This paper proposes and experimentally demonstrates a first wireless local area network (WLAN) system that jointly exploits physical-layer network coding (PNC) and multiuser decoding (MUD) to boost system throughput. We refer to this multiple access mode as network-coded multiple access (NCMA). Prior studies on PNC mostly focused on relay networks. NCMA is the first realized multiple access scheme that establishes the usefulness of PNC in a non-relay setting. NCMA allows multiple nodes to transmit simultaneously to the access point (AP) to boost throughput. In the non-relay setting, when two nodes A and B transmit to the AP simultaneously, the AP aims to obtain both packet A and packet B rather than their network-coded packet. An interesting question is whether network coding, specifically PNC which extracts packet A ⊕ B, can still be useful in such a setting. We provide an affirmative answer to this question with a novel two-layer decoding approach amenable to real-time implementation. Our USRP prototype indicates that NCMA can boost throughput by 100 percent in the medium-high SNR regime (≥10 dB). We believe further throughput enhancement is possible by allowing more than two users to transmit together.

This article presents the first orthogonal frequency-division multiplexing(OFDM)-based digital over-the-air computation (AirComp) system for wireless federated edge learning, where multiple edge devices transmit model data simultaneously using non-orthogonal OFDM subcarriers, and the edge server aggregates data directly from the superimposed signal. Existing analog AirComp systems often assume perfect phase alignment via channel precoding and utilize uncoded analog transmission for model aggregation. In contrast, our digital AirComp system leverages digital modulation and channel codes to overcome phase asynchrony, thereby achieving accurate model aggregation for phase-asynchronous multi-user OFDM systems. To realize a digital AirComp system, we develop a medium access control (MAC) protocol that allows simultaneous transmissions from different users using non-orthogonal OFDM subcarriers, and put forth joint channel decoding and aggregation decoders tailored for convolutional and LDPC codes. To verify the proposed system design, we build a digital AirComp prototype on the USRP software-defined radio platform, and demonstrate a real-time LDPC-coded AirComp system with up to four users. Trace-driven simulation results on test accuracy versus SNR show that: 1) analog AirComp is sensitive to phase asynchrony in practical multi-user OFDM systems, and the test accuracy performance fails to improve even at high SNRs; 2) our digital AirComp system outperforms two analog AirComp systems at all SNRs, and approaches the optimal performance when SNR $\geq$ 6 dB for two-user LDPC-coded AirComp, demonstrating the advantage of digital AirComp in phase-asynchronous multi-user OFDM systems.

Recent studies show that Wi-Fi interference has been a major problem for low power urban sensing technology ZigBee networks. Existing approaches for dealing with such interferences often modify either the ZigBee nodes or Wi-Fi nodes. However, massive deployment of ZigBee nodes and uncooperative Wi-Fi users call for innovative cross-technology coexistence without intervening legacy systems. In this work, we investigate the Wi-Fi and ZigBee coexistence when ZigBee is the interested signal. Typically, the duration of transmitting a ZigBee data packet is longer than that of a Wi-Fi packet. Mitigating short duration Wi-Fi interference (called flash ) in long duration ZigBee data (called smog ) is challenging. To address these challenges, we propose ZIMO: a sink-based MIMO design for harmony coexistence of ZigBee and Wi-Fi networks with the goal of protecting the ZigBee data packets from being interfered by high-power cross-technology signals. The key insight is to properly exploit opportunities resulted from differences between Wi-Fi and ZigBee, and bridge the gap between interested data and cross technology signals. Also, extracting the channel coefficient of Wi-Fi and ZigBee will enhance other coexistence technologies such as TIMO [1] . We implement a prototype in GNURadio-USRP N200, and our extensive evaluations under real wireless conditions show that ZIMO can improve ZigBee network throughput up to 1.9 $\times$ , with 1.5 $\times$ in media, and 1.1 $\times$ to 1.9 $\times$ for Wi-Fi network as byproduct in ZigBee signal recovery.

