Wi-Vi is a device that uses Wi-Fi signals to identify people, their locations, and motions behind walls. It operates in two modes: one to image and track moving objects, and another for gesture-based communication without transmitters. Wi-Vi eliminates flash reflections using interference nulling between transmit and receive antennas. It tracks humans using inverse synthetic aperture radar as they sample the signal at different locations, emulating an antenna array. Gestures like stepping forward or backward encode bits to allow hidden communication. Wi-Vi showed accurate detection up to 5 meters away.

1. Wi-Vi
WIRELESS VISION TECHNOLOGY
Presented By
Dr. SK. Mastan Vali
A SEMINAR ON:

2. INTRODUCTION:
o Wi-Vi, Wireless Vision or Wi-Fi vision, is a see-through-
wall device that employs Wi-Fi signals in the 2.4Gz ISM
band.
o It identifies number of people, their locations and motion
in a closed room.
It can be used for gesture-based communication
without any transmitter.

3. PRINCIPLE:
It uses the principle:
 RF signals can traverse a wall and reflect off objects
behind to wall and reach back the transmitter.
The reflected back signal is imprinted with the signature
of the reflectors.
This is similar to radar and sonar imaging.

4. DESIGN:
Wi-Vi uses Wi-Fi OFDM signals in the ISM band (at 2.4GHz) and
typical Wi-Fi hardware.
It is a 3-antenna MIMO device : two of the antennas being
transmitters and the other a receiver.
 It also employs directional antennas to focus the energy toward the
wall or room of interest.
Its design incorporates two main components:
1. The first component eliminates the flash reflected off the wall.
2. The second component tracks a moving object by treating the
object itself as an antenna array.

5. FLASH EFFECT:
Transmitted RF signals through a wall, get reflected not
only from behind the wall objects but also from the wall.
 These are much stronger and prevent ADC from
registering the minute variations due to reflections from
behind the wall.
This is known as ‘FLASH EFFECT’.

6. OPERATION:
Wi-Vi is operated in two modes:
 In mode1, it can be used to image moving objects behind a wall
and track them.
In mode2, Wi-Vi functions as a gesture-based interface from behind
a wall that enables humans to compose messages and send them to
the Wi-Vi receiver.
Wi-Vi operation uses two basic techniques:
1. MIMO based Nulling to eliminate flash effect
2. Inverse Synthetic Aperture Radar (ISAR) for tracking human
motion

7. ELIMINATING THE FLASH
 Wi-Vi uses interference nulling to cancel both the wall
reflections and the direct signal from the transmit to the receive
antenna.
 To eliminate flash, a sample x is transmitted on each
transmit antenna separately and the ratio p is calculated.
 Then x and px are transmitted concurrently to get the
perceived channel at the receiver.
The process repeats until channel estimates in step1 are
perfect so that the received signal is zero.

8. TRACKING HUMANS:
 Wi-Vi uses Inverse Synthetic Aperture Radar, ISAR technique for
tracking humans behind the wall.
It has one receive antenna that captures one single measurement at
any point of time.
 As the target moves, he/she samples the received signal at
successive locations in space corresponding to successive spatial
locations of the moving target.
 Wi-Vi emulates an antenna array and uses it to track motion behind
the wall.

9. TRACKING HUMANS:

10. Continued..
Single human Multiple human

11. GESTURE-BASED COMMUNICATION:
Wi-Vi enables a human who does not carry any wireless
device to communicate commands or short messages to a
receiver using simple gestures.
It designates a pair of gestures as a ‘0’ bit and a ‘1’ bit.
A human can compose these gestures to create messages that
have different interpretations.

12. Gesture Encoding:
Gestures are encoded using any scheme such that:
The gestures must be composable i.e. at the end of each bit the human
should be back in the same initial state as the start of the gesture
 The gestures must be simple so that a human finds it easy to perform
them and compose them.
The gestures should be easy to detect and decode without requiring
sophisticated decoders.
Here a ‘0’ bit is a step forward followed by a step
backward; a ‘1’ bit is a step backward followed by a step forward. This
modulation is similar to Manchester encoding.

13. Wi-Vi detection of gestures:

14. Gesture Decoding:
 Since each bit is a combination of two steps, forward,
Wi-Vi applies two matched filters: one for the step forward
and one for the step backward.
 Next, Wi- Vi uses a standard peak detector to detect the
peaks/troughs and match them to the corresponding bits.

16. Wi-Vi DESIGN MODEL:
Wi-Vi is built using USRPN210 software radios with SBX daughter boards.
The system uses LP0965 directional antennas, which provide a gain of 6 dBi.
 The system consists of three US- RPs connected to an external clock so that they
act as one MIMO system. Two of the USRPs are used for transmitting, and one for
receiving.
A standard Wi-Fi OFDM modulation is implemented, each OFDM symbol
consisting of 64 subcarriers including the DC. The nulling procedure is
performed on a subcarrier basis.

17. RESULTS:

18. Wi-Vi correctly decoded the performed gestures at all distances
less than or equal to 5m.
A 93.75% accuracy for 6-7m and 75% for 8m was obtained and
above 9m Wi-Vi could not identify any of the gestures.
Continued..

19. ADVANTAGES:
Cheap, small, feasible to general public.
No need to carry transmitting device for communication.
 Accurate detection and good efficiency.

20. APPLICATIONS:
1. Law Enforcement
2. Emergency Situations
3. Personal Security
4. Smart Sensing
5. User Interface Design
6. Entertainment

21. ADVANCEMENTS:
 Can be extended for indoor localization, sensing
and control.
 It can be extended to develop virtual reality
systems.

22. CONCLUSION:
Wi-Vi is a new form of user interface which rely solely
on reflections of a transmitted RF signal to identify
human gestures.
It can be envisioned, promising new direction in virtual
reality.

23. REFERENCES:
Wi-Vi system uses Wi-Fi to see through walls Gizmag
. www.seminarsonly.com
{fadel,dk}@mit.edu
Massachusetts Institute of Technology