We propose a framework called AirID that identifies friendly/authorized UAVs using RF signals emitted by radios mounted on them through a technique called as RF fingerprinting. Our main contribution is a method of intentionally inserting `signatures' in the transmitted I/Q samples from each UAV, which are detected through a deep convolutional neural network (CNN) at the physical layer, without affecting the ongoing UAV data communication process. Specifically, AirID addresses the challenge of how to overcome the channel-induced perturbations in the transmitted signal that lowers identification accuracy. AirID is implemented using Ettus B200mini Software Defined Radios (SDRs) that serve as both static ground UAV identifiers, as well as mounted on DJI Matrice M100 UAVs to perform the identification collaboratively as an aerial swarm. AirID tackles the well-known problem of low RF fingerprinting accuracy in `train on one day test on another day' conditions as the aerial environment is constantly changing. Results reveal 98% identification accuracy for authorized UAVs, while maintaining a stable communication BER of 10-4for the evaluated cases.

5G and open radio access networks (Open RANs) will result in vendor-neutral hardware deployment that will require additional diligence towards managing security risks. This new paradigm will allow the same network infrastructure to support virtual network slices for transmit different waveforms, such as 5G New Radio, LTE, WiFi, at different times. In this multivendor, multi-protocol/waveform setting, we propose an additional physical layer authentication method that detects a specific emitter through a technique called as RF fingerprinting. Our deep learning approach uses convolutional neural networks augmented with triplet loss, where examples of similar/dissimilar signal samples are shown to the classifier over the training duration. We demonstrate the feasibility of RF fingerprinting base stations over the large-scale over-the-air experimental POWDER platform in Salt Lake City, Utah, USA. Using real world datasets, we show how our approach overcomes the challenges posed by changing channel conditions and protocol choices with 99.86% detection accuracy for different training and testing days.

Network densification through the deployment of WiFi access points (APs) is a promising solution towards achieving high connectivity rates required for emerging applications. A critical first step is to discover an AP before an active association between the client and the AP can be established. Legacy AP discovery procedures initiated by the client result in high latency in the order of a few 100 ms and waste spectrum, especially when clients need to frequently switch between multiple APs. We propose CSIscan that exploits the broadcast nature of WiFi channels by embedding discovery related information within an AP's ongoing regular transmissions. The AP does this by intelligently distorting the transmitted OFDM frame by inducing perturbations in the preamble, and these injected `bits' of information are detected via changes in the perceived channel state information (CSI). A deep learning framework allocates the optimal level of distortion on a per-subcarrier basis that keeps the resulting packet error rate to less than 1%. Existing clients perceive no changes in their ongoing communication, while potential new clients quickly obtain discovery information at the same time. We experimentally demonstrate that CSIscan reduces the overall WiFi latency from 150 ms to 10 ms and improves spectrum utilization with ~72% reduction in the probe traffic. We show that CSIscan delivers up to 40 discovery information bits in the outgoing WiFi packet in an indoor environment.

Efficient radio resource allocation is a fundamental optimization problem for wireless networks, and has been widely studied in the past. However, wireless systems are evolving to have a much larger parameter space along with richer set of applications and user requirements leading to significant increase in complexity. This paper draws from recent breakthroughs in applying deep reinforcement learning (RL) for control problems containing large state space dimensionality. In particular, a deep RL approach is explored for the problem of allocating time and frequency resources in OFDMA wireless systems to optimize different objective functions using per-station channel quality and traffic information as inputs. Such approaches hold the potential for agents to learn resource allocation and scheduling policies directly from experience rather than using carefully crafted heuristic algorithms based upon models of the environment and stations. Learning directly from experience also means that the policies that result from the online learning should be more robust to imperfect inputs such as noisy, delayed or missing information compared to model-based heuristic approaches. The results in this paper show promise for a deep RL agent using a policy gradient algorithm to learn policies that approach or exceed the performance of well-known model-based approaches such as max-weight and proportional fair policies. The online adaptation algorithms used for the deep RL agent also demonstrate reasonable adaptability and robustness to varying traffic conditions.

