Wisee Technology

Download Full-Text PDF Cite this Publication

Text Only Version

Wisee Technology

Wisee Technology

B.Tech., Electronics & Communication Engineering

Shrinathji Institute of Technology & Engineering

B.Tech., Electronics & Communication Engineering

Shrinathji Institute of Technology & Engineering

    1. ech., VLSI Design Institute of Technology & Management

      Nathdwara, India Nathdwara, India Bhilwara, India sapnaajmera@ymail.com Navyachouhan2192@gmail.com Nakul.audeechya@gmail.com

      AbstractWisee is an application of gesture recognition technology that utilizes wifi signals to control electronic appliances through body gesture. Wisee is based on the Doppler effect and by manipulating frequency change, the instructions are given to corresponding device. Wisee enables whole home automation without requiring deployment of human with sensors. The basic idea underlying WiSee is to transform the received WiFi signal into a narrowband pulse with the bandwidth of a few Hertz. The receiver then tracks the energy of this narrowband pulse to detect the small Doppler shifts.

      Index Terms Wifi, Doppler shift


        Technology has finally gained the dimension when we do not require any hand held device to control our appliance and they are just replaced by our hand and by directly interacting with our television and computer .For instance for deleting any folder, user just need to wave his hand and removing that folder lie throwing it in dustbin, and using microwave oven for backing cake or change the song playing while showering these would all be done by waving our hand in air like a magician.

        Gesture recognition is the process by which gestures made by the user are made known to the system. There is an arising need of new ways of interacting with computer interfaces .the example of such kind of gesture based technology is Xbox kinect but it requires sensor devices and computer vision. Wisee enables entirely new set of interaction techniques.

        Since these signals do not require line-of-sight and can traverse through walls, very few signal sources need to be available in the space (e.g., a Wi-Fi AP and a few mobile devices in the living room). WiSee works by evaluating the minute Doppler shifts and multi-path distortions that occur with these wireless signals from human motion in the environment.


WiSee is a wireless system that enables whole-home gesture recognition. Since wireless signals can typically propagate through walls, and do not require a line of sight channel, WiSee can enable gesture recognition independent of the users position.

Fig 1 Gesture recognition system Extracting Doppler Shift From Wireless Signals:

Doppler shift is that the amendment within the determined frequencies the transmitter and also the receiver move relative to each alternative. In our context, the object reflecting the signals from the transmitter will be thought of as a virtual transmitter that generates the mirrored signals. Now, because the object (virtual transmitter) moves towards the receiver, the peaks and troughs of the mirrored signals reach the receiver at a quicker rate. Similarly, as associate degree object moves off from the receiver, the crests and troughs attain a slower rate.

  • The ascertained propagation depends on the direction of positive and negative motion with relation to the receiver. As an example, a degree object moving orthogonal to the direction of the receiver ends up in no propagation, whereas a degree object moving towards the receiver maximizes the propagation. Since human gestures usually involve multiple purpose objects moving on totally different directions, the set of Doppler shifts seen by a receiver will, in essence, be accustomed classify totally different gestures.

  • Higher transmission frequencies lead to a better Doppler shift for a similar motion. Thus, a Wi-Fi transmission at five GHz ends up in doubly the Doppler Effect as a Wi-Fi transmission at two.5 GHz. We note, however, that a lot of higher frequencies (e.g., at sixty GHz) might not be

    appropriate for whole-home gesture recognition since they're a lot of directional and generally not appropriate for NLOS situations.

  • Quicker speeds end in larger Doppler shifts, whereas slower speeds lead to smaller Doppler shifts. Thus, it is easier to a human a person's running towards the receiver than to sight a human walking slowly. Further, gestures involving full-body motion (e.g. walking towards or off from the receiver) area unit easier to capture than gestures involving solely elements of the body (e.g., hand motion towards or off from the receiver). This is because a full-body motion involves.


In this section, we tend to show a way to extract the Doppler data and map it to the gestures.

  1. Doppler Extraction:

    WiSee extracts the Doppler data by computing the frequency- time Doppler profile of the narrowband signal. To do this, the receiver computes a series of FFTs confiscate time. Specifically, it computes an FFT over samples within the initial half-a-second interval. Such an FFT provides a doppler resolution of two Hertz. The receiver then moves forward by a five ms interval and computes another FFT over consequent overlapping half-a second interval.

  2. Segmentation:

To try to do this, WiSee uses the structure of the doppler profiles. WiSee segments the profiles into sequences of positive and negative doppler shifts, that unambiguously determine every gesture. In addition to, every gesture contains of a group of segments that have positive and negative Doppler shifts.

