the Wi-Fi receiver. Since our goal is to maximize
range, we should reduce the distance between the passive
Wi-Fi transmitter and the plugged-in device. In the
presence of multiple passive Wi-Fi devices, this would
translate to minimizing the worst-case distance between
the plugged-in device and all passive Wi-Fi transmitters.
2. Where do I place my Wi-Fi router and the pluggedin
device, so that I can have passive Wi-Fi devices work
from anywhere in my home? Fig. 6 shows the 2D coverage
maps for different distances between the plugged-in
device and the Wi-Fi router. The red region represent
points in the 2D space where a passive Wi-Fi transmitter
can be located, while ensuring that the signal from
it to the Wi-Fi router is at least -85 dBm. These maps
show that the coverage area is a union of two circles centered
each at the Wi-Fi router and the plugged-in device.
So, as a general rule of thumb, it is better to deploy the
plugged-in device and the Wi-Fi router at either ends of
the coverage area. Note however that at very large distances
between the plugged-in device and Wi-Fi router
(Figs. 6 (c) and (d)), we end up getting two islands of
coverage. Such large distance deployments are suitable
only when the passive Wi-Fi transmitters are going to be
close to either the plugged-in device or the Wi-Fi router.
Device ID Ack Rate Check Bits
0 : 9 10 11 : 12 13 : 15
Figure 7: Structure of the signaling packet.
3 Passive Wi-Fi Network Stack Design
We first describe how passive Wi-Fi devices share the
ISM band. We then address the issue of ACKs and retransmissions
and finally, present our protocol to associate
passive Wi-Fi devices with the network.
3.1 Sharing the ISM band
Wi-Fi uses carrier sense to share the ISM band. This
however requires a Wi-Fi receiver that is ON before every
transmission. Since Wi-Fi receivers require powerconsuming
RF components like LNA, frequency synthesizers,
mixers and ADCs, this would eliminate the power
savings from our design. Instead, we delegate the task of
carrier sense to the plugged-in device, which also arbitrates
access between multiple passive Wi-Fi devices.
We illustrate this with an example. Say a passive Wi-
Fi transmitter wants to sent a packet on channel 6 and
the plugged-in device transmits its tone between Wi-Fi
channels 1 and 6. Before any of the above transmissions
happen, the plugged-in device first uses carrier sense to
ensure that there are no ongoing transmissions on any the
frequencies including and in between channel 1 and 6.
Once the channels are found free, the plugged-in device
sends a packet signaling a specific passive Wi-Fi
device to transmit. This signal is sent and decoded using
the ultra-low power receiver described in §3.1.1. The
packet starts with an ID unique to each passiveWi-Fi device
(see Fig. 7). When the passive Wi-Fi device detects
its ID, it transmits within a SIFS duration at the end of
the signaling packet. The signaling packet is sent at the
center of channel 1 and 6 as well as in between them.
This prevents other devices in the ISM band from capturing
the channel before the passive Wi-Fi device gets
to transmit. The packet has 16 bits and adds a fixed overhead
of 100 ms for every passive Wi-Fi transmission.
The above description assumes that the plugged-in device
knows when to send the signaling packet to each of
6
the passive Wi-Fi devices in the network. To see how
this can be achieved let us focus on our target IoT applications.
A device sending out beacons is configured
to send them at a fixed rate. Temperature sensors, microphones
andW