Coexistence of Wi-Fi and ZigBee in 2.4 GHz ISM band is a long standing and challenging problem. Previous solutions either require modifications of current ZigBee protocols or Wi-Fi re-configurations, which is not feasible in large-scale wireless sensor networks. In this paper, we present WizBee, a coexistence system using single-antenna sink without changing current Wi-Fi and ZigBee design. WizBee is based on an observation that Wi-Fi signal is about 5 to 20 dB stronger than ZigBee signal in symmetric area, which leaves much room for applying interference cancelation technique to mitigate Wi-Fi interference, and extract ZigBee signals. However, we need to cancel the Wi-Fi interference perfectly for residual ZigBee signal decoding, which needs more accurate channel coefficient across data transmissions in spite of cross technology interference. For robust and accurate Wi-Fi decoding, we use soft Viterbi decoding with weighted confidence value over interfered subcarriers. Consequently, our solution uses decoded data for channel coefficient estimation instead of conventional training symbol based methods. The key insight is that, the signal recovery opportunity for cross technology coexistence, lies in multi-domain information, such as power, frequency and coding discrepancies. Using these information properly will improve the coexistence network throughput effectively. We implemented WizBee in USRP/GNURadio software radio platform, and studied the decoding performance of interference cancelation technique. Our extensive evaluations under real wireless conditions show that WizBee improves ZigBee throughput up to 1.9x, with median throughput gain of 1.2x.

Recent studies show that WiFi interference has been a major problem for low power urban sensing technology ZigBee networks. Existing approaches for dealing with such interferences often modify either the ZigBee nodes or WiFi nodes. However, massive deployment of ZigBee nodes and uncooperative WiFi users call for innovative cross-technology coexistence without intervening legacy systems. In this work we investigate the WiFi and ZigBee coexistence when ZigBee is the interested signal.Mitigating short duration WiFi interference (called flash) in long duration ZigBee data (called smog) is challenging, especially when we cannot modify the WiFi APs and the massively deployed sensor nodes. To address these challenges, we propose ZIMO, a sink-based MIMO design for harmony coexistence of ZigBee and WiFi networks with the goal of protecting the ZigBee data packets.The key insight of ZIMO is to properly exploit opportunities resulted from differences between WiFi and ZigBee, and bridge the gap between interested data and cross technology signals. Also, extracting the channel coefficient of WiFi and ZigBee will enhance other coexistence technologies such as TIMO [1]. We implement a prototype for ZIMO in GNURadio-USRP N200, and our extensive evaluations under real wireless conditions show that ZIMO can improve up to 1.9x throughput for ZigBee network, with median gain of 1.5x, and 1.1x to 1.9x for WiFi network as byproduct in ZigBee signal recovery.

This paper presents a first real-time network-coded multiple access (NCMA) system that jointly exploits physical (PHY)-layer network coding (PNC) and multiuser decoding (MUD) to boost the throughput of a wireless local area network (WLAN). NCMA is a new design paradigm for multipacket reception wireless networks, in which the access point can receive and decode several packets simultaneously transmitted by multiple users. Conventionally, multipacket reception is realized using MUD only, whereas the key idea of NCMA is to use PNC together with MUD to realize multipacket reception. Although the feasibility of NCMA has previously been studied by the authors, our previous NCMA prototype was a version with offline signal processing. In addition, our previous investigation left open a number of theoretical and implementation issues, the resolution of which is critical to the adoption of NCMA in real practice. The current investigation makes the following state-of-the-art contributions toward NCMA: 1) we demonstrate a first NCMA system with integrated real-time PHY and MAC-layer decoding; 2) we construct a new unified framework for MAC-layer decoding that yields higher throughput with faster decoding-the faster decoding is one of the key enablers of our real-time implementation; and 3) we design new PHY-layer decoding techniques that overcome the poor performance of the first-generation NCMA prototype at low SNR. Experimental results show that, compared with the previous NCMA prototype, our new NCMA prototype improves real-time throughput by more than 100% at medium-high SNR (≥ 8 dB).