Performance of Radar and Communication Networks Coexisting in Shared Spectrum Bands

Abstract

Recent technological advancements are making the use of compact, low-cost, low-power mm-wave radars viable for providing environmental awareness in a number of applications, ranging from automotive to indoor mapping and radio resource optimisation. These emerging use-cases pave the road towards networks in which a large number of radar and broadband communications devices coexist, sharing a common spectrum band in a possibly uncoordinated fashion. Although a clear understanding of how mutual interference influences radar and communications performance is key to proper system design, the core tradeoffs that arise in such scenarios are still largely unexplored. In this paper, we provide results that help bridge this gap, obtained by means of an analytical model and extensive simulations. To capture the fundamental interactions between the two systems, we study mm-wave networks where pulsed radars coexist with communications devices that access the channel following an ALOHA policy. We investigate the effect of key parameters on the performance of the coexisting systems, including the network density, fraction of radar and communication nodes in the network, antenna directivity, and packet length. We quantify the effect of mutual interference in the coexistence scenario on radar detection and communication network throughput, highlighting some non-trivial interplays and deriving useful design tradeoffs.

Cooperative Detection for mmWave Radar-Communication Systems

“…The BSs are uniformly distributed in R 2 according to an independent homogeneous Poisson Point Process (PPP) denoted as Φ = {x j ∈ R 2 , j ∈ N + } with density λ, and transmit with power equal to P t . We assume that all BSs are connected to a central unit, also known as fusion center (FC), through an ideal report channel [4]. In addition, we assume that each BS can operates either in RM or CM with probabilities and 1 − , respectively, where ∈ [0, 1].…”

“…The coverage probability is well-examined in the literature [4], and therefore, in the rest of this paper we focus on the detection performance of RadCom systems.…”

“…In the considered CFA operating mode, the detection threshold is selected based on a fixed target probability of false alarm P fa . In order to derive compact and insightful expressions for the detection performance, the detection threshold is selected solely based on the power of the strongest interfering device [4]. The aforementioned assumption provides an approximation of the actual network's detection performance, which is shown to be tight from numerical results in Section V. Hence, we assume that the required false alarm rate, is equal to P fa = αP[ I τ > ϑ], where I represents the power of the strongest device that causes interference to the serving RBS at u.…”

“…The aforementioned assumption provides an approximation of the actual network's detection performance, which is shown to be tight from numerical results in Section V. Hence, we assume that the required false alarm rate, is equal to P fa = αP[ I τ > ϑ], where I represents the power of the strongest device that causes interference to the serving RBS at u. The constant α is the probability that the strongest interferer is active at least once during the M − 1 time-slots that radar waits for the echo, and is equal to 1 − 1 M + (1 − ) [4]. In the following lemma, the cdf of I τ is analytically evaluated.…”

“…The work in [3] considers the trade-off emerging in networks, where the terminals switch between radar mode (RM) for target detection and communication mode (CM) for data exchange, sharing the same bandwidth. This work is further extended in [4], where the authors quantify the effect of mutual interference on the radar detection and the communication network throughput.…”