A WiSee receiver uses these properties to initial realizes segments then cluster segments into a gesture. If the quantitative relation between this energy and background level is larger than a threshold, then the receiver checks the start of a packet. Similarly, if this quantitative relation falls below a threshold, the receiver detects the end of the packet. Likewise, in our system, the energy in every phase initial will increase then decreases. Therefore the WiSee receiver computes the average energy within the positive and negative doppler frequencies (other than the DC and therefore the four frequency bins around it). If the quantitative relation between this average energy and therefore the background level is larger than three dB, the receiver detects the start of a segment. Once this quantitative relation falls below three dB, the receiver detects the end of the segment. To cluster segments into one gesture, WiSees receiver uses a straightforward algorithm: if 2 segments are separated by less than one second, we tend to cluster them into one gesture.

  1. Gestures Classification:

    As depicted earlier, Doppler profiles could also be thought of as a sequence of positive and negative Doppler shifts. Further, from the plots, we tend to see that the patterns are distinctive and totally different across the 9 gestures. Thus, the receiver will classify gestures by matching the pattern of positive and negative doppler shifts. Specifically, there are 3 sorts of segments: segments with solely positive doppler shifts, segments. With solely negative doppler shifts, and segments with each positive and negative doppler shifts. These may be described as 3 numbers, 1, -1, and 2. Every gesture in will now could be written as a novel sequence of those 3 numbers.

    Now, gesture classification may be performed by examination and matching the received range sequence with te set of pre- determined sequences. We tend to note that our classification rule works with different to completely different users performing gestures at different speeds. This is often as a result of, totally different speeds solely modification the length of every phase and also the specific doppler frequencies that have energy, however don't alter the pattern of positive and negative shifts. Thus, the gestures performed at totally different speeds lead to identical pattern of numbers and thus may be classified.


Human motion leads to a really little Doppler effect that may be exhausting to discover from a typical wireless transmission (e.g., Wi-Fi, WiMax, LTE, etc.). For example, take into account a user moving her hand towards the receiver at 0.5 m/sec. This end in a Doppler effect of regarding seventeen Hertz for a Wi-Fi signal transmitted at five rates. Since the information measure of Wi-Fis transmissions is a minimum of twenty megahertz, the ensuing Doppler Effect is orders of magnitude smaller than Wi-Fis bandwidth. Distinguishing such little Doppler shifts from these transmissions may be difficult.

Our Solution:

WiSee presents a receiver design that may identify Doppler shifts at the resolution of a couple of Hertz from Wi-Fi signals. The fundamental idea underlying WiSee is to remodel the received Wi-Fi signal into a narrowband pulse with the bandwidth of a couple of Hertz. The receiver then tracks the frequency of this narrowband pulse to detect the little Doppler shifts. WiSee is meant for OFDM-based systems OFDM is that the modulation of choice for many modern wireless systems admitting 802.11 a/g/n, WiMAX, and LTE. OFDM divides the used RF bandwidth into multiple sub-channels and modulates data in each sub-channel. as an example, Wi-Fi typically divides the 20 MHz channel into 64 sub-channels each with a bandwidth of 312.5 KHz. The time-domain OFDM symbol is generated at the transmitter by taking an FFT over a sequence of modulated bits transmitted in each OFDM sub-channel. Specifically, the transmitter takes blocks of N modulated bits (N = sixty four in 802.11), and applies an

N-point Inverse fast Fourier transform (IFFT), Equations.

Case 1: Transmitter sends a similar OFDM symbol. In this case, rather than playing an FFT over every OFDM Symbol l, WiSees receiver performs an oversized FFT over M consecutive OFDM symbols. As a consequence of this operation, the information measure of every OFDM sub- channel is reduced by an element of M. to check this, say the receiver performs a 2N-point FFT over 2 consecutive, identical OFDM symbols.

Thus, WiSee will produce multiple narrowband signals focused at every sub-channel by continuation an OFDM symbol and performing an oversized FFT operation. Now, by applying an oversized FFT over a one-second period, the WiSee receiver will produce a one-Hertz wide narrowband signal. The WiSee receiver tracks this narrowband signal to capture the Doppler Effect .Note that one will average the Doppler shifts ascertained across all the OFDM sub-channels to considerably cut back the noise within the Doppler measurements.

Case 2: Transmitter sends capricious OFDM symbols. Our description up to now presumes that the transmitter repeatedly sends constant OFDM symbol. Typical 802.11 transmitters but send utterly totally different data across symbols. WiSee achieves this by springing up with a data-equalizing re- encoder at the receiver that transforms every received OFDM symbol into constant symbol. To do this, the receiver 1st decodes the symbols exploitation the quality 802.11 decoder. Specifically, the receiver performs an FFT on every time- domain OFDM symbol and transforms it into the frequency- domain. The receiver, then, decodes the modulated bits in every sub-channel, transfers the modulated bits through the detector and so the convolutional/Viterbi decoder to urge the transmitted bits.