This paper describes the architecture and performance of ORACLE, an approach for detecting a unique radio from a large pool of bit-similar devices (same hardware, protocol, physical address, MAC ID) using only IQ samples at the physical layer. ORACLE trains a convolutional neural network (CNN) that balances computational time and accuracy, showing 99% classification accuracy for a 16-node USRP X310 SDR testbed and an external database of >100 COTS WiFi devices. Our work makes the following contributions: (i) it studies the hardware-centric features within the transmitter chain that causes IQ sample variations; (ii) for an idealized static channel environment, it proposes a CNN architecture requiring only raw IQ samples accessible at the front-end, without channel estimation or prior knowledge of the communication protocol; (iii) for dynamic channels, it demonstrates a principled method of feedback-driven transmitter-side modifications that uses channel estimation at the receiver to increase differentiability for the CNN classifier. The key innovation here is to intentionally introduce controlled imperfections on the transmitter side through software directives, while minimizing the change in bit error rate. Unlike previous work that imposes constant environmental conditions, ORACLE adopts the `train once deploy anywhere' paradigm with near-perfect device classification accuracy.

This paper presents FreeIoT, a control plane paradigm that allows fine grained signaling for city-scale IoT deployments without installing any additional infrastructure. FreeIoT overlays control/wake-up information for sensors over existing standards compliant LTE through the following contributions: First, we develop a novel encoding scheme that changes the spatial positioning of Almost Blank Subframes (ABS) within a standard LTE frame to convey control information. ABS was originally defined in the standard to allow coexistence between the macro-cell eNB and nearby small cells, which FreeIoT leverages as a side channel for IoT signaling. Our approach works with any number of ABS settings chosen by the LTE eNB, and accordingly adjusts the encoding of control messages at maximum possible transmission rates. Second, a session management protocol is introduced to maintain contextual information of the control signaling. This allows FreeIoT to handle situations where the control message may span multiple frames, or when the LTE operator temporarily reduces the number of ABS. FreeIoT also incorporates an error detection and correction mechanism to counter channel and fading errors. Finally, we implement a proof of concept testbed to validate the operation of FreeIoT using a software defined LTE eNB and custom-designed RF energy harvesting circuit interfaced with off-the-shelf sensors.

Emerging network architectures in 60GHz millimeter wave bands will likely use dense deployment of access points (APs) given the high attenuation and frequent line-of-sight related outages. We show in this paper that naively connecting to any available AP, or even multiple APs, may not fully realize the promise of efficiently utilizing the extremely large bandwidths available in this band. This paper holistically addresses these problems through the Multi-AP Association Protocol (MAP) that: (i) indicates ideal durations for beam-searching at the physical/link layers of the protocol stack that will result in minimal interruptions to user-traffic, (ii) devises a Multi-AP association framework based on multipath TCP, which increases the network robustness by allowing immediate redirection of traffic to alternative APs whenever an existing connection is interrupted, and (iii) designs an exploration-exploitation aware flow scheduling algorithm that dynamically activates the sub-flows to optimize transport layer performance. MAP is a lightweight, client side system that needs zero modifications to existing protocol stack, though it interacts with the latter to opportunistically trigger standards-defined functions. The paper also presents convergence analysis of throughput, experimental validation on a 60GHz network testbed, and trace-driven simulation studies that shows MAP achieving 5-7x reduction in re-buffering rate in HD video streaming, rapid convergence to the best possible AP, and fair allocation of resources among clients.

This paper proposes the revamp of architecture of an eNodeB (eNB) for 5G cellular communication networks capable of opening up the crunched spectrum resources. We have developed a novel priority driven Resource Block (RB) handoff algorithm for Full-Duplex Multi-User MIMO communication, where the eNB creates beams for establishing communication links. The use of Full-Duplex operation enables simultaneous in-band uplink and downlink operation without sacrificing temporal resource in each link. Space-division multiple access (SDMA) technique has been used to dynamically create a beam towards the intended mobile users as well as track them through beam steering. The possibility of co-channel interference (CCI) due to mobility of users is eliminated by the introduction of RB handoff mechanism. This allows sharing of a RB or RBs by multiple users. In the proposed architecture, along with coexistence of multiple mobile UEs in same RBs diversity gain is introduced both in uplink and downlink. The work includes related simulations showing the conceptual validity of the proposed architecture to revamp. The proposed eNB model leads to improved QoS and a significant decrease in cellular spectrum requirement worth millions of dollars.