KSKannan Srinivasan

Kannan Srinivasan

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Kannan Srinivasan has a total of 41 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Co-Designing Agentic AI and Wireless/Edge Networks: A Vision for Self-Evolving, Trustworthy Network Intelligence

    by Ness Shroff, Yingbin Liang, Jia Kevin Liu, Anish Arora, Raef Bassily, Elisa Bertino, Constantine Caramanis, Kaushik Chowdhury, Eylem Ekici, Atilla Eryilmaz, et al.

  • Open Access

    0cal: Zero-Cost Calibration for mmWave Networks

    by Xin Liu, Wei-Han Chen, Kannan Srinivasan

  • Open Access

    HORCRUX: Accurate Cross Band Channel Prediction

    by Avishek Banerjee, Xingya Zhao, Vishnu Chhabra, Kannan Srinivasan, Srinivasan Parthasarathy


  • Open Access

    Fewer Demands, More Chances: Active Eavesdropping in MU-MIMO Systems

    by Xingya Zhao, Anwesha Roy, Avishek Banerjee, Kannan Srinivasan

  • Open Access

    RFDrive: Tagged Human-Vehicle Interaction for All

    by Wei Sun, Kannan Srinivasan

    Human–vehicle interaction is an important factor for safe driving. The driver needs to interact with the in-vehicle steering wheel and infotainment system properly during driving. Specifically, driving guidelines require the driver to hold the steering wheel at the 3 o’clock and 9 o’clock positions. Moreover, the in-vehicle infotainment system should be more adaptive for the driver and front-seat passenger during driving (i.e., the in-vehicle infotainment system should be part and even fully disabled for the driver, whereas the front-seat passenger should be able to enjoy the full in-vehicle infotainment system). However, affordable vehicles are usually designed to achieve basic driving functions without considering safe human–vehicle interactions, which require an add-on, affordable, and ready-to-use human–vehicle interaction monitoring system. In this article, we present RFDrive, a system that can simultaneously locate the driver’s hand positions on the steering wheel and automate in-vehicle infotainment system touch discrimination for safe driving using commodity passive RFID tags. Since these commodity passive RFID tags are low cost (i.e., around 5 cents per tag), battery free, and are small, like a sticker, our design will enable not only safe driving but is also low cost, which can lead to sustainable solutions. To do so, we attach RFID tags on the steering wheel for the driver’s hand position location and attach RFID tags on the roof of the vehicle’s interior for in-vehicle infotainment system touch discrimination (i.e., differentiating the driver’s infotainment system touch and front-seat passenger’s infotainment system touch). However, the wheel steering will distort the wireless channel-based driver’s hand position location on the steering wheel. Thus, we propose a novel tag ID-based algorithm to locate the driver’s hand position on the steering wheel by harnessing the human body as part of the RFID tag’s antenna. Since the in-vehicle infotainment system touch from the driver or front-seat passenger will affect different RFID tags attached to the roof of the vehicle’s interior, we propose to use the differential amplitude of backscattered signals from all the tags to discriminate in-vehicle infotainment system touch sources. Our experiments show that RFDrive can achieve the average accuracy of 0.98 and 0.98 for in-vehicle touch source discrimination and driver’s hand position location, respectively.


  • Open Access

    Fast and Optimal Beam Alignment for Off-the-Shelf mmWave Devices

    by Wei-Han Chen, Xin Liu, Kannan Srinivasan, Srinivasan Parthasarathy

    Overcoming the very high path loss is considered a prerequisite for operating millimeter-wave (mmWave) beamforming to meet the high data-rate requirements of next-generation wireless systems. Existing work chose the optimal beam pattern outside the codebooks to maximize the SNR. However, it is nontrivial to quickly find the optimal beam for off-the-shelf mmWave devices, especially as the standardization efforts are moving towards incorporating large-scale antenna arrays. In this paper, we present SWIFT, a system that can significantly reduce the beam scan time while achieving optimal beam alignment for off-the-shelf mmWave devices. To implement SWIFT, we tackle algorithm complexity and calibration cost, making SWIFT independent of any phase measurements and instrument or specialized circuitry. Our results show that SWIFT can achieve near-optimal SNR performance on the off-the-shelf devices and significantly save beam scan time over the state-of-the-art. Our design is generic and has the potential to scale to future-large scale antenna arrays (e.g., massive MIMO in 5G and beyond).


  • Open Access

    Malicious Relay Detection and Legitimate Channel Recovery

    by Xingya Zhao, Wei-Han Chen, Kannan Srinivasan

    Full-duplex devices can compromise the integrity of wireless channel measurements through signal relaying and several attacks have been proposed based on this vulnerability. Existing source authentication methods relying on previously-collected signatures face significant challenges in detecting these attacks because a relay attacker can gradually inject the channels so that the manipulated channels will fall within the tolerance range of the authentication methods and are mistaken as new signatures. In this paper, we propose RelayShield, a system for detecting malicious relays and recovering the legitimate transmitter-receiver channels from the manipulated channels. RelayShield requires only one channel measurement at the receiver. It analyzes signal path information resolved from input channels to detect relays and recover channels. RelayShield achieves over 95% detection accuracy with channels collected in two typical indoor environments. The recovered channels can support a wide range of applications, including secret generation protocols and sensing systems.


  • Open Access

    Embracing Collisions to Increase Fidelity of Sensing Systems with COTS Tags

    by Jiaqi Xu, Wei Sun, Kannan Srinivasan

    RFID techniques have been extensively used in sensing systems due to their low cost. However, limited by the structural simplicity, collision is one key issue which is inevitable in RFID systems, thus limiting the accuracy and scalability of such sensing systems. Existing anti-collision techniques try to enable parallel decoding without sensing based applications in mind, which can not operate on COTS RFID systems. To address the issue, we propose COFFEE, which enables parallel channel estimation of COTS passive tags by harnessing the collision. We revisit the physical layer design of current standard. By exploiting the characteristics of low sampling rate and channel diversity of RFID tags, we separate the collided data and extract the channels of the collided tags. We also propose a tag identification algorithm which explores history channel information and identify the tags without decoding. COFFEE is compatible with current COTS RFID standards which can be applied to all RFID-based sensing systems without any modification on tag side. To evaluate the real world performance of our system, we build a prototype and conduct extensive experiments. The experimental results show that we can achieve up to 7.33x median time resolution gain for the best case and 3.42x median gain on average.


  • Open Access

    Embracing collisions

    by Jiaqi Xu, Wei Sun, Arjun Bakshi, Kannan Srinivasan

    There is a growing interest in backscatter-based sensing systems in recent years. RFID techniques can be used due to low cost and structural simplicity. However, collision caused by simultaneous tag responses is one of the key issues in backscatter systems. Existing anti-collision techniques try to enable parallel decoding without sensing based applications in mind, which can not operate on COTS RFID systems. To address the issue, we propose COFFEE, which is the first work to enable parallel channel estimation of COTS passive tags by harnessing the collision. By exploiting the characteristics of low sampling rate and channel diversity of passive tags, we separate the collided data, extract the channels and identify the tags. COFFEE is compatible with current RFID standards which can be applied to all RFID-based sensing systems without any modification on tag side. To evaluate the real world performance of our system, we build a prototype and conduct extensive experiments. The experimental results show that we can achieve up to 7.33x median time resolution gain for the best case and 3.42x median gain on average.


  • Open Access

    Enabling real-time interference alignment

    by Kyle Miller, Atresh Sanne, Kannan Srinivasan, Sriram Vishwanath