Human gestures modification the half and amplitude of the received symbols. a standard decoder accounts for these changes by exploitation the pilot bits that are present in every OFDM symbol. Specifically, the receiver decodes by removing these halves and amplitude changes that code the gesture data. To avoid this, throughout the re encryption section before conniving the IFFT, the WiSee receiver re- introduces the section and amplitude changes that were removed by the decoder. This assures that the gesture data isn't lost in decoding.

Multiple Humans:

WiSee utilizes MIMO to improve the accuracy and lustiness of the system, and to enable it to work in the presence of multiple humans .MIMO decoding requires a known preamble to compute the MIMO channel of the target user. WiSee uses a repetitive gesture as a preamble. Specifically, the user pushes her hand towards and away from the receiver, and repeats this gesture to form the preamble. This produces a sequence of alternating positive (+1) and negative (-1) Doppler shifts, i.e., an alternating sequence of +1 and -1 symbols. The WiSee receiver uses this sequence to correlate

and detect the presence of a target human. Note that, similar to communication systems, this correlation works even in the presence of interfering users, since their movement is uncorrelated with the preambles alternating sequence of positive and negative Doppler shifts.

Next, WiSee finds the MIMO channel that maximizes the Doppler energy from the target user.

Fig 2 MIMO System

Providing security:

One of the risks of using a whole-home gesture recognition system is enabling an unauthorized user outside the home to control the devices to address this problem; one may use a mystical pattern of gestures as a secret key to get access to the system. Once the system is locked onto the user, the receiver can track the authorized user and perform the corresponding gestures.


In this paper, we tend to take the initiative towards transmuting Wi-Fi into a gesture-recognition detector. We tend to present WiSee, a unique gesture recognition system that utilizes wireless signals to modify whole-home sensing and recognition of human gestures. Since wireless signals don't need line-of-sight and might pass over through walls, WiSee will modify whole home gesture recognition exploitation few signal sources. WiSee will extract associate degree ample set of gesture info from wireless signals and modify whole-home gesture recognition exploitation solely 2 wireless sources placed within the lounge


  1. ACQUITEK Inc. Fury GPS Disciplined Frequency Standard.

  2. R. Block. Toshiba Qosmio G55 features SpursEngine, Visual Gesture Controls.

  3. W. Carrara, R. Goodman, and R. Majewski. Spotlight

  4. Synthetic Aperture Radar: Signal Processing Algorithms.Artech House, 1995

  5. K. Tan, J. Fang, Y. Zhang, S. Chen, L. Shi, J. Zhang, and Y. Zhang. Fine-grained Channel Access in Wireles LAN. In SIGCOMM, 2010.

  6. J. Wilson and N. Patwari. Radio Tomographic Imaging with Wireless Networks. In Trans. Mobile Computing, 2009.

  7. J. Wilson and N. Patwari. Through-Wall Motion Tracking Using Variance-Based Radio Tomography Networks. In ARXIV, 2009.

  8. J. Xiong and K. Jamieson. Towards fine-grained radio-based indoor location. In HotMobile, 2012.

  9. J. Xiong and K. Jamieson. ArrayTrack: A Fine Grained Indoor Location System. In NSDI, 2013.

  10. M. Youssef, M. Mah, and A. Agrawala. Challenges:

    Device-free passive localization for wireless environments. In Mobicom, 2007.

  11. Y. Zhao and N. Patwari. Robust Estimators for

    Variance-Based Device-Free Localization and Tracking. In ARXIV, 2011. I. S. Jacobs and C. P. Bean, Fine particles, thin films and exchange anisotropy, in Magnetism, vol. III, G. T. Rado and H. Suhl, Eds. New York: Academic, 1963

    S. Gollakota and D. Katabi. Zigzag decoding: combating hidden terminals in wireless networks. In ACM SIGCOM, 2008.

  12. S. Gupta, D. Morris, S. Patel, and D. Tan. Soundwave: using the doppler effect to sense gestures. In HCI 2012.

  13. C. Harrison, D. Tan, and D. Morris. Skinput: appropriating the body as an input surface. In CHI 2010.

  14. D. Kim, O. Hilliges, S. Izadi, A. D. Butler, J. Chen, Oikonomidis, and P. Olivier. Digits: freehand 3d interactions anywhere using a wrist-worn gloveless sensor.In UIST 2012.

  15. Y. Kim and H. Ling. Human Activity Classification Based on Micro-Doppler Signatures Using a Support Vector Machine. In Trans. Geoscience and Remote Sensing, 2012.

  16. K. Lin, S. Gollakota, and D. Katabi. Random Access Heterogeneous MIMO Networks. In SIGCOMM, 2011.

  17. J. Paradiso, C. Abler, K.-y. Hsiao, and M. Reynolds. The magic carpet: physical sensing for immersive environments. In CHI, 1997.

Leave a Reply

Your email address will not be published. Required fields are marked *